University of Southern California Dissertations and Theses

Institutional analysis of evolution of housing cooperatives in India

(USC Thesis Other)

Institutional analysis of evolution of housing cooperatives in India

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INSTITUTIONAL ANALYSIS OF EVOLUTION OF HOUSING COOPERATIVES IN INDIA by Sukumar Ganapati A Dissertation Presented to the FACULTY OF THE SCHOOL OF POLICY, PLANNING, AND DEVELOPMENT UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (PLANNING) December 2003 Copyright 2003 Sukumar Ganapati R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. UMI Number: 3133271 Copyright 2003 by Ganapati, Sukumar All rights reserved. INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. ® UMI UMI Microform 3133271 Copyright 2004 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90089-1695 This dissertation, written by SUKUMAR GANAPATI under the direction o f h i s dissertation committee, and approved by all its members, has been presented to and accepted by the Director o f Graduate and Professional Programs, in partial fulfillment o f the requirements for the degree o f D O C T O R O F P H IL O SO P H Y . Director Date Oprpmhsr 1 7 f 2003 Dissertation Committee H.W.RTGHARDSON W , ftJsAc-rf.W, NTBA.T VERMA S . TANG R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Dedication This dissertation is dedicated to my father, V.S.G. Raman and my deceased mother, G. Padamavati. I owe everything in my life to them. ii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Acknowledgements My dissertation would not have seen the light of the day if it were not for the material and intellectual support I received during my fieldwork and my writing. I should first acknowledge the support I received from three research institutes in India, which acted as intellectual bases for my fieldwork. I was affiliated with Madras Insti tute of Development Studies in Chennai; Indira Gandhi Institute of Development Re search in Mumbai; and National Institute of Public Finance and Policy in New Delhi. I had full access to library and computer facilities in all three institutes. I must thank Paul Appasamy, Kirit Parikh, and O.P.Mathur, who were instrumental in providing affiliation to the three institutes respectively. They also engaged me in stimulating dialogues relating to my dissertation. In Indira Gandhi Institute of Development Re search, Madhura Swaminathan was additionally helpful in crystallizing some of my research ideas. In each institute, I presented my findings in a seminar, where the fac ulty and students actively participated. I thank them for their criticisms and inputs. A difficult part of any fieldwork is the availability of some of the key people who could give insights into research questions. I was lucky that many of these peo ple were willing to spend much time talking to me, exploring some of the questions in more depth than I had anticipated. In Chennai, the following eagerly spent many hours explaining the intricacies of housing and cooperative policies: G. Dattatri, re tired Chief Urban Planner, Madras Metropolitan Development Authority; K.N. Nan- jundan, retired Special Officer of the Tamil Nadu Cooperative Housing Federation; D. Paranjothi, Principal of the Natesan Institute of Cooperative Management; M.S.Somasundaram, General Manager of the Tamil Nadu Cooperative Housing Fed eration; M. Nelson, Director of Center for Development Madras. J. Sukumar, the li brarian at Madras Metropolitan Development Authority, was very enthusiastic in pro- iii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. viding me with relevant documents. I was lucky to be staying with my father, brother (Raja), Mala, and now my sister-in-law (Uma) who went out of their way to help me with small chores related to my research. In Mumbai, the following were enthusiastic in providing inputs for my disser tation: V. K. Phatak, Chief of Planning Division and K. R. Ghorpade, Senior Planner in the same division in Bombay Metropolitan Regional Development Authority; Uma Adusumilli, Senior Planner in City and Industrial Development Corporation of Ma harashtra (CIDCO); D.T. Joseph, a former secretary in the Urban Development De partment and an ex Municipal Commissioner; Raghuvir Samant, an avid promoter of cooperatives and the editor of Housing Times (a magazine devoted to cooperative housing, published by Mumbai District Cooperative Housing Federation), and Sudhir Thakare, Divisional Joint Registrar of Cooperatives. I am especially indebted to R.N.Sharma, a housing expert in the Urban Studies program at Tata Institute of Social Sciences, who not only provided me with his insights into the housing issues in Mum bai, but also gave logistical support in carrying out a survey of housing cooperatives in Navi Mumbai. Sheela Patel, Celine D’Cruz, and Sunder Burra, who are associated with Society for Promotion of Area Resource Centers, along with Alan Jockin, Presi dent of National Slum Dwellers Federation provided me valuable information on how NGOs have adapted cooperatives in Mumbai. My stay in Mumbai could not have been productive without the hospitality of my brother, G. Bhaskar, his wife Meena, and their two children (Sandhya and Akshay). In Delhi, I met with H.C. Joshi, General Manager of Delhi Co-operative Hous ing Finance Corporation; M. L. Khurana, Managing Director of National Cooperative Housing Federation (NCHF); Vinay D. Lall, Director of Society for Development Studies; Saurabh Pillay, Manager of Cooperative Program for SHARAN (a NGO de- iv R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. voted to housing issues in Delhi); and Kiran Wadhwa, Chief Economist in the Hous ing and Urban Development Corporation. I am grateful to all of them for not only the information they provided, but also for their thoughtful comments that helped shape some of my arguments. Delhi also holds a special place for me since many of my un dergraduate classmates from Indian Institute of Technology, Kharagpur have their of fices for architectural practice. I am grateful to Chakresh Jain, Sanjay Mehra, Alok Saxena, and Sharad Trivedi, who generously helped me in various ways. I could not have gone through the motions of writing this dissertation without the emotional and other support from some of my close friends here in Los Angeles. Deepak Bahl helped me in so many ways that it is hard to synthesize even some of them in a few sentences—suffice it to say that his role as the “best man” has not been limited to my wedding only. Shin Lee was a constant source of strength and encour agement; without her, my life as a doctoral student could have been much duller. Ash- ish Sinha, his wife (fondly called Kavitaji), and their two children (Mallavi and Avanya) were a constant source of joy, and a welcome diversion from writing. Ajay Garde and his wife (fondly, Smitaji) were strong emotional pillars. Ajay and I share a strange sense of bondage, where we seem to go through similar pattern of ebb and flow. I must thank Aseem Inam, who was helpful in more ways that he probably real izes. Lastly, Jose Delgado, CAD Director of Facilities Management Services and Koichi Mera, Director of Center for International Policy, Planning and Development at USC, who were the main sources of my bread and butter, helped me by easing my workload at a crucial time. Irina Kotsinian, my other colleague at USC-FMS, pro vided a boost with her constant interest and provocative questions. Among my professors, Martin Krieger left a deep imprint in me. I learnt much from him to see beyond the obvious in making phenomenological explanations. Many v R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. of my arguments are drawn from imaginary conversations with him. I learnt from Tridib Banerjee how to identify common threads in seemingly disparate events, and to stitch a coherent argument out of them. Although both of them were not a part of my final dissertation committee, I am certain that my dissertation reflects some of their styles. I could not have asked for a better dissertation committee than that composed of Niraj Verma, Shui-Yan Tang, and Harry Ward Richardson. I owe Niraj much grati tude for his intellectual contribution to my dissertation. His keen sense of detail and his extraordinary capability to transform even the abstruse and vague ideas into clear and precise sentences helped me concretize my basic research ideas. Yan introduced me to the exciting world of “new institutionalism” and my analytical framework fun damentally draws on some of the ideas I derived from his class and from informal conversations with him. As my dissertation committee’s chair, Harry gave me much freedom in exploring various ideas. At the same time, he had a disarming way of en suring that the intellectual strength and rigorousness are maintained—if ideas did not excite him, he marked them as “boring.” I wish I could emulate his research capacity. Last, but not the least, this research endeavor would not have been worth the effort if it were not for my parents and Emel, my better half. My deceased mother al ways wanted the best effort from us, and my father taught never to stray from the path of honesty. I hope the dissertation does justice to both of them, to whom this disserta tion is dedicated. Emel entered my life in a fortuitous way in Rome. She exposed me to such facets of life that were only a dream before. At the same time, she grounded me in the reality of daily life-world. She has been my emotional reprieve in the lone some process of writing. She makes my dissertation worthwhile. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Contents Dedication ii Acknowledgements iii List of Tables x List of Figures xiii Abstract xiv 1. Introduction 1 1.1. Research Obj ectives 2 1.2. Research Background 5 1.3. Analytical Approach 10 1.4. Research Methodology 15 1.5. Organization of Chapters 23 2. Analysis of Housing Cooperatives 25 2.1. Organizations and Institutions 26 2.2. Endogenous Level Analysis of Housing Cooperatives 30 2.3. Exogenous Level Analysis of Housing Cooperatives 43 2.4. Significance of Institutional Structure 52 2.5. Conclusion 54 3. Evolution of Housing Cooperatives in India 57 3.1. Housing Cooperatives in the Pre-Independence Period (1904-1947) 57 3.2. Housing Cooperatives in the Post-Independence Period (1947- 62 present) 3.3. The National Institutional Framework 66 v ii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3.4 The Role of National Institutional Structure 80 3.5. Conclusion 87 4. Housing Cooperatives in Mumbai 89 4.1. Pre-Independence Period 90 4.2. Post-Independence Period 109 4.3. Mumbai’s Institutional Context, 1947-2000 115 4.4. Cooperative Institutional Structure in Mumbai 131 4.5. Diverse Roles of Housing Cooperatives 139 4.6. Conclusion 151 5. Housing Cooperatives in Chennai 153 5.1. Pre-Independence Period 154 5.2. Post-Independence Period 166 5.3. Chennai’s Institutional Context, 1947-2000 183 5.4. Cooperative Institutional Structure in Chennai 198 5.5. Conclusion 203 6. Differential Evolution of Housing Co-ops in Mumbai and Chennai 206 6.1. Endogenous Level Analysis 207 6.2. Exogenous Level Analysis 213 6.3. Role of Cooperative Institutional Structures 218 6.4. Policy Conclusions 221 7. Potential of Housing Cooperatives for Low-Income Households 226 7.1. Endogenous Level Analysis 227 7.2. Exogenous Level Analysis: The National Institutional Context 233 viii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 7.3. Exogenous Level Analysis: Cases of Mumbai and Chennai 239 7.4. Conclusion: Role of Institutional Structure 246 8. Conclusion 248 8.1. Analytical Framework 249 8.2. Institutional Conditions for Growth of Housing Cooperatives 251 8.3. Potential of Housing Cooperatives for Low-income Households 257 8.4. Significance of Institutional Structure 261 8.5. Housing Cooperatives in Sweden, Turkey, and United States 265 8.6. Conclusion 279 Bibliography 281 Appendix 1: Housing Cooperatives in New Delhi 299 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. List of Tables Table 1.1. Sources of Information (National level) 21 Table 1.2. Sources of Information (City level) 22 Table 3.1. Growth of Housing Cooperatives in Bombay and Madras Prov- 61 inces, 1927-28 to 1946-47. Table 3.2. Growth of Housing Cooperatives in India, 1950-2000 62 Table 3.3. Five year contribution of Housing Cooperatives to the housing 64 stock Table 3.4. Primary Cooperatives’ Activities in Eight States, 1994-95 65 Table 3.5. Investment in Housing by Plan Periods 69 Table 3.6. Phase I (1947-1970) Schemes and Role of Housing Cooperatives 71 Table 3.7. Phase II (1970-1988) Programs/ Organizations and Role of 74 Housing Cooperatives Table 3.8. Phase III (1988-present) Programs and Role of Housing 79 Cooperatives Table 3.9. Sources of Funds for Housing Cooperatives (percent share) 84 Table 4.1. Five Year Average Merchandise Export and Import through 92 Mumbai, 1871-1939 (Rupees, million) Table 4.2. Population Growth and Sex Ratio in Mumbai, 1849-1941 94 Table 4.3. Contribution of migration to population growth of Mumbai, 94 1901-1941 Table 4.4. Growth of Primary Housing Cooperatives in Maharashtra, 109 1960-61 to 1999-2000 Table 4.5. Employment in Greater Mumbai by Sector, 1961-1991. 117 Table 4.6. Contribution of migration to population growth of Mumbai, 118 1941-1981 Table 4.7. Population Growth of Greater Mumbai, 1941-2001 120 x R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 4.8. Population Growth of Mumbai Metropolitan Region, 1971-2001 121 Table 4.9. Sex Ratio in Mumbai, 1951-1991 123 Table 4.10. Residential Density of Greater Mumbai 123 Table 4.11. Maharashtra State Cooperative Housing Finance Corporation’s sources of funding, 1995-96 135 Table 4.12. Loans Disbursed by Maharashtra State Cooperative Housing Finance Corporation, 1960-61 to 1995-96. 136 Table 5.1. Population Growth of Chennai, 1872-1946 159 Table 5.2. Cooperative Building Societies in Madras State, 1945-46 to 1949-50 167 Table 5.3. Types of Housing Cooperatives in Tamil Nadu and their Principal Activities 169 Table 5.4. Growth of Housing Cooperatives in Tamil Nadu, 1949-50 to 1993-94 174 Table 5.5. Activities of Taluk Cooperative Housing Societies, 1979-80 to 1993-94 176 Table 5.6. Housing Cooperatives in Chennai, 1990-91 182 Table 5.7. Sector wise delivery of Housing Units in Chennai Metropolitan Area, 1981-1991 183 Table 5.8. Post-Independence Growth of Chennai’s Population, 1951 to 2001 186 Table 5.9. Post-Independence Growth of Households and Housing Stock in Chennai, 1961 to 1991 186 Table 6.1. Summary of Comparison between Mumbai and Chennai 222 Table 7.1. Income-group wise distribution of loans sanctioned by Apex Cooperatives, 1988-89. 238 Table 7.2. Income-group wise distribution of houses assisted by Apex 239 Cooperatives, 1988-89. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table Table Table Table 8.1. Tenure by share of housing stock in Sweden, 1945-1990 266 8.2. Share of Cooperative and Condominium Housing among owner 272 occupied units in United States 8.3. Comparison of Housing Cooperatives in Sweden, Turkey, United 275 States and India A -l. Growth of Housing Cooperatives in New Delhi, 1946-47 to 301 1995-96 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. List of Figures Figure 4.1. Sector-wise Contribution to Housing Stock in Greater Mumbai, 112 1956-57 to 1965-66 Figure 4.2. Sector-wise Contribution to Housing Stock in Greater Mumbai, 113 1973-74 to 1981-82 Figure 4.3. Sector-wise Contribution to Housing Stock in Greater Mumbai, 114 1982-83 to 1990-91 Figure 5.1. Urban/ Rural Distribution of Housing Cooperatives in Tamil 175 Nadu, 1949-50 to 1993-94. Figure 5.2. Growth of Membership in Housing Cooperatives, 1949-50 to 177 1993-94 Figure 5.3. Houses constructed by Housing Cooperatives, 1954-55 to 178 1993-94 Figure 5.4. Distribution of Housing Cooperatives by Type in Tamil Nadu, 179 1949-50 to 1993-94 Figure 5.5. Distribution of Membership by Type of Housing Cooperatives in 180 Tamil Nadu, 1949-50 to 1993-94 Figure 5.6. Construction of Flats/ Tenements in Chennai, 1981-82 to 1990-91 192 Figure 5.7. Proportion of Urban Housing Constructed with TNCHF funding 201 Figure 5.8. Proportion and Loan Amount disbursed by TNCHF to Primary 202 Housing Cooperatives in Chennai Figure 7.1 TNCHF funding and number of Housing Units constructed with 245 such funding in Chennai, 1982-83 to 1995-96 xm R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Abstract Housing cooperatives emerged in India in early twentieth century. While their activities were limited to Bombay and Madras provinces before India’s Independence in 1947, their activities grew phenomenally nationwide after Independence. However, the evolution of cooperatives differs between Mumbai and Chennai in the post- Independence period. Cooperatives’ activities grew dramatically in Mumbai, and have been used in diverse ways for low-income households. Cooperative’s activities grew modestly in Chennai, and have been hardly used for low-income households. I analyze the Indian experience to answer two questions: (i) What are the insti tutional conditions under which housing cooperatives grow? and (ii) What is the scope of cooperatives in fulfilling the housing objectives of low-income households? In their explanation of growth or decline of cooperatives, agency theorists focus on en dogenous organizational analysis and Marxist critiques focus on exogenous institu tional analysis. I propose that exclusive focus on either levels of analysis leads to bi ased policy conclusions. Arguing for a combined approach, I propose that the institutional structure plays a key role in reinforcing the development of housing cooperatives. At the organ izational level, the structure provides support to enhance cooperatives’ internal collec tive strengths, while mitigating attendant weaknesses. At the institutional level, the structure reduces transaction costs of formation and functioning of cooperatives. The structure enables growth of cooperatives by providing: (i) administrative, legal, and procedural support; (ii) access to finance; and (iii) access to land. In India, cooperatives were promoted nationally in the political economic con text of democratic socialism following Independence. Although India underwent eco nomic liberalization in the 1990s, elaborate national institutional structures enabled xiv R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. growth of cooperatives. In Mumbai and Chennai, cooperatives emerged in response to the specific local institutional conditions and housing market. In Mumbai, coopera tives emerged for collective ownership and management in response to multi-family housing development; in Chennai, they emerged as financing mechanisms for individ ual plotted housing development. Cooperatives’ autonomy and robust support struc tures in Mumbai helped to diversify their activities and to serve low-income house holds. State controls coupled with weak support structures in Chennai restrained co operatives’ growth and limited their use for low-income households. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1. Introduction Housing cooperatives emerged in India in the early twentieth century, and they were active principally in Bombay and Madras provinces until India’s Independence in 1947. Since Independence, cooperatives’ activities grew phenomenally nationwide, when housing cooperatives in other states were modeled after those in either of the two provinces. However, there are differences in the evolution of housing coopera tives in Mumbai (Bombay city) and Chennai (Madras city) in the post-Independence period. While cooperatives’ activities grew phenomenally in Mumbai, their activities grew modestly in Chennai. In the pre-Independence period, housing cooperatives hardly were hardly con sidered for serving the housing needs of low-income households. In the post- Independence period, Indian government policies explicitly promoted cooperatives as means for low-income households. However, utilization of cooperatives for low- income households differs between Mumbai and Chennai. In Mumbai, housing coop eratives have been utilized in multiple ways in both formal and informal sectors for low-income households; in Chennai, cooperatives have been hardly considered for low-income households. In my dissertation, I make an institutional analysis of the historical evolution of housing cooperatives in India. Examination of the extraordinary growth of housing cooperatives in the post-Independence period in general, and the differential evolution of housing cooperatives in Mumbai and Chennai in particular, holds broader policy lessons for expanding the role of cooperatives in the housing market, particularly for serving the housing needs of low-income households. The study is particularly sig nificant in the present context of international emphasis on “enabling” policies, where the focus is on improving organizational and institutional mechanisms in the housing 1 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. process. “Enabling” policies aim to reduce direct intervention of the state in the hous ing process, and increase the role of Third Sector organizations like housing coopera tives to fulfill the consequent housing gaps. Agency theorists take an endogenous organizational perspective in their expla nation of growth or decline of cooperatives. Their explanation is based on analysis of the comparative advantages of cooperatives in controlling agency problems. Marxist critiques like Elster (1989) take an exogenous institutional perspective in their expla nation. They examine the broader institutional constraints and opportunities in the evolution of cooperatives. Although analysis at both levels are indeed important, I argue that focusing exclusively on one level would lead to biased policy lessons. Hence, my analytical framework and research methodology combine both levels of analysis. In synthesizing the two levels, I propose that the cooperative institutional structure—that is, cooperatives’ manifest relationships with each other and with other organizations—is significant for the development of housing cooperatives. The rest of this Introduction is organized as follows. First, I outline the context of my research objectives. Second, I explain the significance of my research in the present housing policy context of “enabling” approach adopted by several countries including India. Third, I outline my analytical approach. Fourth, I explain my re search methodology. Fifth, I provide a layout of the chapters in my dissertation. 1.1. Research Objectives A brief background of the historical evolution of housing cooperatives is nec essary in order to explain the objectives of my research. Cooperatives emerged in In dia in the beginning of twentieth century. The British formally introduced coopera tives as rural credit mechanisms with the Cooperative Societies Act in 1904. As coop eratives came to be used for non-credit purposes, the Act was amended in 1912 to in- 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. elude other sectors. After the amendment, activities of housing cooperatives grew principally in Bombay and Madras provinces; cooperatives’ activities were quite lim ited in other provinces. In Bombay province, housing cooperatives were engaged in construction and management of multi-family housing (called “flats”). Cooperatives owned the housing, and each member had occupancy rights to the dwelling unit. By 1946-47, there were 167 such tenure based housing cooperatives. In contrast, housing cooperatives in Madras province were engaged in financing and construction of de tached housing in individual plots. Members owned the housing individually. By 1946-47, there were 128 such finance cooperatives. Until India gained Independence in 1947, cooperatives were hardly considered as vehicles for providing housing to low-income households. Since Independence, however, housing cooperatives were promoted nationally as vehicles for housing low- income households. By 1990, every state had a state-level secondary cooperative (called Apex cooperative) to mobilize finance for primary housing cooperatives. Housing cooperatives proliferated nationwide. The number of primary housing coop eratives in the country grew from 1,482 in 1950-51 to over 90,000 in 1999-2000. Dur ing the same period, their membership grew from 0.09 million to 6.5 million. The proportion of contribution of housing cooperatives to the national housing stock rose from less than 1 percent to nearly 10.8 percent. Since the states of Maharashtra, Gujarat (which were carved out of Bombay province) and Tamil Nadu (which was carved out of Madras province) had a past ex perience with housing cooperatives, they had an historical edge over other states. Other states followed either the Bombay or the Madras model of housing cooperatives. While many northern states (e.g. New Delhi, Rajasthan) adapted the Bombay model (i.e. tenure based cooperatives for multi-family housing), several southern states (e.g. 3 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Kerala, Andhra Pradesh) followed the Madras model (i.e. financial cooperatives for advancing loans). Despite an historical edge, housing cooperatives evolved differently in Mum bai (capital of Maharashtra) and Chennai (capital of Tamil Nadu). In Mumbai, hous ing cooperatives’ activities grew enormously over the last five decades since Inde pendence. Cooperatives assumed diverse roles in the housing market as the public, private and the Third sectors adapted them. In Chennai, however, the activities of housing cooperatives grew modestly over the last five decades. Cooperatives hardly diversified from their initial role as housing finance mechanisms; few tenure-based cooperatives emerged in Chennai. In 1994, while there were over 11,000 registered housing cooperatives in Mumbai, there were 100 housing cooperatives in Chennai. Formal housing cooperatives do not serve low-income households in either Mumbai or Chennai. Yet, in Mumbai, public agencies and non-government organiza tions capitalized on the diverse roles of cooperatives to use them as vehicles for pro viding housing to low-income households. For example, cooperatives have been used to define slum-dwellers’ groups for slum upgrading and redevelopment; to define groups for micro-finance; and to define informal groups among pavement dwellers for mobilizing collective action. In Chennai, cooperatives have been hardly used as simi lar vehicles for low-income households. Examination of the extraordinary growth of housing cooperatives in the post- Independence period in general, and the differential evolution of housing cooperatives in Mumbai and Chennai in particular, holds broader policy lessons on the role of hous ing cooperatives. I analyze the historical evolution of housing cooperatives to obtain policy insights into two related questions that have received little attention in housing research. The first question is—What are the conditions under which housing coop- 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. eratives grow? Answering this question allows us to understand the elements of insti tutional environment that favor or constrain the development of Third sector organiza tions like cooperatives. I answer the question by examining the conditions that lead to the phenomenal growth of housing cooperatives in India, and the conditions that lead to differential evolution of housing cooperatives in Mumbai and Chennai. The second question is—what is the scope of cooperatives in fulfilling the housing objectives of low-income households? Answering this question allows us to understand the internal organizational potential of housing cooperatives for serving the housing needs of low- income households. I answer the question by examining the utilization of housing co operatives for low-income households India in general, and the two cities of Mumbai and Chennai in particular. 1.2. Research Background Answering the aforementioned questions has gained significance in the present international context, when there is a renewed policy interest in Third sector organiza tions like housing cooperatives with the emergence of “enabling” approach since the late 1980s. Several developing countries including India adopted the approach follow ing the lead of international agencies (principally UNCHS and World Bank). The ap proach emerged essentially in response to the failure of conventional public sector strategies in addressing housing problems (UNCHS, 1996, p. 424), and it puts greater emphasis on institutional and organizational mechanisms in the housing process. Dif ferent versions of India’s National Housing Policies adopted during the 1990s reflect the basic principles of the approach. Three broad phases can be identified in the evolution of international housing policies (including India) since the 1950s, which culminated in the adoption of ena- 5 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. bling approach.1 In the first phase, until 1970, housing policies followed a “top- down” state centered approach, in which public sector agencies were directly involved in the housing provision. These strategies were essentially “project” based, in which public housing schemes were implemented along with slum clearance schemes. Dur ing the second phase, between 1970 and 1988, housing policies evolved in response to critiques like Turner who argued for a “bottom up” self-help approach.2 Strategies during this phase included decentralized projects like sites and services schemes and in-situ slum upgrading schemes. Policies were also aimed at improving the overall performance of housing sector, with a special emphasis on accessibility to housing fi nance. During these two phases, the public sector played an important role in housing, although the scale of its involvement reduced in the second phase. Direct government investment in housing was also relatively high during these two periods. The third phase of housing policies since 1988 is characterized by the enabling approach, which marks a shift lfom a “project” to a “process” based approach. Hous ing is considered as a part of the wider economy, along with other urban sectors (e.g. transportation, economic development, environment). In taking this “holistic” view, the role of public, private, and Third sector organizations is redefined according to their comparative strengths. Public, private and Third sector organizations are broadly distinguishable based on their organizational characteristics. While public and private sector agencies are based on the principles of vocation and income maximization respectively, Third sector organizations are broadly based on the principle of voluntarism (Brett, 1993). Public sector agencies are governed by the polity and they can invoke the coercive 1 Pugh (1994a, 1994b) provides a detailed account o f how the enabling approach evolved. 2 For example, see Turner and Fichter (1972). 6 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. power of law to commandeer resources (e.g. through eminent domain, zoning, taxa tion, preferential regulations). Private sector agencies are governed by the market and their resources depend on their profitability. Third Sector organizations are usually self-governing and mobilize their resources from various sources. Hence, “in the pub lic sector, this means bureaucratic hierarchy associated with democratic accountability and, in the private sector, profit and accountability to the market,” while voluntary or ganizations are private bodies motivated by “principles of obligation and reward based on solidarity or purposive incentives” (Brett, 1993, p. 276). Besides cooperatives, the Third Sector includes non-government organizations (NGOs), community based or ganizations (CBOs), grassroots organizations (GROs), non-profit organizations (NPOs).3 The organizational characteristics also give rise to specific sectoral failures. The public sector faces government failure, principally brought about by the rigidity of public bureaucracies (as exemplified in public housing inefficiencies). The private sector faces market failures, principally due to housing market externalities. Third sector organizations face voluntary failure, brought about by voluntary insufficiency, particularism, paternalism and amateurism (Salamon, 1987). In addition, Third sector organizations like housing cooperatives face collective action problems. In the enabling approach, the comparative strengths and failures determine each sector’s role. Thus, the public sector is viewed as a facilitator in the housing process rather than a direct housing provider. The government takes a “central role in 3 Although Third sector organizations are broadly distinguishable from public and private sector agencies, they are also not homogenous. They could differ in terms o f their membership characteristics, financial sources, and the groups they serve. For example, housing cooperatives are voluntary member organizations based on collective ownership and management; their finance is largely internally generated through their members. NGOs and NPOs are typically voluntary organizations funded by other agencies (public, private, or philanthropic) to provide housing related services. CBOs and GROs are loosely structured local associations to promote specific collective interests o f their members. See Vakil (1997) for classification problems o f NGOs. 7 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. setting the framework for urban development but a lesser role in providing the invest ment” (UNCHS, 1996, 424). It has the responsibility of setting up supportive institu tional frameworks for other actors (i.e. private and Third sector organizations) to func tion effectively. Thus, the enabling approach has an overall focus on institutional and organiza tional mechanisms in the housing process. On the institutional front, the enabling ap proach seeks to remove institutional constraints hindering the efficiency of housing markets, thus enhancing role of the private sector. Towards this end, World Bank’s policy paper on enabling housing markets (World Bank, 1993) identified several de mand and supply side institutional reforms to remove housing market distortions. On the organizational front, enabling approach seeks to build partnerships between the various sectors to synthesize “top-down” and the “bottom-up” approaches. The Habi tat II Conference in Istanbul (UNCHS, 1996) reinforced the formation of such partner ships to “integrate and mutually support objectives of broad-based participation through, inter alia, forming alliances, pooling resources, sharing knowledge, contribut ing skills and capitalizing on the comparative advantages of collective actions” (p. 22). The enabling approach is, however, not without its problems. It has been cri tiqued from two perspectives: (i) its overt emphasis on efficiency of housing market, and (ii) its neglect of political economic context in building partnerships. With respect to the first, Baken and Linden (1992) argue that housing markets are not efficient by their very nature. The supply and demand of land, for example, is different from other commodities—increase in its demand does not necessarily lead to an increase in its supply. Housing market is also rife with externalities. Hence, although private sector participation may increase, institutional reforms to enhance housing market efficiency may not necessarily lead to equitable housing conditions. With respect to the second, 8 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Sanyal (1998) critiques that the normative prescription of forming partnerships ne glects the political context in which such partnerships emerge. He argues that the mo tivations for such partnerships are “deeply influenced by the specific nature of the lo cal politics” (p. 84). Indeed, in the face of housing market inefficiency and retreat of the state from direct intervention, Third sector organizations like housing cooperatives have a greater onus in extending the welfare arm of the state to fulfil the housing objectives of those excluded from the formal housing market. While housing cooperatives and philan thropic organizations have been active for several decades, many newer Third sector organizations like NGOs and NPOs have emerged over the last two decades for filling the welfare gap. In this context, enabling policies need to explicitly take into account the extent to which Third sector organizations like housing cooperatives can indeed fulfil the traditional welfare need of housing low-income households, since they are also limited by their inherent sectoral failures. Further, enabling policies are hardly carried out in a vacuum. While Sanyal has a valid point in emphasizing the local political economic context, these policies are also intimately linked with the local institutional history. Housing cooperatives, for example, exist in both socialist and capitalist political economic systems, but both offer different incentives for cooperatives to emerge. The local institutional frame work is important for expanding or constraining the role of housing cooperatives. The local housing market characteristics are important for the extent to which there is a demand for such collective organizations. Thus, policies to enhance the role of Third sector calls for a contextual understanding of the institutional setting. Therefore, my analytical approach is explicitly institutional, taking into account the historical contin gencies that have shaped the evolution of housing cooperatives. 9 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.3. Analytical Approach Two distinctive approaches can be identified in the analysis of evolution of co operatives. The first approach is at micro level, where the focus is on endogenous or ganizational level analysis. The second approach is at the macro level, where the fo cus is on the exogenous institutional level analysis. Taking the first approach, agency theorists typically explain survival of organizations “in terms of the comparative ad vantages of the characteristics of the residual claims in controlling the agency prob lems of an activity” (Fama and Jensen, 1983, p. 345). In this vein, Alchian and Dem- setz (1972) highlight the intrinsic inefficiency of monitoring in collectively owned or ganizations like cooperatives.4 Similarly, in their comparison of cooperatives and condominiums, Goodman and Goodman (1997, p. 232) are suspicious of the growth of housing cooperatives in the U.S. due to their intrinsic disadvantages. According to them: Looking forward, it is difficult to project a resurgence for cooperative housing. Its market disadvantages relative to condominium housing appear to exceed its advantages. The relatively low valuations of co ops, and their declining market share, despite some cost advantages, suggest that demand for this form of housing falls considerably short of condo demand. In his analysis of growth of condominiums and cooperatives vis-a-vis rentals in the context of United States, Hansmann (1996) considers market contracting costs (i.e. costs due to market power (e.g. monopoly), lock-in, and contractual risks) and owner ship costs (i.e. costs of management (monitoring), decision making and ownership risks). In his framework, cooperatives emerge in a given context to the extent the costs of the collective ownership and operating in the market are minimized as com 4 Alchian and D em setz’s discussion o f collectively owned “socialist firms” in Yugoslavia (p. 787) is relevant in this context. 10 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. pared to other types of ownership. Yet, he finds that subsidies favoring ownership ac counted for the growth of cooperatives and condominiums in the United States (Hansmann, 1991). Taking the second approach, Marxist critiques like Elster set aside the internal problems of cooperatives as something that could be overcome. Instead, Elster argues that collective organizations like cooperatives are systematically discriminated in a capitalist economy. Elster (1989, p.96) claims, “it could be that the isolated coopera tive in an otherwise capitalist economy suffers difficulties that would not arise within a fully cooperative context.” Similarly, Chouinard (1990a; 1990b; 1989) also focuses on the institutional framework of the capitalist state in her analysis of the growth of housing cooperatives in Canada. Using a Marxist theory of “uneven development of capitalist states,” she argues, “changing relations and procedures within the state have played an extremely important role in regulating conflict over management and delivery of state-assisted cooperative housing” (Chouinard, 1990b, p. 1441). In her view, cooperative housing policies emerged in specific conjunctures of class conflicts, which are geographically contextual. In other words, politics of class struggles has been key to the development of cooperatives in Canada. In my analysis, I acknowledge that analyses at both levels are indeed important to understand the growth of housing cooperatives in India. At the external institu tional level, although it is arguable whether India’s post-Independence context of de mocratic socialism constituted a “fully cooperative context,” some of the difficulties raised by Elster were overcome as the state ideology provided a favorable environ ment for development of cooperatives. However, state support to cooperatives re duced over the last five decades, when India’s political economy transitioned towards 11 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. economic liberalization. In spite of the reduction in direct state support, housing co operatives flourished nationwide. Hence, state ideology and direct state support, by themselves, do not explain the post-Independence growth of housing cooperatives in India in general. Unlike Elster, I argue that the internal organizational features of cooperatives cannot be ignored. As collective organizations, cooperatives have both strengths and weaknesses in controlling agency problems, which endow them with both advantages and disadvantages as compared to other types of organizations. The demand for col lective organizations like cooperatives could depend on the local housing market char acteristics. Analysis at the organizational level is thus important for how the internal features of housing cooperatives are significant for responding to the specific needs of the local housing markets. Hence, my approach consists of both levels of analysis: the endogenous organ izational level and the exogenous institutional level.5 I make this analytical difference of levels to understand the relationship between organizations as endogenous variable and institutions as exogenous environmental variable. In making these two levels of analysis, I take the approach of “new institutionalism” to make a deliberate distinction between organizations and institutions. I understand organizations as groups of indi viduals who come together for achieving some common objectives; institutions refer 5 M y approach is comparable with Keogh and D ’arcy’s (1999a p. 2407) three-level hierarchy in the institutional analysis o f property markets: “At the topmost level, the property market exists within an institutional framework defined by the political, social, econom ic and legal rules and conventions by which the society in question is organ ised. At the next level, the property market itself can be considered as an institution with a range o f characteristics which determine its structure, scope and function. Finally, at the bottom level, the main organisations which oper ate in the property market can be considered in terms o f the way they are structured and the way they change. The relationship between institutions and organisations at each level, and between levels, can best be described as inter active, defined in relation to one another and capable o f change in response to action and experience.” In my analysis, the first two levels o f Keogh and D ’Arcy are collapsed into the exogenous level analysis. 12 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. to the broader framework of informal and formal rules in which organizations evolve (Chapter 2). Some analysts do not make the distinction—they subsume organizations under the rubric of institutions. Distinguishing organizations and institutions allows me to explore the relationship between the internal organizational features and the external institutional framework (e.g. broader policies, laws) in the evolution of housing coop eratives. Organizations and institutions have a two-way relationship: while institu tions influence the type of organizations that evolve in a particular context, organiza tions also influence institutional change. Institutions form the incentive framework in which organizations evolve; in turn, organizations act as agents of institutional change while pursuing their objectives. At the endogenous level, organizational characteristics endow specific strengths and weaknesses for housing cooperatives to achieve their objectives. As col lective organizations, housing cooperatives have distinctive internal governance fea tures, which endow them with both advantages and disadvantages in achieving their housing objectives. Members come together to form cooperatives based on the advan tages that these collectives hold for them. At the exogenous level, institutions form the framework of constraints and op portunities in the development of housing cooperatives. Institutions could lower or raise the transaction costs of housing cooperatives as compared to other organizations. The nature of such institutions is contingent on the broader political economic context. In particular, the state has an important role in the design and implementation of for mal institutions (i.e. laws and policies). State ideology, for example, has a crucial im pact on policies towards cooperatives. Local housing market characteristics gain sig 13 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. nificance for the demand of such collective organizations, and the structural con straints at the local level. In combining the two levels of analysis, I hypothesize that an enabling institu tional structure plays a crucial role in the evolution of housing cooperatives. Institu tional structure refers to the manifest arrangement o f relationships between organiza tions. With respect to housing cooperatives, the supportive institutional structure re fers to the mutually reinforcing relationships of cooperatives with each other and with other organizations. Cooperative institutional structure could thus include a hierarchal organization of cooperatives (e.g. tertiary, secondary, and primary cooperatives), or a horizontal relationship between cooperatives (e.g. federation of primary cooperatives). The cooperative institutional structure could also span horizontal relationships with other public, private, or Third sector organizations. The institutional structure could embody administrative, financial, and other functional relationships. Furthermore, institutional structure could be formal or informal. While a formal structure is based on legally binding relationships (e.g. contractual relationships) and is well defined, informal structure is based on customary relationships (e.g. based on reciprocity, trust, promise) and is loosely defined. I propose that the cooperative institutional structure reinforces the growth of cooperatives’ activities by providing mutually supportive relationships. In the absence of the supportive institutional structure, cooperatives’ activities are likely to decline over time. The institutional structure is contingent on both the internal organizational features and the external institutional framework. At the internal organizational level, enabling institutional structure enhances the internal organizational strengths, w h ile overcoming attendant internal weaknesses. The structure provides the reinforcing mechanism for housing cooperatives to persist and to expand their activities. The 14 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. supportive structure also offers scope for tapping the internal collective strengths of the cooperative for low-income households. At the external institutional level, the institutional structure reduces the trans action costs of the formation and functioning of housing cooperatives. The form of the institutional structure also depends on the institutional context. In India, the state played a key role in setting up an elaborate cooperative institutional structure in India. The structure was established in the political economic context of democratic social ism. Once set-up, the structure enabled growth of housing cooperatives despite reduc tion of direct state support with India’s political economic transition towards economic liberalization. In other institutional contexts, the cooperative institutional structure need not necessarily be established by the state; such structures could emerge with lit tle state support. 1.4. Research Methodology I test the significance of the institutional structure by making an institutional analysis of historical evolution of housing cooperatives in India in general, and in Mumbai and Chennai in particular. The analysis is useful in identifying the institu tional conditions in which housing cooperatives have evolved in the two cities. Identi fying such conditions holds broader policy lessons for enhancing cooperatives’ activi ties in other contexts. The methodology is necessarily historical in being sensitive to the events influencing the evolution of housing cooperatives. Indeed, in “new institutionalism,” history is significant for explaining diver gent institutional evolution, although it is not unified in its methodological approach with respect to how institutions originate, how they persist, and why they change. In their seminal article, Hall and Taylor (1996) identify three strands of “new institution alism,” which approach the historical feature differently: historical institutionalism 15 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. (exemplified by the works of Immergut (1992); Thelen and Steinmo (1992); Skocpol (1995, 1985)); rational choice institutionalism (exemplified by the works of North (1981); Riker (1980); Shepsle (1989)); and sociological institutionalism (exemplified by the works of Berger and Luckmann (1966); DiMaggio and Powell (1983); Zucker (1987)). While rational choice institutionalists are deductive in their approach, the other two are inductive. Historical institutionalists emphasize path dependence and unintended conse quences in institutional development. According to them, “institutions structure po litical interactions and in this way affect political outcomes” (Thelen and Steinmo, 1992, p. 13); hence, similar policies could produce different results due to differences in contextual features. Rational choice institutionalists focus on methodological indi vidualism, whereby individuals act strategically. Institutions evolve in the strategic interactions (games) to reduce transaction costs and to mitigate principal-agent and collective action problems. Institutional change is path dependent since it is character ized by increasing returns to scale, thus producing lock-in effects. Sociological insti tutionalists take a deeply cultural approach to include contextual symbols and meaning in their approach to institutions. According to them, “institutions do not simply affect the strategic calculation of individuals, as rational choice institutionalists contend, but also their most basic preferences and very identity” (Hall and Taylor, 1996, p. 948). Thus, choices of new institutions may be based on social legitimacy, rather than effi ciency considerations. History becomes significant in this approach as the focus is on “processes whereby those developing new institutions ‘borrow’ from the existing world of institutional tem plates” (Hall and Taylor, 1996, p, 953). Methodologically, my approach is similar to that used by Greif (1994, 1998; 2001) and Evans (1995) who span rational choice and sociological traditions to em- 16 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. phasize the significance of historical context in comparative institutional analysis. Greif (1994) uses comparative institutional analysis to examine the divergent evolu tion of the twelfth century Maghribi and Genoese society, and Evans (1995) uses comparative institutional analysis to examine the role of the state in the growth of high technology industries in Brazil, South Korea, and India. Greif (2001) argues for com bining inductive and deductive approaches, wherein “history is necessary to identify the relevant institutions and theory is required to deductively restrict hypotheses, re veal lines of causation, and generate predictions that can be compared with the evi dence.” Inductive approach is required for establishing “the relevance of particular institutions based on evaluating and synthesizing micro-level, historical and compara tive evidence and insights from context-specific, micro-theoretical models” (Greif, 1998). Stressing the need for sensitivity to context in comparative institutional analy sis, Evans (1995, p. 29) also argues, “a comparative analysis that starts with contextual differences and then looks for underlying regularities is the only way to proceed.” Al though Evans is not explicit in combining inductive and deductive approaches, he ar gues (Evans,1995, p. 19): A comparative institutional approach implies a strategy of gathering evidence. Obviously, one central aim is to collect evidence that will locate specific state policies and societal responses in the larger institu tional context that produces them, showing how that context helps de fine interests, aspirations, and strategies. At the same time, demonstrat ing variation across cases requires delving into specifics. Whether the focus is on society or within the state, the central methodological pre cept of a comparative institutional approach is to ground assertions of institutional effects in the analysis of the actions of specific groups and organizations. In my approach, I combine the inductive and deductive approaches to identify the relevant historical contextual factors in the evolution of housing cooperatives in 17 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Mumbai and Chennai. My historical review is neither stylized nor descriptive; it is a selective, yet rich, history with the purpose of identifying the underlying contextual patterns in the evolution of housing cooperatives in the two cities. The contextual dif ferences and similarities are used for analyzing the role of institutional structure in en hancing the development of housing cooperatives. Methodologically, Skocpol and Somers (1994) call this “contrast of contexts,” where the focus is on contextual par ticularities. I draw broader policy lessons through comparative juxtaposition of the historical evolution of housing cooperatives in the two cities. I combine both endogenous organizational level analysis and exogenous insti tutional level analysis for understanding the role of institutional structure. At the en dogenous organizational level, the internal characteristics of the housing cooperatives are analyzed for the role of cooperatives under the specific institutional conditions. At the exogenous institutional level, the contextual features of the political economic conditions and housing market characteristics are analyzed for why cooperatives evolved in the way they did. The two levels of analysis are combined for how specific institutional structures emerged in response to the endogenous organizational charac teristics and exogenous institutional conditions to reinforce the evolutionary path of housing cooperatives. The analytical framework provides the basis for consistency of comparative study of housing cooperatives across the two cities. There are two levels of compari son in the evolution of housing cooperatives: the organizational (internal to housing cooperatives) and the institutional (external to housing cooperatives). The first level of analysis explains the role of housing cooperatives w ithin an institutional context; the second level of analysis explains the institutional conditions in which housing co 18 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. operatives evolved. The organizational role of housing cooperatives is considered as a dependent variable, and the institutional context is considered as independent variable. At the first level, the endogenous characteristics of cooperatives gain signifi cance for their potential to serve low-income households. I examine membership characteristics of housing cooperatives in the two cities to analyze this potential. While cooperatives offer potential for low-income households, they also face weak nesses in achieving the potential. At the institutional level, there are two facets of comparison: (i) the broader political economic conditions, including state’s role in the formal institutional framework; and (ii) the local housing market characteristics. These two aspects are analyzed from a historical perspective. I obtained information on the role of housing cooperatives and the conditions under which cooperatives evolved from a variety of sources. I conducted my field re search in Chennai, Mumbai, and New Delhi over a span of nearly ten months in 1997. In each of these cities, I interviewed key informants on the state policies and the evo lution of housing cooperatives. Information gathered through interviews was supple mented using government documents, statistical data, and secondary literature on housing cooperatives. Tables 1.1 and 1.2 list the various cooperative, government, and non-government organizations from which I obtained information on housing co operatives at the national and city levels respectively. To obtain information on the broader political economic conditions influencing the evolution of housing cooperatives nationally, I have relied on government policy documents, Annual Reports of Cooperative Registrars, and secondary literature. For evolution of housing cooperatives during pre-Independence period, I used secondary research and other historical accounts. For information on broader role of state and national policies on housing cooperatives in post-Independence period, I have used 19 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Government of India’s Five Year Plan documents. More specific information on housing policies and the role accorded to housing cooperatives were obtained from government reports and policy documents. I compiled national statistics on housing cooperatives for various years from Statistical Statements relating to Cooperative Movement in India (published by Reserve Bank of India until mid 1970s, and subse quently by National Agricultural Bank and Rural Development). With respect to the Mumbai and Chennai, I gleaned information on housing cooperatives through interviews with several key informants, who included state/ city government officials, policymakers, Cooperative Registrars, and officials of local housing authorities, metropolitan development agencies, housing finance agencies, non-governmental organizations, Apex cooperatives, and primary housing coopera tives. I obtained historical information on the state policies, local housing market, and the role of housing cooperatives through secondary sources. I compiled statistical evi dence on housing cooperatives for each city based on information from Cooperative Registrar’s office and the Annual Reports. Annual Reports of the state level Apex co operatives provided useful information with respect to financing. There is scarce data on income groups served by housing cooperatives, since income data with respect to cooperatives are hardly available. Hence, I have depended on various primary and secondary sources for this information. I have used qualitative ordinal classification of income groups (i.e. low income, middle income, and high in come) rather than a precise classification based on monthly income used by Govern ment of India. Qualitative descriptions of the membership characteristics have been used to infer the potential o f housing cooperatives to serve low -in com e households. I was affiliated with a reputed local research institute in each of the three cities during my field research. In Chennai, I was affiliated with Madras Institute of Devel- 20 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. opment Studies; in Mumbai, I was affiliated with the Indira Gandhi Institute of Devel opment Research; and in New Delhi, I was affiliated with the National Institute of Public Finance and Policy. The affiliation provided me with intellectual and logistical support for my research. The affiliated institutes provided me with an opportunity to interact with local researchers and experts to test out my ideas. They also provided me with considerable library support and access to government documents. In each city, I presented my findings in a seminar in the affiliate institute to verify my findings. The seminar discussions were most useful in crystallizing my research ideas. Table 1.1. Sources of Information (National level) National Level Cooperative Or ganizations National Cooperative Housing Federation (NCHF) National Cooperative Union of India (NCUI) Government Or ganizations Ministry of Urban Development Non-government Organizations Society for Development Studies, New Delhi Housing Finance Agencies National Housing Bank (NHB) Housing and Urban Development Corporation (HUDCO) Housing Finance and Development Corporation (HDFC) 21 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 1.2. Sources of Information (City level) Organizations Mumbai Chennai Cooperative Or ganizations Apex Cooperatives Tamil Nadu Cooperative Housing Federation (TNCHF) Maharashtra State Cooperative Housing Finance Corporation Limited (MSCHFC) District Cooperative Fed eration Mumbai District Cooperative Housing Federation — Registrar’s Office Office of Cooperative Registrar (various districts) Office of Cooperative Registrar (Housing) Cooperative Educational Institute ___ Natesan Institute of Cooperative Management Government Or ganizations Local Housing Agency Maharashtra Housing and Area Development Authority (MHADA) Tamil Nadu Housing Board (TNHB) Metropolitan Agency Mumbai Metropolitan Regional Development Authority (MMRDA) Chennai Metropolitan Develop ment Authority (CMDA) Non-Governmental Organizations National Slum Dwellers Federa tion (NSDF); Mahila Milan; Society for Promotion of Area Re sources Center (SPARC) Center for Development, Madras (CEDMA) to to 1.5. Organization of Chapters The dissertation is divided into eight chapters. The first chapter is this intro ductory chapter, which provides the research context, the analytical framework, and the research methodology. The second chapter provides an institutional analysis of housing cooperatives. In this chapter, I analyze the internal organizational characteris tics of housing cooperatives with respect to their governance structure, housing objec tives, and membership characteristics. Then, I analyze housing cooperatives at the exogenous level, including the influence of institutional framework and local housing market on the development of cooperatives. In the third chapter, I examine the growth of housing cooperatives in the Indian national context. I first analyze the growth of housing cooperatives in the pre- Independence period. Then I analyze the development of housing cooperatives in the post-Independence period. I identify three phases of political economic context in In dia, when housing policies and the role of cooperatives distinctively changed. In Phase I (1947 to 1970), state policies emphasized directly involvement in the housing process in the context of democratic socialism; in Phase II (1970 to 1988), state poli cies emphasized creation of other autonomous agencies for housing finance; in Phase III (1988 to present), the policies aimed at retreat of direct state intervention in the context of economic liberalization. Chapters 4 and 5 examine the development of cooperatives in Mumbai and Chennai respectively. I trace the pre-Independence historical roots of urban and hous ing development, which shaped the distinctive evolution of housing cooperatives in the two cities. Then, I analyze the growth of housing cooperatives in the post- Independence period. In Chapter 6 ,1 make an institutional analysis of the differential evolution of housing cooperatives in the two cities. I compare the two cities to iden- 23 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. tify the institutional elements that aided or constrained the development of coopera tives. Chapter 7 deals with the institutional potential of housing cooperatives to reach low-income households. The chapter summarizes the experience of India in general, and the two cities of Mumbai and Chennai in particular to draw broader lessons on the scope of cooperatives to serve the housing needs of low-income households. Chapter 8 concludes with the policy prescriptions gleaned from the study. These policy pre scriptions are compared with the historical evolution of cooperatives in three countries of Sweden, Turkey, and United States. In this, I examine if the policy conclusions conform with the conditions that lead to phenomenal growth of housing cooperatives in Sweden and Turkey, and the modest role of cooperatives in United States. 2 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2. Analysis of Housing Cooperatives In this chapter, I analyze housing cooperatives at the two levels identified in Chapter 1: (i) the endogenous organizational level, and (ii) the exogenous institutional level. I differentiate between organizations and institutions in making the analysis. In my analysis, organizations are considered as groups of individuals bound by some common purpose to pursue certain objectives, and institutions are considered as the matrix of rules in which organizations like housing cooperatives evolve. At the endogenous organizational level analysis, I take into account three in ternal features of cooperatives: their governance structure; their membership charac teristics; and the nature of their housing objectives. The governance structure of co operatives is characterized by collective ownership and democratic management. Such a governance structure holds both prospects and problems for members to form or join cooperatives. Lastly, different types of housing cooperatives could emerge based on the characteristics of housing objectives (e.g. asset specificity). I identify three types of housing cooperatives in India: Building Cooperatives, Finance Coopera tives, and Tenure Cooperatives. At the exogenous level, I consider two external features affecting housing co operatives: the institutional framework and the local housing market. The institutional framework forms the incentive framework in which cooperatives evolve, and it deter mines cooperatives’ transaction costs. The local housing market characteristics shape the specific types of cooperatives that evolve in a particular context. In combining the two levels of analysis, I propose that the cooperative institu tional structure plays a key role in fostering housing cooperatives. Institutional struc ture refers to the manifest relationships of housing cooperatives with each other and 25 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. with other organizations. A supportive institutional structure could reinforce the growth of housing cooperatives by lowering their transaction costs. The chapter is organized in the following way. First, I expand on the key as pects of organizations and institutions to give a background to the two levels of analy sis. Second, I analyze housing cooperatives from an internal organizational perspec tive. Third, I analyze housing cooperatives from an external institutional perspective. Fourth, I outline the significance of institutional structure for housing cooperatives. Fifth, I conclude with a summary of analysis of housing cooperatives. 2.1. Organizations and Institutions I draw on rational choice institutionalists, particularly Coase, North, and Wil liamson to distinguish between institutions and organizations. Coase’s two seminal articles were instrumental in explaining how organizations emerge and why institu tions are significant. In his first article, Coase established that organizations arise in order to internalize certain marketing costs (Coase, 1937). In his second article, he proposed that in a world of zero costs of market transactions (simply, transaction costs), the initial delimitation of legal rights would not matter as they can be re arranged through the market at no cost (a proposition now known as Coase Theorem) (Coase, 1960). However, in a world of positive transaction costs, the institutional framework gains significance, which is, in turn, significant for organizations to emerge.1 In the housing market, for example, transaction costs are “partly the market costs—such as legal fees, realtor fees, title insurance, and credit rating searches—and partly the costs of time each party must devote to gathering information, to searching, 1 The optimality o f initial arrangement o f legal rights matters since “the costs o f reaching the same result by alter ing and combining rights through the market may be so great that this optimal arrangement o f rights, and the greater value o f production which it would bring, may never be achieved” (Coase, 1960, p. 16). 26 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and so forth” (North, 1990; p. 62). For Williamson, transaction costs are the frictional costs of exchange; for Barzel (1997) they are costs “associated with the transfer, cap ture, and protection of rights” (p. 4). Transaction costs are broadly costs—tangible and intangible—incurred in facilitating social exchange. North (1990) and Williamson (1985) extend Coase’s argument from a macro • • • .9 and micro-analytic perspective respectively. North (1990, p. 5) argues that institu tions are constraints that reduce uncertainty by providing a structure to everyday life. The institutional framework affects transaction costs of organizations. Organizations, as opposed to institutions, are “groups of individuals bound by some common purpose to achieve objectives” (North, 1990, p. 5) defined by opportunities afforded by the in stitutional framework. Unlike North, whose focus is on institutions affecting transaction costs, Wil liamson’s focus is on the transaction cost analysis of organizations that involves “comparative costs of planning, adapting, and monitoring task completion under alter native governance structures” (Williamson, 1985, p. 2, my emphasis). For him, or ganizations are thus distinct from production functions—they are governance struc tures. Governance structure influences an organization's monitoring characteristics and distinguishes one type of organization from another (Williamson, 1985; Fama and Jensen, 1983). In Williamson’s terms, asset specificity is important for why organiza tions evolve. Asset specificity refers to “durable investments that are undertaken in support of particular transactions, the opportunity cost of which investments is much lower in best alternative uses or by alternative users should the original transaction be prem aturely term inated” (W illiam son, 1985, p. 55). W illiam son (1996) identifies six forms of asset specificity—site specificity, physical asset specificity, human asset 2 See W illiamson (1996, p. 93). 27 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. specificity, dedicated assets, brand name, and temporal specificity. For Williamson, forming organizations to incur the asset specific investments could be more efficient than undertaking such investments through the market. While markets are character ized by discrete transactions, organizations internalize such transactions within their governance structure. Thus, Williamson argues that organizations provide non-market alternatives for achieving objectives. I draw on the above authors to distinguish between organizations and institu tions. Similar to North, I understand organizations as groups o f individuals bound by -3 some common purpose to pursue certain objectives. They are collective entities that “perform a function when there are common or group interests, and though organiza tions often also serve purely personal, individual interests, their characteristic and pri mary function is to advance the common interests of groups of individuals” (Olson, 1965, p. 7). Housing cooperatives are thus organizations that emerge to fulfill the common housing objectives of their members. Institutions form the matrix of rules that act as constraints as well as opportu nities for organizations like housing cooperatives evolve.4 Rules are prescriptions that “refer to which actions are required, prohibited, or permitted” (Ostrom, 1986, p. 5). Institutions are both formal and informal in nature. Formal institutions are usually well defined, and include laws, guarantees, standards, and public policies; informal 3 The concept o f organization is less controversial than that o f institutions; various approaches emphasize similar dimensions o f collectivity, common purpose, and objectives/ goals. Etzioni (1964, p.3), for example, argues, “or ganizations are social units (or human groupings) deliberately constructed and reconstructed to seek specific goals.” Scott (1998) approaches organizations from three perspectives— rational system, natural system, and open system. In all three, organizations are considered as collectivities or coalitions with a common goal or interest. Their governance structures, however, differ. In the first, they are relatively formalized, in the second, they are informally structured, and in the third, they are influenced by environmental factors. 4 This is a very particular approach to institutions; alternative approaches highlight different dimensions. Schotter (1981, p. 155), for example, argues that social institutions are “not rules o f the game, but rather the alternative equi librium standards o f behavior or conventions o f behavior that evolve from a given game described by its rules.” U phoff (1986, p. 9) similarly argues that “institutions are com plexes o f norms o f behavior that persist over time, by serving collectively valued purposes.” 28 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. institutions are more tacit that evolve in a particular social context (e.g. socially sanc tioned norms, codes of conduct, conventions, shared belief systems, and customs). In the context of housing, institutions mediate the housing market through a matrix of laws, rules, traditions, and conventions governing construction, property ownership, buying and selling of property, and so on. These institutions affect the types of hous ing organizations that evolve and the performance of such organizations: Realtors, title insurance, credit bureaus, and savings and loan associa tions that affect the mortgage market all will be influenced. The effi ciency of these organizations is a function of the structure of property rights and enforcement and of the capital market (including voluntary as well as governmental guarantees and subsidies and other instruments that exist in the capital market). (North, 1990, p. 63) These institutions are contingent on local culture and history and vary across contexts. Property rights and tenure, for example, are specific to a local housing market and evolve over time. Keogh and D’Arcy (1999a, p. 2407) argue that the property market “can be considered as an institution with a range of characteristics which determine its structure, scope and function.” Based on the above distinction between organizations and institutions, I ana lyze housing cooperatives at two levels: the endogenous organizational level, and the exogenous institutional level. Agency theorists have generally focused on the organ izational level, while Marxist critiques have focused on the institutional level in the analysis of cooperatives. As I highlighted in Chapter 1, neither levels of analysis can be ignored in the development of housing cooperatives in India. Thus, my analysis recognizes the importance of both levels of analysis. At the endogenous organizational level, three key dimensions gain significance for the analysis of housing cooperatives. The first dimension, as Williamson high lights, is the governance structure, which distinguishes cooperatives from other organ- 29 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. izational forms. The second dimension is cooperative’s membership characteristics, including the motivation of members to join or form cooperatives and the incentives that cooperatives offer. In this, I examine the strengths and weaknesses of coopera tives from members’ viewpoint. The third dimension is the nature of housing objec tives—different types of housing cooperatives could emerge based on the housing ob jectives. The characteristics of these types also differ based on the asset specificity implied by the objectives. At the exogenous institutional level, two related aspects require examination for the development of housing cooperatives: the institutional framework and the lo cal housing market characteristics. The institutional framework forms the incentive structure in which housing cooperatives are embedded. In this, the political economic context gains significance for the extent to which the institutional framework is favor able to the growth of housing cooperatives. The characteristics of local housing mar ket determine the specific institutional conditions under which housing cooperatives emerge. In this, the historical context of the local housing market gains significance for the emergence of housing cooperatives. 2.2. Endogenous Level Analysis of Housing Cooperatives As I outlined before, there are three key dimensions in the endogenous organ izational level analysis of housing cooperatives. The first dimension is the governance structure. Housing cooperatives are based on distinctive principles of collective own ership and management. These collective principles endow them with comparative internal strengths and weaknesses that affect their performance in achieving their ob jectives relative to other organizations. The second dimension is the membership characteristics. Different organiza tions offer different incentives for members to achieve their objectives, and members 30 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. would join or form cooperatives to the extent that they can achieve their objectives more efficiently than through other organizations. In this, the strengths and weak nesses of cooperatives from members’ viewpoint are particularly important for under standing for why households would form or join cooperatives. The third dimension is the nature of housing objectives. The housing process consists of both production and consumption, which encompass a broad range of hous ing objectives (construction, land development, finance, tenure, management). Three types of housing cooperatives could be identified in India based on the housing objec tives: Building Cooperatives, Tenure Cooperatives, and Finance Cooperatives. The asset specificity implicit in these three types varies, which affects their stability. 2.2.1. Cooperative’s Governance Structure Cooperative principles emerged with the first cooperative store established in Rochdale, England in 1844. Initially, these principles included: open membership; democratic voting; cash trading; membership education; political and religious neu trality; no unusual risk assumption; limited interest on stock; goods sold at regular re tail prices; limitation on the number of shares that can be owned by members; distribu tion of net savings to members according to their patronage; and regular auditing and meetings (Birchall, 1997). Subsequently, the International Cooperative Alliance (ICA) formed in 1895 modified and adopted Rochdale principles to define an interna tionally accepted meaning of the cooperative. The ICA has since become a major in ternational organization in influencing the cooperative movement worldwide, and its tenets have been broadly accepted in India. The cooperative principles stipulated by ICA serve a useful background for analyzing the governance structure of housing cooperatives. The principles evolved over a series of ICA Congresses held in 1937, 1966 and 1995. Following the last Con- 31 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Congress, ICA defined a cooperative as “an autonomous association of persons united voluntarily to meet their common economic, social, and cultural needs and aspirations through a jointly-owned and democratically-controlled enterprise.” Additionally, the cooperative is based on values of “self-help, self-responsibility, democracy, equality, equity, and solidarity. .. .[C]o-operative members believe in the ethical values of hon esty, openness, social responsibility, and caring for others” (ICA, 1995). There are seven cooperative principles prescribed by the ICA. First, they are voluntary in nature and have open membership. Second, they are democratically con trolled membership organizations. Third, they entail member economic participation, with each member contributing equitable share capital. Fourth, cooperatives are autonomous and independent organizations, where the decision making process is con trolled by members. Fifth, cooperatives provide education, training, and information to members. Sixth, cooperatives cooperate with each other for establishing wider in stitutional structures for enhancing cooperative movement. Last, cooperatives entail a concern for community. The above principles are, however, not uniformly exhibited in all cooperatives. Indeed, some of the principles could be undermined in many contexts. In many Indian states, for example, overt government intervention in the day-to-day operations limits the autonomy of cooperatives. Yet, two of the above principles are central to most cooperatives: collective ownership and democratic management. Shah (1996) also suggests, “that an enterprise is owned and controlled by its member-users should, in our judgement, be viewed as the core and the defining feature of a co-operative” (p. 33). Collective ownership entails that members jointly own resources of the coopera tive. Members collectively pool their investment resources in the cooperative for achieving certain common objectives. Staber (1992, p. 1193) argues, “what the differ- 32 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ent types of cooperatives have in common, and what distinguishes them from conven tional organizations, is that they pool the capital investment on which they are built from among their members who thus ‘own’ the organization.” A members’ invest ment is often in the form of share capital, but it could also be in the form of material inputs, labor, or land. Collective management entails participation of members in monitoring the co operative. They are self-governing organizations, owned and managed by members as a group. Members share cooperative’s benefits so that they all are “residual claim ants” (i.e. share benefits) in Alchian and Demsetz’s (1972) terms. Management deci sions are taken in a democratic way i.e. they are based on “one member, one vote” principle, irrespective of the number of shares owned by a member. Each member has an equal voice in the decision making process. In this, cooperatives differ from other shareholding firms where decisions are made based on a “one share, one vote” princi ple. In terms of housing, cooperatives entail collective ownership and management of one or more resources according to their housing objectives. There are several di mensions of housing along which cooperatives could define their objectives, e.g. land, building materials, construction, financing, maintenance, management, and service provision. In self-help cooperatives, members pool their labor resources for planning and construction. In cooperatively owned housing, the superstructure, land, or both are collectively owned or leased by members. In tenant management cooperatives, residents come together for housing maintenance and management. 2.2.2. Membership Characteristics Analysis at the membership level allows insights into the motivations of indi viduals to form or join cooperatives, including the incentives that the cooperatives of- 33 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. fer for the members. Since one of the basic principles of cooperatives’ governance structure is open membership, cooperatives are not explicitly oriented towards any particular group based on income, or any other affiliation. They could be potentially formed by any group of individuals who can derive mutual housing benefits from the collective. For example, housing cooperatives could be formed by individuals from a range of income groups—from high to low. High-income households could form co operatives to obtain exclusivity in housing (Hansmann, 1996). Low-income house holds could also form cooperatives for obtaining any cost-savings that the collective may offer. Individuals with similar lifestyles and tastes could form cooperatives to preserve their common outlook (e.g. in intentional communities). Furthermore, coop eratives could be vehicles for group living in a cultural or ideological context (e.g. co- housing schemes in Scandinavian countries). The motivation for individuals to form cooperatives could thus differ. In this context, it is important to understand the strengths and weaknesses that cooperatives hold for members to join or form cooperatives. Cooperatives hold sev eral strengths as collective organizations. First, cooperatives entail pooling of re sources in the collective (Shah, 1996; Staber, 1992), so that members benefit from economy of scale. There are several dimensions of housing—land, building materials, construction, financing, maintenance, management, and service provision—where the collective pooling offers such economy of scale. For example, building material, con struction, maintenance and management costs could be shared among households. Second, cooperatives entail “peer monitoring” (Stiglitz, 1990); such social sanctions may make cooperatives a superior alternative to conventional banks for those with little credit or access to collateral (Baneijee, Besley, and Guinnane, 1994). 34 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In small cooperatives, members know each other’s background information, which is also effective for group lending and microcredits (Guinnane, 2000, 2001; Rogaly, 1998). Third, cooperatives foster community action (Clapham and Kintrea, 1995), community empowerment (Saegert and Winkel, 1996), and act as organizational vehi cles to foster collective action and self-help (UNCHS, 1989). Many government agencies and NGOs have used cooperatives as a means for fostering community action and self-help among slum-dwellers (for example, see Panwalkar (1995)). Fourth, cooperatives could emerge to internalize housing externalities. For ex ample, in multi-family housing, externalities could arise due to common spaces, com mon land, interdependence among tenants for building repairs and services, and so on. Cooperatives internalize such externalities through collective tenant ownership and management (Williams, 1993). Thus, cooperatives have emerged as alternatives to public housing management in the United States and United Kingdom to the extent they are able to self-manage and internalize the rental externalities that otherwise arise in public housing (Birchall, 1995; Clapham and Kintrea, 2000; Rohe, 1995; Rohe and Stegman, 1995; Miceli, Sazama, and Sirmans, 1994; Miceli and Sazama, 1998). In spite of their strengths, however, cooperatives also face several weaknesses arising principally from collective action problems. First, cooperatives typically face entry problems and are seldom voluntarily formed. As Lewin (1981, p. 23) observed, “usually an organization such as a trade union, local authority, commercial enterprise, or even the government is the moving force behind the co-operative.” Second, collective decision-m aking could be inherently costly (Hansmann, 1991, p. 34-36) and housing cooperatives frequently have low levels of participation. Also, as Olson (1965) observed, the tendency of members to free ride (i.e. to obtain 35 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. collective benefits without participating) increases as the group size increases (i.e. as the cooperative becomes larger). Hechter (1990) argues that since costs of monitoring and sanctioning can be high, the degree of cooperative’s success depends on mecha nisms developed by the group to minimize such costs. Motivation for participation could be induced in various ways: imposing selective incentives (Olson, 1965); pro viding good leadership (Shah, 1996); increasing the perception of psychological own ership (Vandewalle, Dyne, and Kostova, 1995) or social control by members (Cooper and Rodman, 1994). Third, heterogeneous preferences of members could adversely affect the alignment of the individual preferences with the collective decisions. As Hansmann (1996) illustrates, “a decision by a cooperative apartment building to accelerate re payment of the principal on the building’s mortgage may affect members differently depending on their personal liquidity and tax status even if they occupy identical apartments and have identical leases” (p. 39). Devereux and Fishe (1993) also argue that homogeneity is significant for group lending programs to succeed since it lessens potential default by members. The internal strengths and weaknesses of housing cooperatives give them comparative advantages and disadvantages when compared with public and private sectors. Public and private agencies are typically based on the principles of vocation and income maximization respectively (Brett, 1993). Public agencies have a political mandate, and are resistant to “exit” in Albert Hirschman’s (1970) terms. Private agen cies rely on profits. As voluntary member organizations, housing cooperatives are not necessarily motivated by capital—members could form cooperatives to achieve com mon objectives that each member could not have otherwise achieved individually. 3 6 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. However, similar to other voluntary organizations, cooperatives are also susceptible to “exit” under unfavorable institutional conditions. Thus, while cooperatives offer collective advantages, they also have inherent collective action problems. Members have incentive to form cooperatives to the ex tent they obtain collective advantages while overcoming the disadvantages. House holds could defect from cooperatives if disadvantages outweigh benefits of joining cooperatives (for example, if participation costs are too high). Bonus (1986) resolves the tension between collective benefits and concurrent problems of the cooperative by positing it as an organization that is characterized by a precarious equilibrium between a member’s benefits of being in the collective (centripetal forces) and benefits of in dependent operation (centrifugal forces). As the forces shift, cooperative’s equilib rium changes in nature—members would continue to participate in the cooperative as long as centripetal forces exceed centrifugal forces. Members could exit the coopera tive if they could obtain similar services at lower costs on their own or through other means. Asset specificity gains significance in this context: if members’ investments in the cooperative were highly asset specific, they would lose their investments if they leave the cooperative. In such conditions, ‘ ‘ ‘ ‘ exit is therefore not a realistic option for members of a cooperative, and voice is thus indispensable within the cooperative” (Bonus, 1986, p.334). 2.2.3. Nature of Housing Objectives The housing process is characterized by a peculiar set of characteristics, and is important to consider for why housing cooperatives might evolve. Briefly, the hous ing process can be conceived in terms of production and consumption of housing. The production process is marked by the delivery of housing and is usually of finite term. It consists of several facets—land acquisition, construction, service provision, and fi- 37 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. nancing. For Barlow and Duncan (1988), the production process raises “questions of housing as an industrial process, as a process of value production, questions of how profits are made, or what agents are involved in housing production and so on” (p. 227). They identify three types of agents in providing housing: individualized (where individual households construct housing by themselves or through a builder); through intermediate agents (the state, housing associations, cooperatives, private developers) which use their own labor or contract through private enterprise; and through building developers that develop and build housing. In this context, cooperatives would emerge to the extent the provision of housing through cooperatives is more efficient (or more desired by households) than provision through alternative modes. The consumption process typically begins after household occupation of hous ing units. Consumption relations are more enduring than production relations since consumption is extended over the useful life of housing. Housing is a durable good and shelter services are consumed over a long period of time. The consumption proc ess also includes several facets: ownership, housing management, maintenance, re pairs, and improvement. Households are the chief agents of consumption as they are the occupants, although the consumption process may involve participation of other agents (e.g. service utilities firms, neighbors, owners, and so on). In this, households may form cooperatives for taking on the role of various aspects of consumption. Co operatives may emerge for collective ownership, housing management, and mainte nance. Cooperatives that are solely oriented towards production are likely to last for the duration of the production process only. Asset specificity in production-oriented cooperatives is limited to the production process. For example, the need for dedicated assets exists during the production process; such dedicated assets have little role after 38 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the production. Unlike public or private housing agency whose clientele base is unre stricted, a cooperative principally serves its members. Hence, unless the membership base expands or changes, the cooperative is likely to cease to exist once the housing production objectives for all members are achieved. Since the consumption process is more enduring than the production process, cooperatives involved in consumption process are likely to be more stable. Asset specificity in consumption-oriented coop eratives exist over an extended period. Consideration of the production and consumption process of housing is useful in delineating the role of cooperatives. Production oriented cooperatives are site de velopment cooperatives, building construction cooperatives, finance cooperatives, and so on. Consumption oriented cooperatives are associations for collective management or maintenance (e.g. tenant management cooperatives). Indeed, some cooperatives could span both production and consumption processes. For example, hybrid coopera tives could be involved in housing construction, finance, and management as well. In the Indian context, three types of housing cooperatives can be identified based on their housing objectives. First, they are Building Cooperatives that are usu ally involved in land development and housing construction. Second, they are Tenure Cooperatives (i.e. based on ownership), which collectively own and manage housing (i.e. cooperative housing). Third, they are Finance Cooperatives, which provide hous ing finance to their members for new housing construction or repairs. The classifica tion is, however, neither mutually exclusive nor mutually exhaustive. For example, often Tenure and Finance Cooperatives may also assume the functions of Building Cooperatives. Self-help housing cooperatives could also span the functions of all three types of cooperatives. Yet, the three types cover the majority of functions per 3 9 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. formed by primary housing cooperatives in India. I examine the characteristics of these three types of cooperatives below. Building Cooperatives Building Cooperatives are predominantly production-oriented cooperatives, where members pool their capital resources for various objectives like land develop ment; planning, design, and construction of housing; and allocation to members. They could specialize in one or more of these tasks. For example, Cooperative House Sites Societies specialize in site development, infrastructure provision, and subdivision of sites into plots; these plots are then sold to members after they are developed. Simi larly, Cooperative House Construction Societies undertake housing construction on behalf of their members; after construction, housing units are allocated to members. Other Cooperative House Building Societies span both land development and house construction. In Building Cooperatives, members’ investments are bound to the cooperative (i.e. they have a high degree of dedicated assets) until they obtain a house or plot. Thus, centripetal forces of collective benefits are strong during the production process. After members obtain house or plot, their transactions with the cooperative are over, and there are no incentives binding members to the cooperative. After production, members have few benefits of being in the cooperative and they tend to exit the coop erative. Hence Building Cooperatives are often transitory in nature, and expire after their purpose is served. Finance Cooperatives Finance Cooperatives—also known as House Mortgage Societies in some In dian states—principally lend money to their members for house construction and re- 40 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. pairs. Although their operations could extend over the consumption process, their principal functions lie in the production process. Members usually join Finance Co operatives by buying minimum number of shares in the cooperative. While the share capital forms the core, cooperatives also take deposits from members and loans from outside sources to enhance their capital base for on-lending. Credible commitment to repay is ensured through collateral or peer monitoring. Unlike Building Cooperatives, Finance Cooperatives are less transitory in na ture—they survive as long as their capital base is stable. However, from a member’s perspective, the advantage of being a member exists so long as s/he expects to borrow. After loan repayments, there is little incentive binding a member to the cooperative. Hence, the member could exit after making the repayments. Thus, although these co operatives are not entirely transitory, they tend to have higher membership turnover. Enduring cooperatives of this type also function similar to banks—they bind members through investments in savings and deposits schemes that yield a competitive return. Such cooperatives give preference in lending to members who are account holders with good credit standing. Informal micro-credits for housing could also be classified as Finance Coop eratives, although they are qualitatively distinctive from formal Finance Cooperatives in their functioning. Informal micro-credits typically use peer monitoring and mutual sanctions to ensure repayments. Hence, informal micro-credits are typically small in size. Formal Finance Cooperatives could be comparatively larger with formal admin istrative structures. In both types of Finance cooperatives, however, members have an incentive to remain in the cooperatives as long as they have dedicated assets or expect to borrow in the future. 41 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Tenure Cooperatives In Tenure Cooperatives, ownership of housing is based on collective tenure through the cooperative. Typically, cooperative housing takes two basic forms. First, both land and housing could be collectively owned, while households have occupancy rights to their housing. Second, land and common properties could be collectively owned, while housing is owned on an individual basis. The cooperative possesses le gal title to the entire housing occupied by members; each member owns shares in the cooperative and has occupancy rights to the housing unit. The cooperative is respon sible for housing management (maintenance of common areas, selection of new ten ants, repairs, etc.). Cooperatively owned housing is distinctive from condominium ownership (also known as “apartments” in India). In condominiums, each household owns the housing unit occupied by it, and has undivided rights to the common property. While decisions in a cooperative are based on democratic voting of members, decisions in condominiums are usually based on various other formulae (e.g. shares invested by members, area of housing unit, and so on). Further, while the cooperative is responsi ble for the mortgage and debt repayments, in a condominium, individual households are liable for mortgage and debt repayments. There are several varieties of Tenure Cooperatives in India. For example, in New Delhi, they are Cooperative Group Housing Societies that undertake planning, construction, and management of housing; these housing developments are predomi nantly multi-family dwelling units. In Bombay, they are Tenant Ownership Housing Societies (where cooperative owns the land and members own the housing unit) or Tenant Co-Partnership Housing Societies (where cooperative owns the land as well as housing units, and members have occupancy rights to their dwelling units). 42 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In Tenure Cooperatives, both members and non-members can enjoy the coop erative’s services. For, a non-member can rent a housing unit from a member of the cooperative. The presence of the non-members does not, by itself, modify the formal characteristics of a cooperative since they do not have ownership rights and the right to vote in decision making process. Yet, they may affect the cooperative’s functioning since renters may have less affinity to their homes than owners (e.g. in terms of main tenance). Also, it may affect collective participation since a member who lets the housing unit may not participate to the same extent as she would if she herself were occupying it. Size of these cooperatives is finite since the number of housing units is fixed. Among the three types of cooperative societies, Tenure Cooperatives are the most stable. They continue to exist due to perpetuity of co-ownership. Investments of members carry a high degree of asset specificity (particularly in terms of site specific ity and dedicated assets). Thus, centripetal force of collective ownership is strong and members tend to exercise voice. When a member exercises the exit option (i.e. trans fers his/her membership), a new member is inducted with similar co-ownership rights, thus continuing the cooperative’s life. 2.3. Exogenous Level Analysis of Housing Cooperatives The exogenous level analysis is useful for identifying the external institutional constraints and opportunities in the development of cooperatives. As I identified be fore, two related aspects require examination at this level. The first aspect is the insti tutional framework, which forms the incentive structure in which housing cooperatives are embedded. In this, the political economic context gains significance for the extent to which the institutional framework is favorable to the growth of housing coopera tives. The second aspect is the characteristics of local housing market, which deter- 43 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. mines the specific institutional conditions under which housing cooperatives emerge. In this, the historical context of the local housing market gains significance for the emergence of housing cooperatives. 2.3.1. The Institutional Framework North (1990, 5) argues that “both what organizations come into existence and how they evolve are fundamentally influenced by the institutional framework.” Insti tutions form the incentive framework in which organizations like housing cooperatives evolve. The institutional framework affects transaction costs of formation and func tioning of organizations like housing cooperatives. The costs of formation for housing cooperatives include entry and establishment costs; the costs incurred in the function ing include administrative and monitoring costs, costs of resolving disputes (between members or cooperatives or both), and so on. Higher transaction costs—that is, higher costs of formation and functioning—could lower the range of activities of an organiza tion in a given institutional context. Similarly, lower transaction costs could facilitate an increase in an organization’s activities. However, as North (1990) argues, “institu tions everywhere are a mixed bag composed of those that lower [transaction] costs and those that raise them” (p. 63). Prohibitions, unclear property rights, and information asymmetry raise transaction costs, and preferential treatment (e.g. subsidies, tax con cessions), clearly defined procedures/ property rights lower them. Thus cooperatives would emerge to the extent their transaction costs are lowered in comparison to other types of organizations (e.g. public, private, and other Third sector organizations). Housing policies, particularly as they relate to housing cooperatives, form a significant component of the formal institutional framework in the development of co operatives. Among others, such policies relate to direct provision of housing by the state, access to housing finance, and access to land. These policies are set in the 44 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. broader political economic context, wherein the state ideology has a central role. Al though cooperatives exist in both socialist and capitalist political economic systems, they offer different incentives for cooperatives to emerge. In the Indian context, in centives offered by both systems are relevant, since the political economic context shifted from democratic socialism to economic liberalization over the past five dec ades. Hence, both systems need to be taken into account for tracing the evolution of housing cooperatives. Socialist political economic systems favor cooperatives due to their ideological alignment with cooperative’s collective principle and self-management. Gamble and Kelly (1996) argue that “the real successes of the socialist tradition lie not in the or ganization of centrally planned economies but in the achievement of forms of collec tive action, such as trade unions, cooperatives, community groups and political movements” (p. 82). Horvat (1982) also argues that the cooperative movement could contribute greatly to “the delegitimation of private ownership with its social hierarchy and authoritarianism” (p. 461). The socialist state promotes cooperatives through preferential treatment. For example, housing policies could include cooperatives as alternative means to public housing. Housing cooperatives could be provided preferential access to finance and land. Further, cooperatives could be given concessions in terms of subsidies, license, tax rebates, and so on. The overall state support thus reduces the transaction costs of the cooperatives as compared to private enterprises. Consequently, cooperatives may proliferate in comparison to private entities. However, in the process, the state could also co-opt cooperatives and control their functioning. Shah (1996) argues that “since co-operatives are people intensive and frequently promoted by the state they often get co-opted by the state and the national political elite into the larger policy framework 45 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. which invariably draws them into political and bureaucratic power play” (p. 39). Ex cessive state controls could undermine the autonomy of cooperatives, so that the func tioning of cooperatives could be constrained by the state. The institutional context of the capitalist market system does not explicitly fa vor cooperatives. The competitive market determines the type of firms that will sur vive (Alchian, 1950). Cooperatives will survive in the capitalist context to the extent they have comparative advantage over other organizations. In this context, coopera tives may not necessarily be preferred vehicles of public housing. Access to finance and land could also depend on the market mechanisms. Since there is little state sup port, autonomy of cooperatives is greater. Yet, cooperatives may face structural diffi culties in obtaining finance and land, which are two key components of housing. Co operatives may not be able to raise housing finance if lending mechanisms (e.g. mort gage instruments) are not adequately developed to recognize cooperative tenure. The limitations on ability to raise finance could also impair cooperative’s access to land. Indeed, Marxist critiques like Elster, argue that the capitalist system structur ally discriminates against collective organizations like cooperatives. While asking the question, “If cooperative ownership is so desirable, why there are so few coopera tives?” Elster (1989, p. 96) suggests that the internal collective action problems of co operatives are not insurmountable. Instead, he argues that isolated cooperatives could face four types of difficulties in a capitalist context that they would otherwise not face in a “fully cooperative context.” These difficulties include: endogenous preference formation, adverse selection, discrimination, and externalities. According to him, en dogenous preference formation is an obstacle if the presence of cooperatives is a pre condition for the desire to have them. Adverse selection could occur since going against the dominant social conventions requires qualities that do not always aid the 46 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. endeavor. Discrimination could occur as a result of unfair competition between coop eratives and other types of firms. Lastly, isolated cooperatives may not be able to in ternalize the positive externalities generated by them (e.g. entrepreneurial and political externalities) in a capitalist system. At the same time, cooperatives may not be able to avoid the negative externalities generated by other capitalist firms (e.g. opportunistic and collective bargaining externalities). Bengtsson (1992, p. 87-88) interprets Elster’s arguments as “fundamental” and “transitional” problems that affect housing coopera tives in a capitalist system: Fundamental problems could be that tenants do not want to participate in the management of their estate, or that they lack the competence to do it efficiently, or that cooperative housing does not provided tenants and management with the right incentives. Transitional problems may be that only eccentrics or idealists are prepared to join cooperatives in a housing sector dominated by the capitalists and the state. Furthermore, the market may not be ready for cooperative solutions, and the legal and institutional framework may be unfavorable. Although it is arguable if Elster’s “fully cooperative context” can indeed be achieved, he raises some fundamental constraints with respect to emergence of collec tive organizations in a capitalist context. Some of these had been broadly overcome in the India due to the initial national ideological context of democratic socialism that emerged after Independence. The state explicitly promoted cooperatives through pref erential treatment, and discrimination was in favor of cooperatives. 2.3.2. Characteristics of Local Housing Market Although housing cooperatives proliferated nationwide in India, there are re gional differences in the growth of housing cooperatives. To understand the differen tial evolution of housing cooperatives at the local level, the local housing market char acteristics need to taken into account to identify the specific institutional conditions 47 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. under which housing cooperatives emerged. On one hand, the evolution of such con ditions is contingent on the cultural and historical context of the local housing market. On the other hand, the demand for collective organizations like cooperatives also de pends on the local housing market. The specific characteristics of housing market need to be considered for emer gence of housing cooperatives. Housing is unlike other consumer goods. It has pecu liar physical and legal characteristics. Housing is principally composed of two physi cal components—land and superstructure (i.e. shelter). Land is characterized by its permanence (i.e. it is non-destructible), immovability (i.e. it is fixed in geographical space), contiguity (i.e. continuous over space), and non-fungibility (i.e. no two parcels are the same) (Doebele, 1983). Land is thus a durable commodity. As classical theo ries of land rent show, its value depends fundamentally on location. Durability of the housing superstructure depends on the quality of construction,5 but it is also prone to obsolescence due to aging, changes in design style and so on. Once the superstructure is constructed, it could be costly to replace (Keogh and D’Arcy, 1998).6 The super structure may take many forms—informal squatter settlements (which are typically temporary structures), single-family homes (usually on detached individual plots), multi-family housing (flats or apartments), shared housing, and so on. On the legal side, the local institutional context matters for the delineation of property rights. Tenure defines the ownership and occupancy rights to the housing 5 In India, structures are classified a spucca and kutcha. Pucca structures last between 30 to 50 years, while kutcha structures in slums last for a few months and require periodic repairs. The Census o f India defines a pucca struc ture as one which has walls and roofs made o f “materials such as cement, concrete, oven burnt bricks, stone and stone blocks, jack board tiles, timber, galvanised or corrugated iron sheets, etc.” A kutcha structure is one which has both walls and roofs made o f materials such as “unbumt bricks, bamboo, mud, grass, leaves, reeds, and/or thatch etc.” (NBO, 1996, p. xii). 6 In classical econom ics, however, replacement o f housing incurs the least cost. Yet, replacement o f housing (even temporary structures in slums) has attached political and social costs. 48 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. unit. Ownership determines the incentives for households to invest in home improve ments. Tenure is closely associated with the physical characteristics of housing. Dif ferent sets of tenure arrangements could arise based on ownership of land and housing superstructure. Doebele (1983) identifies seven basic proprietary categories of land tenure: (i) non-formal defacto tenure (“squatting”); (ii) private freehold; (iii) private leasehold; (iv) public freehold; (v) public leasehold; (vi) communal tribal ownership; and (vii) communal neighborhood ownership. The specific form of housing tenure depends on the cultural and historical context of the local housing market (Dale, 1997). There are two primary types of tenure with respect to the housing superstruc ture: (i) ownership; and (ii) rental. Home ownership could include individual, shared, or informal ownership. Ownership is usually perpetual, with transferable and heredi tary rights. In rental accommodation, landlords own the property, and are distinct from renters, who have occupancy rights. Rental is temporally limited. While capital gains accrue for an owner, they do not accrue for a renter. The tenure form of housing superstructure is intimately linked with land tenure, so that a range of housing tenures could emerge through a combination of the two. Due to the physical characteristics of land and superstructure, housing invest ments obtain a high degree of site specificity and dedicated assets in terms of asset specificity. Investments in housing are highly site specific since location matters. Proximity with other superstructures also matters for sharing of common spaces, amenities, and infrastructure (typically utilities). Housing requires a high degree of dedicated assets due to its durability. As Buckley (1996) argues, “because a house is long-lived and its rental cost requires a significant share of annual income, the present value of housing is a multiple of annual income” (p. 15). Housing investments are 49 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. thus lumpy and form a significant percentage of a household’s income. Further, in developing country contexts like India where housing mobility is typically low, hous ing investments are usually long term spanning several generations. Housing is also rife with externalities. Improvement or degeneration of a par cel of land affects other parcels of land. Provision of public services could benefit several parcels of land together; pollution of any form (noise, air, water) could nega tively affect many parcels of land in the vicinity. There are also neighborhood exter nalities. As Keogh and D’Arcy (1999b, p.918) illustrate: If collectively they [buildings] generate an image of dereliction and de cline, then this may contribute to the economic decay of whole areas and a loss of entrepreneurial capacity. Conversely, a high-quality built environment can be expected to support a positive city image, encour aging economic growth. Lastly, housing morphology gains significance for the degree of excludability. Although housing is a private good in terms of consumption, it has certain dimensions of collective good based on the morphology. In single family detached housing units, property boundaries (e.g. lots) are clearly defined. Consequently, such units have high degree of excludability in terms of facilities and services, which are clearly under the control of the individual unit owners. In multi-family dwelling units, there is no clear demarcation of property boundaries. Common facilities and services such as stairs, corridors, yards, elevators, water and sewage utilities need to be shared among the owners. An owner cannot freely undertake repairs on his/her part of the building, since such repairs could structurally affect other units. Thus, multi-family units have low degree of excludability. Housing cooperatives could emerge to the extent they are more optimal than other organizational forms in addressing the above issues of asset specificity, external ities, and excludability. Thus cooperatives could emerge to address the adjacency is- 50 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. sues due to site specificity of housing and to achieve economy of scale in dedicated assets. In terms of production, cooperatives for land development typically provide common site services. Self-help and other building cooperatives typically mobilize materials and labor required for construction. In terms of consumption, housing coop eratives could emerge to internalize externalities, which typically happens by exercis ing collective control over housing. Cooperative housing tenure could take two basic forms. Land and common properties could be collectively owned, while housing is owned on an individual basis. Alternatively, both land and housing could be collec tively owned, while households have perpetual occupancy rights. Housing morphology could also be important for the emergence of coopera tively owned housing. In a housing market that is predominantly made up of single family units, there is little need for collective organization of owners. Each owner could carry out his/ her activities independently (i.e. centrifugal forces dominate over centripetal forces). Consequently, the demand for cooperative organization is likely to be low. In a market consisting mainly of multi-family dwelling market, alternative organizational structures are required since no unit-owner has the freedom to operate independently. Consequently, the demand for cooperative ownership is likely to be higher. In his model of tenure choice, Williams (1993, p. 94) also argues: With single-family houses the extra maintenance [by owners] is not sufficient to compensate for the reduced care [by tenants]. As a result, houses are optimally owned by their occupants. Also, landlords have economies of scale in maintaining large apartment complexes. ... In this case, there is a critical number of residences per property, such that all units in multifamily dwellings of equal or smaller size are optimally owned as condominiums or cooperatives, while all units in larger prop erties are optimally rented as apartments, (p. 94) 51 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2.4. Significance of Institutional Structure While the endogenous level analysis highlights the internal strengths and weaknesses, the exogenous level analysis highlights the structural constraints and op portunities in the development of housing cooperatives. The comparative advantage and disadvantage of housing cooperatives over other organizations is a function of both the endogenous characteristics and the exogenous institutional framework. Hence, both levels of analysis need to be taken into account for the development of housing cooperatives. In combining the two levels of analysis, I propose that the cooperative institu tional structure is significant for the development of housing cooperatives. Institu tional structure refers to the manifest arrangement o f relationships between organiza tions fostered by the institutional framework. The institutional structure mediates both the endogenous and the exogenous levels. At the endogenous level, a supportive structure enhances the internal organizational strengths, while overcoming the concur rent weaknesses of cooperatives. At the exogenous level, the structure reduces the transaction costs of formation and functioning of housing cooperatives. A supportive institutional structure provides a reinforcing mechanism for or ganizations to persist and to enhance their activities. In the absence of the structure, their activities are likely to decline over time. The cooperative institutional structure could consist of the broad structure of primary, secondary, and tertiary cooperatives. While this structure may not be strictly hierarchical in terms of vertical integration, it is also not strictly a horizontal network in terms having equal footing. Primary coop eratives could operate independently, but maintain associational relationship with sec ondary cooperatives. Such institutional structure is intra-sectoral, i.e. relationships between organizations are maintained within the cooperative framework. Inter- 52 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. sectoral institutional structures may include private and public agencies that may pro vide specific types of external support to primary cooperatives. The institutional structure could support development of housing cooperatives in three specific ways, which act at both internal organizational level and the external institutional level. First, the institutional structure could provide administrative, legal, and procedural support. Providing such support could enhance the internal strengths while overcoming the internal weaknesses of cooperatives. Administrative support could facilitate primary cooperatives to overcome some of the collective action prob lems inherent in cooperatives, especially in terms of monitoring and educating the members about management practices. Legal support could reduce the conflicts that arise between members or between members and cooperatives. Lastly, procedural support could be in the form of establishing routine procedures for formation and functioning of cooperatives. Such support is crucial for reducing the transaction costs of formation and functioning of housing cooperatives. Second, the institutional structure could enhance access to finance, which is a key component of housing. As cooperatives are member organizations, they usually raise their finances through their members. However, given the inherent characteris tics of housing market in terms of dedicated assets and lumpy investments, finance raised through members needs to be leveraged with external funding for adequate in vestment in housing. Cooperatives could also face obstacles in obtaining such finance in the open market, i.e. their transaction costs for obtaining finance could be high. In the absence of external financial support, cooperatives could grow only to the extent they are able to mobilize finance internally for productive housing investments. Third, the institutional structure could enhance access to land, which is another key component of housing. As housing market is characterized by high degree of site 53 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. specificity, cooperatives typically require large parcels of land for contiguous housing development. Cooperatives could face high transaction costs in access to such large parcels. If such land were unavailable, development of cooperatives may be limited. Thus, both housing finance and land are two critical factors of housing where exoge nous institutional support is required to enhance the activities of housing cooperatives. 2.5. Conclusion As the foregoing analysis reveals, housing cooperatives need to be considered at both the endogenous organizational level and the exogenous institutional level. While analysis at the endogenous level highlights the cooperative’s comparative or ganizational strengths and weaknesses, analysis at the exogenous level reveals the broader institutional constraints and opportunities in the development of housing co operatives. In combining the two levels of analysis, I have argued that the cooperative institutional structure acts as a reinforcing mechanism in the development of housing cooperatives. At the first level of analysis, there are three distinctive organizational aspects of cooperatives. The first aspect is the governance structure: cooperatives are charac terized by collective ownership and management. While such a governance structure holds strengths for achieving collective action, it also engenders collective action problems. The second aspect is the membership characteristics. As cooperatives are not oriented towards any particular group, members may join or form cooperatives if a member’s benefits of being in the collective exceed the attendant costs, or if such benefits exceed the benefits from acting independently. The third aspect is the nature of housing objectives—different types of housing cooperatives could emerge based on the asset specificity implied by such objectives. Three types of housing cooperatives could be identified in India: Building Cooperatives, Finance Cooperatives, and Tenure 54 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Cooperatives. Consideration of production and consumption process of cooperatives reveals that the three types have different characteristics in terms of asset specificity and excludability. For example, stability varies among the three types. Building Co operatives are least stable due to low asset specificity after production process is over. Tenure Cooperatives emerge in multi-family housing and are most stable due to high degree of asset specificity and indivisibility; such cooperatives may not emerge in sin gle-family housing due to the low degree of asset specificity and high degree of ex cludability. Stability of Finance Cooperatives could vary depending on their financial base. Members could tend to defect after their financial commitments are over. There are two distinctive aspects at the exogenous level of analysis. First, the institutional framework forms the incentive framework in which cooperatives evolve. The framework influences the broader institutional constraints and opportunities; it determines the transaction costs of cooperatives’ formation and functioning. In the Indian context, the institutional frameworks under both socialist and capitalist states become relevant since India’s political economic context evolved from democratic socialism to that of economic liberalization over the last fifty years. Second, the char acteristics of local housing market determine the specific institutional conditions under which housing cooperatives emerge. The local housing market affects the demand for collective organizational forms like cooperatives. In combining the two levels of analysis, I propose that the cooperative institu tional structure is key for fostering housing cooperatives. Institutional structure refers to the manifest relationships of housing cooperatives with each other and with other organizations. It is a function of both the external institutional framework and the in ternal organizational features of cooperatives. The institutional structure could lower transaction costs of formation and functioning of housing cooperatives by providing 55 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. support in three ways: (i) administrative, legal, and technical support; (ii) access to finance; and (iii) access to land. While the first addresses the internal weaknesses of cooperatives’ governance structure, the other two address the external structural con straints in the development of cooperatives. Implementing such institutional structure is crucial for the development of housing cooperatives. 56 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3. Evolution of Housing Cooperatives in India Although housing cooperatives were introduced in India in the early decades of the twentieth century, their activities were limited to Bombay and Madras provinces until the country’s independence in 1947. Since independence, the activities of hous ing cooperatives grew phenomenally nationwide. In this chapter, I make an institu tional analysis of this extraordinary evolution of housing cooperatives’ activities in the country. I examine the role of cooperative institutional structure in the phenomenal growth of housing cooperatives. The chapter is structured in four sections. The first section deals with the evo lution of housing cooperatives during pre-Independence period. The second section deals with the phenomenal growth of housing cooperatives in post-Independence pe riod. The third section examines the institutional framework in which housing coop eratives developed. The fourth section deals with the role of institutional support structures in the extraordinary growth of housing cooperatives nationwide. The last section concludes with the national significance of supportive institutional structures for housing cooperatives. 3.1. Housing Cooperatives in the Pre-Independence Period (1904-19471 Although indigenous mutual loan associations were prevalent in some parts of India,1 cooperatives were not indigenous to India. They were, “to begin with, the creation of the state” (Catanach, 1970, p. 3). Some officials of British government in India had knowledge of cooperatives in Europe, and sought to introduce similar or ganizations in India. W olff, an influential leader o f cooperative m ovem ent w orldw ide, 1 For example, nidhis (which were similar to provident funds and friendly societies in Europe) were common in Madras province; chit funds and rotating savings and credit associations (ROSCAs) also existed in a few provinces (Hough, 1932, p. 51). 57 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. observed as early as 1920 that there were no pioneering cooperative leaders like Schulze or Luzzati or Raiffeisen in India who provided indigenous leadership for the movement. He wrote: . . . the Co-operation eventually introduced was distinctly a government measure, necessarily the product of Government action, issuing from headquarters, with the Government stamp plainly impressed upon it— so plainly, in fact, that often enough, quite naturally, have its character and its objects come to be altogether misunderstood by those for whose benefit it was introduced,. . . (Wolff, 1920, p. 11) The British introduced cooperatives in India in the turn of twentieth century to alleviate peasant indebtedness to moneylenders and to improve rural credit. In 1904, Co-operative Societies Act was passed as “one of a series of measures which, while by no means intended to extinguish the money-lender, did aim to put some fairly definite limits on his activities” (Catanach, 1970). They were expected to provide the peasant with a “suitable institution in which to lay by his savings and teaches him the valuable lesson of self-help through the sense of responsibility he feels in being its member” (Reed and Sheppard, 1928; p. 412). One of the principal pillars of the cooperative institutional structure—the Reg istrar’s Office—was established under the 1904 Act. The Registrar was a special gov ernment officer who controlled the development of cooperatives in each province. The Registrar’s Office gained immense significance over the years and became pivotal for the development of cooperatives in India. As the main objective of cooperatives under the 1904 Act was to provide rural credit, non-credit cooperatives like housing cooperatives could not be formally regis tered. Often, non-credit cooperatives included a credit component in their operations to get registered. Non-credit cooperatives were legally recognized following the amendment of Co-operative Societies Act in 1912, which also laid the legal basis for 58 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. housing cooperatives to function. The amendment redefined cooperative as “a society which has as its object the promotion of the economic interests of its members in ac cordance with the co-operative principles, or a society established with the object of facilitating the operations of such a society” (Hough, 1932, p. 265). It formalized the use of the word co-operative as a legal nomenclature and prohibited its use as “part of the title of any business concern except as a registered society” (Reed and Sheppard, 1928, p. 413). In 1919, the British undertook extensive administrative reforms with the Gov ernment of India Reforms Act (also called the Montague-Chelmsford Act). Under the “dyarchy” principle of the Act, some of the central government responsibilities were transferred to provincial governments. The responsibility of administering coopera tives was also transferred provincial governments in this process. Provinces could now adapt cooperative laws to their regional needs. The Act laid early seeds of diver gence in the development of cooperatives between provinces in India. In the following years, several provinces introduced cooperative legislations, many of which were modeled after the 1912 Cooperative Societies Act. Provincial legislations refined the scope of non-credit cooperatives like housing by defining their objectives more clearly. Laws governing housing cooperatives were the same as those applicable to cooperatives in general. Although housing was recognized as a separate sector, there was no separate legislation governing housing cooperatives per se. In Bombay province, Bombay Co-operative Societies Act was passed in 1925, which classified cooperatives into five categories: resource, producer, consumer, housing, and general. In Madras province, Madras Cooperative Societies Act was passed in 1932, which went beyond the Bombay legislation to recognize five sub-categories of housing cooperatives: Ordinary Building Societies; House Building Societies; House 59 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Construction Societies; Cooperative Townships; and Cooperative House Mortgage Banks. Although Cooperative Acts were enacted in other provinces, the activities of housing cooperatives were principally limited to Bombay and Madras provinces dur ing the pre-Independence period. In Bombay, cooperatives were principally Tenure Cooperatives. They were in the form of Cooperative Housing Societies, which were of two types: (i) tenant co-partnership and (ii) tenant ownership. Under the first type, the cooperative owned both land and buildings; under the second, cooperative owned the land, and members leased housing units from the cooperative. Cooperatives in Madras were principally Finance Cooperatives, predominantly in the form of Ordinary Building Societies, which were “confined mainly to the provision of capital for build ing schemes and only occasionally extend[ed] to the joint purchase of land or of build ing materials” (Reed and Sheppard, 1928, p. 420). Housing cooperatives were much less developed in other parts of India until India’s Independence. There were nine co operative housing societies in United Provinces in 1929-30 (Hough, 1932, p. 191). Bihar and Orissa had one cooperative society each; in Hyderabad, cooperatives were still in the developing stages. Thus, the history of cooperatives in the pre- Independence period is principally that of housing cooperatives in Bombay and Ma dras provinces. The activities of housing cooperatives in Bombay and Madras provinces accel erated since 1922, when the provincial governments started to provide loans. At the end of 1930, there were 67 housing cooperatives in Bombay province, of which 23 w ere in B om bay city and its suburbs. In Madras, b y 1926-27, Ordinary B uilding S o cieties grew to 109 (an almost five-fold increase since 1922-23) with 3,383 members; they had constructed 681 houses (Government of Madras, 1928). However, the activi- 60 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ties of housing cooperatives were affected since then in both provinces due to the 1929 depression and World War II. Also, government loans were discontinued in Bombay since 1933. Although construction activities of housing cooperatives reduced in Bombay, the number of housing cooperatives grew from 79 to 136 during this period. These cooperatives were mainly Tenure Cooperatives. For example, in 1944, there were 114 cooperative societies in Bombay, of which “101 [were] working on the ten- ant-ownership system and 10 on the co-partnership system, the remaining 3 of the mixed type” (Government of India, 1951, p. 120). In Madras, several cooperatives defaulted on their repayment of loans to the government; consequently, their number decreased from 135 to 113 between 1930-31 and 1945-46. These cooperatives were principally Finance Cooperatives, in the form of Ordinary Building Societies. Table 3.1 summarizes the growth in number of cooperatives, their membership, and the number of houses constructed in Bombay and Madras provinces between 1930-31 and 1945-46. Table 3.1. Growth of Housing Cooperatives in Bombay and Madras Provinces, 1927-28 to 1946-47. Year Bombay Madras Co operative Housing Societies Members Tenements constructed Ordinary Building Societies Members Houses completed 1927-28 52 6,260 N.A 109a 3,383a 681a 1930-31 79 6,420 N.A 135 4,269 1,402 1935-36 80 5,617 N.A 118 4,040 1,809 1940-41 101 6,988 N.A 117 4,492 2,346 1945-46 136 8,597 3,250 113 4,027 879b Sources: Unless otherwise mentioned, Reserve Bank of India (1950), p. 43 and p.51. Notes: (i) Data (a) obtained from Government of Madras (1928) and relates to 1926- 27; (ii) Data (b) obtained from Government of Madras, Administration Report o f the Co-operative Department for 1946-47, p. 55, and relates to 1946-47. 61 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3.2. Housing Cooperatives in the Post-Independence Period (1947-present) In the post-Independence period, the activities of cooperatives grew phenome nally nationwide. In the first two years following Independence, the number of coop- erative housing societies in Bombay state increased more than three-fold (there were 480 cooperatives in 1948-49). In Madras state, Ordinary Building Societies grew by 50 percent in the same period (there were 174 of them in 1948-49). Although there were only 553 housing cooperatives in other states by 1949-50 (Mathur, 1977, p. 464), the activities of housing cooperatives grew in most states since then. The number of primary housing cooperatives in the country increased from 1,482 in 1950-51 to over 90,000 in 1999-2000; during the same period, membership of primary housing coop eratives grew from 0.09 million to 6.5 million (Table 3.2). Table 3.2. Growth of Housing Cooperatives in India, 1950-2000 State level (Apex) Cooperatives Primary Housing Cooperatives Membership (in million) 1950-51a N.A. 1,482 0.09 1954-55 2,118 0.15 1959-60 6 5,558 0.32 1964-65 12 10,975 0.70 1969-70 17 16,308 1.11 1974-75 17 25,633 1.64 1979-80 17b 34,036 1.85 1984-85 2 ib 49,317 3.27 1990-91 25b 64,384 3.S7 1994-95 25b 68,121 4.13 1999-00° 26 90,000 6.50 Sources'. Unless otherwise mentioned, Statistical Statements Relating to Co-operative Movement in India (for respective years). Notes: (i) Data (a) obtained from Mathur (1971, pp. 498-499). (ii) Data (b) calculated from Khurana (1995, p. 17) (iii) D ata (c) obtained from G overnm ent o f India (2002, p. 622) (iv) The interpretation of this table should be done with some caution. The statistical statements since 1979-80 do not include all states (e.g. Tamil 2 In the post-Independence period, provinces have been referred to as states. 62 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Nadu); presumably, the figures are based on those states for which data were available at the time. Hence, the figures are undercount. I have overcome this difficulty for State level (Apex) Cooperatives by calculating their numbers based on when each such cooperative was established. The growth in number and membership alone, however, does not indicate a growth in their activities. Their increase in activities is also indicated by their partici pation in the housing market i.e. their contribution to the national housing stock and to the volume of housing finance. Based on the Indian National Report submitted at the Habitat II conference in 1996 (Government of India, 1996), I estimated that coopera tives contributed about 10.8 percent of the annual housing stock between 1991 and 1995.3 The contribution of cooperatives to the housing stock also increased consid erably over the years. Housing cooperatives contributed 168,961 units during the five year period between 1960-61 and 1964-65. They contributed 1,289,650 housing units during the five year period between 1995-96 to 1999-2000, reflecting an over seven fold increase (Table 3.3). Housing cooperatives also play a crucial role in the dis bursement of formal housing finance. In 1987-88, cooperative housing finance formed 12.9% of the total volume of formal housing finance in India (Struyk and Ravicz, 1992, pp. 33). Bombay and Madras states provided a guide for the development of housing cooperatives in other states where they were yet at their infancy. Other state govern ments adapted the institutional framework and the form of cooperatives according to their local housing conditions. For example, New Delhi adapted Bombay’s Coopera tive Societies Act in 1949; Tenure Cooperatives emerged in New Delhi in the form of 3 Comparatively, Sweden, one o f the countries where activities o f housing cooperatives are most pronounced, co operative housing stock accounts for about 15 percent. In the United States, where housing cooperative activities are minimal, cooperative housing stock accounts for less than 1 percent. See Chapter 8 for comparison o f interna tional development o f housing cooperatives. 63 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Cooperative Group Housing Societies. In Kerala, housing cooperatives emerged prin cipally as Finance Cooperatives, similar to those in Madras state. Table 3.3. Five year contribution of Housing Cooperatives to the housing stock Independent Houses Tenements Total Constructed by Cooperatives Constructed by Members 1960-61 to 1964-65 40,904 64,618 63,439 168,961 1965-66 to 1969-70 25,592 75,475 90,784 191,851 1970-71 to 1974-75 57,409 115,835 122,544 295,788 1975-76 to 1979-80 43,058 68,741 93,515 205,314 1980-81 to 1984-85 140,781 137,212 93,897 371,890 1985-86 to 1989-90 81,364 231,992 151,777 465,133 1990-91 to 1994-95 124,235 232,122 163,517 519,874 1995-96 to 1999-003 85,202b 403,498b 567,766b l,289,650c Source: Unless otherwise indicated, calculated from Statistical Statements Relating to Cooperative Movement (for respective years) Notes: (i) Data (a) obtained from “Progress of Housing Programmes (During 1998- 99 to 2000-2001)” Table presented at Economic Editors Conference, (Oc tober 17-19, 2001), Ministry of Urban Development & Poverty Allevia tion, Government of India. (ii) Data (b) do not include figures for 1998-99 and 1999-00. The Total hous ing units for 1998-99 was 175,000, and for 1999-00 the Total was 58,184 units. Data (c) includes these two Totals. Housing cooperatives’ activities spread nationwide after Independence. Nearly all states have a state level housing cooperative called the Apex cooperative. Primary cooperatives’ activities have been more prominent especially in Maharashtra, Gujarat, and Tamil Nadu, which inherited the institutional conditions of their parent states of 6 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Bombay and Madras.4 Table 3.4 summarizes cooperative’s activities in eight states where the activities were most pronounced in 1994-95. In terms of number of primary housing cooperatives, the top three states were Maharashtra, Gujarat, Andhra Pradesh; but in terms of membership, Tamil Nadu, Maharashtra, and Karnataka ranked among the first three states. Construction activities are proxied by the number of housing units (independent and tenement type) constructed during the year. In this respect, Maharashtra, Kerala, and Tamil Nadu were among the top three states. Financial ac tivities are proxied by the total working capital of primary housing cooperatives. Ma harashtra, Gujarat, and Karnataka were among the top three states in this respect. Table 3.4. Primary Cooperatives’ Activities in Eight States, 1994-95 Number of Primary Cooperatives Membership Housing Units Constructed during year Total Working Capital (Million ru pees) Andhra Pradesh 3,164 493,224 7,079 401.05 Gujarat 14,563 595,759 7,716 5,552.37 Karnataka 1,588 613,118 2,874 3,053.07 Kerala 342 187,061 11,329 1,982.43 Maharashtra 42,926 1,576,425 72,903 20,182.84 Punjab 886 128,560 1,117 1,433.46 Rajasthan 1,297 247,178 30 1,036.21 Tamil Nadu l,025a 2,273,918a 10,802a l,190.97a b Sources'. Unless otherwise indicated, Statistical Statements Related to Co-operative Movement in India (for respective years). Notes: (i) Data (a) obtained from the Office of the Registrar of Cooperatives (Hous ing), Chennai, and is for 1993-94. (ii) Data (b) represents sum of paid up share capital and reserve/other funds. 4 Bombay and Madras states were divided along linguistic lines in 1960 and 1956 respectively. Maharashtra and Gujarat were formed out o f the erstwhile Bombay state, and Tamil Nadu was formed out o f erstwhile Madras state. 65 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3.3. The National Institutional Framework Why did housing cooperatives grow phenomenally, and spread nationwide fol lowing Independence? The national institutional framework provided a congenial en vironment for the growth of housing cooperatives. In the post-Independence context of democratic socialism, the state provided the initial thrust for development of hous ing cooperatives. The socialist context promoted cooperatives due to its ideological alignment with the internal organizational feature of cooperatives as collective organi zations. The national five year plans, housing policies, housing task forces and other committees recognized the significance of cooperatives. Yet, housing cooperatives played only a small role in implementing state sponsored schemes. Further, as the po litical economy evolved towards economic liberalization over the last five decades, direct state support to housing cooperatives waned. In spite of such reduction, the ac tivities of housing cooperatives continued to grow. As I argue in Section 3.4, direct state support, by itself, does not explain the growth of housing cooperatives. Rather, cooperative institutional structures that were formalized in the context of democratic socialism played a key role in the development of housing cooperatives. Once set-up, these structures supported the growth of housing cooperatives, despite waning of di rect state support. Three distinctive phases of the institutional framework can be identified in In dia, when housing policies changed with broader political economic context. Over the last five decades, India’s political economy evolved from democratic socialism to eco nomic liberalization. The role accorded to housing cooperatives also changed with the broader political econom ic conditions. A s it w ill be evident from b elow , the internal advantages of housing cooperatives as collective organizations consistently recurred in the justifying the promotion of housing cooperatives. During Phase I (1947-1970), 66 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1970), housing policies emphasized state intervention and public housing in the con text of democratic socialism. Housing policies explicitly promoted housing coopera tives, which were considered as alternatives to public (state) ownership. In Phase II (1970-1988), socialism waned gradually, just as there were pressures for economic liberalization (especially in the late eighties). Although many schemes and policies of the first phase were continued, emphasis on public housing schemes waned. Housing policies were directed towards increasing accessibility to finance and land, wherein cooperatives received preferential treatment. In Phase III (1988-present), economic liberalization policies emphasized greater role of private sector. “Enabling” housing policies, which emerged in this context, put emphasis on institutional reforms for more participation by private agencies. Although these housing policies recognize the sig nificance of housing cooperatives, direct state support to housing cooperatives reduced compared to the previous two phases. 3.3.1. Phase I (1947-1970) Following independence, cooperatives received a boost in general. Jawaharlal Nehru, the first Prime Minister of India, championed the growth of cooperatives as a part of his paradigm of democratic socialism. The socialist paradigm held much sway during Phase I. The Second Five Year Plan (1956-1961) explicitly laid out the foun dations for a “socialistic pattern of society,” which was reaffirmed in the Third Five Year Plan (1961-66). In this paradigm, cooperatives were conceived as alternative to state ownership (Government of India, 1998a, p. 391). Jain and Coelho (1996, p. 39) argue that cooperatives “began to be adopted by the new national leaders as key insti tutional instruments in the development of the economy and in transformation of the society towards a more equitable, democratic, and socialistic model.” Cooperatives’ 6 7 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. internal features of democracy and mutual self-help were viewed as being conducive to democratic society: The principle of mutual aid, which is the basis of co-operative organi sation, and the practice of thrift and self-help which sustain it, generate a sturdy feeling of self-reliance which is of basic importance in a de mocratic way of life. (Government of India, 1950, p. 163) Although in Nehru’s model, cooperatives were expected to be “a people’s movement and not something imposed from above,” (Government of India, 1998, p. 441) it was rarely achieved in reality. Yet, the national government could hardly in tervene directly in the functioning of housing cooperatives since both housing and co operatives were state government subjects. Hence, state patronage depended on the role of the state government, mediated through the Registrar’s Office. State govern ment controls were imposed either because either because cooperatives depended on the government for financial and other resources, or the state government wanted to ensure a “proper” development of cooperatives. The extent of such controls differed between states. The political context of democratic socialism provided a conducive environ ment for housing cooperatives. Housing cooperatives were given preference in land allocation and supply of building materials (especially steel and cement) that were controlled by the government following World War II (Reserve Bank of India, 1950). The idea of the cooperative as an organizational vehicle for housing gained political currency and was reinforced consistently over time. Cooperatives were considered as mechanisms of collective self-help, and as means to mobilize capital. For example, the Planning Commission noted in the First Five Year Plan (1951-56): We lay special emphasis on co-operative housing societies not only be cause they can mobilise private capital, which otherwise would remain dormant, but because they open the way for aided self-help in the con- 68 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. struction of houses which should be encouraged for reducing the cost as much as possible (Government of India, 1950, p. 599). In the socialist context, the state directly intervened in the housing process through public housing schemes during the first phase. These schemes were essen tially project based, implemented in a top-down fashion through metropolitan devel opment authorities or state housing boards. By the end of the Third Five Year Plan, most states had established a Housing Board to undertake housing construction. Pub lic sector investment in housing was also at its peak, ranging between 22% and 27% (Table 3.5). Table 3.5. Investment in Housing by Plan Periods Plan Pe Investment (Rupees, million) Percent Share riod Public Private Total Public Private First Plan 1951-56 2,500 9,000 11,500 22% 78% Second Plan 1956-61 3,000 10,000 13,000 23% 77% Third Plan 1961-66 4,250 11,250 15,500 27% 73% Annual Plans3 1966-69 2,500 9,000 11,500 22% 78% Fourth Plan 1969-74 6,250 21,750 28,000 22% 78% Fifth Plan 1974-79 7,960 36,400 44,360 18% 82% Sixth Plan 1980-85 14,910 180,000 194,910 8% 92% Seventh Plan 1985-90 24,580 290,000 314,580 8% 92% Sources: Unless otherwise mentioned, National Buildings Organisation (1996), p. 262. Note: Data (a) obtained from Reserve Bank of India (1978). Since cooperatives were considered as suitable alternative to the state, housing cooperatives received direct state support in implementing some of the government schemes. Table 3.6 shows the schemes implemented during Phase I, and the role of housing cooperatives. In reality, however, cooperatives played a minor role in the im plementation of the public housing schemes. For example, in the First Five Year Plan, two schemes—the Subsidised Industrial Housing Scheme (1952) and Low Income Group Housing Scheme (1954)—were introduced. Under the first scheme, coopera 69 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. tive societies of industrial workers were offered subsidies and loans in the construction of housing. Under the second scheme, loans were given to individuals and coopera tives whose members’ belonged to low-income groups. At the end of the First Plan, however, 85 percent of housing construction under Subsidised Industrial Housing Scheme was carried out by state governments; 13 percent was done by private em ployers, and a mere 1.6 percent by housing cooperatives (Government of India, 1956, p. 557). Cooperatives played little role in implementing the Low Income Group Hous ing Scheme. Village Housing Scheme included a minor role for cooperatives—in sev eral states, cooperatives of backward castes were constituted for providing financial assistance and allocating land (Government of India, 1956, p. 41). Other schemes such as Rental Housing for State Government Employees and Middle Income Group Housing did not include cooperatives. The most significant institutional support for housing cooperatives came with the Report o f the Working Group on Housing Co-operatives in 1964. The Central government had formed the Working Group to explore the role of housing coopera tives. It identified two inherent advantages of the organizational features of housing cooperatives and sought to expand their role in the housing process. These advantages were: (i) scale economy in planning, building materials, labor and equipment; and (ii) more efficient and more satisfactory surroundings according to community standards. Based on the Working Group’s recommendations, the National Cooperative Housing Federation (NCHF) was established in 1969. NCHF has since been instrumental in advocating for housing cooperatives, and helped establish state level Apex coopera tives across the country for m ob ilizin g finance for primary cooperatives (Section 3.4). 7 0 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 3.6. Phase I (1947-1970) Schemes and Role of Housing Cooperatives Year Scheme Purpose and role of cooperatives, if any 1952 Subsidized Industrial Housing Scheme Scheme for subsidy and loans to industrial em ployers, cooperatives of industrial workers, and State governments. 1954 Low Income Group (LIG) Housing Scheme Scheme for loans of upto 80 percent to indi viduals and co-operatives of low-income households; loans also given to housing boards for housing construction. 1956 Subsidized Housing Scheme for Plantation Workers Scheme for loans up to 80% of cost of housing construction for plantation workers. 1956 Slum Clearance and Improvement Scheme Scheme for financial assistance to State gov ernments and local bodies for slum clearance and re-housing, including land acquisition 1957 Village Housing Scheme Scheme for implementation of lay-out plans and rebuilding of houses in clusters; co-ops organ ized for manufacturing building components 1959 Rental Housing for State Government Em ployees Scheme for loans from Life Insurance Corpora tion (LIC) to State governments for housing to government employees 1959 Middle Income Group (MIG) Housing Scheme for LIC loans to Middle Income Group households through State governments for housing construction 1959 Land Acquisition and Development Scheme Scheme for loans to State governments for large scale acquisition and development of urban land for plots for Low Income Group households, especially cooperatives. 3.3.2. Phase II (1970-1988) In Phase II, although many of the public housing schemes were continued, policies distinctively changed from an interventionist approach. The Fourth Five Year Plan (1969-1974), recognized that public sector could not solve the housing problem, a sentiment that would recur in later years: “The experience of public housing so far is that its unit costs are high and that with the constraint of resources it is not possible for public operations to touch even the fringe of the problem” (Government of India, 1970, p. 402). The Fifth Five Year Plan recognized that “the bulk of the housing ef 71 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. fort will necessarily be in the private and the cooperative sectors” (Government of In dia, 1974, pp. 261). Policy focus shifted from public housing projects to broader poli cies that considered the housing problem within the broader context of urban and re gional development programs. The Fifth Five Year Plan consolidated urban develop ment programs by instituting the Task Force on Planning and Development of Small and Medium Towns and Cities in 1975. The Task Force’s recommendations resulted in the Integrated Development of Small and Medium Towns, which was introduced in the Sixth Five Year Plan (1980-85). The program aimed at a balanced distribution of urban population and at slowing the growth of metropolitan areas by providing infra structure and other facilities in the small towns. In keeping with the broader policy context, housing policies in Phase II were directed towards two basic elements of housing—access to housing finance and land. Although both issues were not new, policies were now explicitly directed towards ad dressing them. The difficulty of obtaining formal housing finance induced households to depend on informal financing mechanisms.5 The Banking Commission formed in 1971 and the Shah Working Group on Housing Finance formed in 1978 had high lighted the lack of adequate sources of institutional finance for housing and barriers to providing housing finance (Reserve Bank of India, 1978). Commercial banks hardly specialized in housing finance. Life Insurance Corporation (LIC) had been the main source for formal housing finance until the 1970s. Housing cooperatives became important vehicles for provision of housing fi nance in an environment where other formal housing finance mechanisms were yet undeveloped. Initially the state provided direct financial support to State level A pex 5 Dependence on informal housing finance has been extensively covered in housing literature. See Boleat (1987) and Renaud (1987). 72 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cooperatives. Finance through other sources was also was also channeled through co operatives. LIC became the most significant source of finance for cooperatives, al though three other major financial organizations came up in Phase II. The first or ganization is Housing and Urban Development Corporation (HUDCO), which was established as a techno-financial organization in 1970. One of its main objectives is to provide financial and technical assistance for housing and infrastructure programs in urban and rural areas. Its scope of lending was initially limited to public agencies. HUDCO expanded its scope for lending to cooperatives in 1974-75, but the major thrust to lend to Apex cooperatives started in 1985, “with a view to giving a big push to the cooperative sector during the Seventh Plan period” (Sharma, 1995, p. 34). The second organization is Housing Development and Finance Corporation (HDFC), a private sector housing finance organization that was setup in 1977. It lends mainly to individuals, and its procedures are well established for borrowing by indi vidual members of housing cooperatives. HDFC lends mainly at market interest rates, although it has a line of credit through KfW, Germany for providing a limited amount of loans to non-governmental organizations and State level public housing agencies for social housing schemes. The third organization is National Housing Bank, which was set up in 1988 as a wholly owned subsidiary of the Reserve Bank of India for promoting, funding (espe cially refinancing), and regulating private Housing Finance Institutions (HFIs). Its refinancing program has been especially significant in an environment of regulated interest rate—it gave an impetus for the private sector to lend for housing purposes.6 6 A s a subsidiary o f Reserve Bank o f India, NHB could raise funds at low costs and use them for refinancing at lower interest rates. 73 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In 1995-96, about 20 percent of NHB’s disbursement for refinancing was through Apex housing cooperatives. With respect to land, the Fifth Five Year Plan reinforced the need for a “well formulated land policy” since “one of the main impediments in the way of increased housing activity is the non-availability of land at reasonable prices” (Government of India, 1974, p. 261). Large scale migration to cities in the post-Independence period created massive urban housing shortage. On one hand, slum population increased drastically (reaching over 30 percent in some cities); on the other hand, land prices escalated. Table 3.7. Phase II (1970-1988) Programs/ Organizations and Role of Housing Cooperatives Year Program/ Organization Purpose and role/ scope of cooperatives, if any 1970 Housing and Urban Development Corpo ration (HUDCO) HUDCO finances projects for low-income hous ing; it funds Apex cooperatives for undertaking such schemes. 1974 Environmental Im provement of Urban Slums (EIUS) Program to provide a minimum level of services in slums for selected cities. 1976 Urban Land (Ceiling and Regulation) Act Law for government acquisition of land owned by individuals over a prescribed ceiling; coopera tives exempt from provisions of the law. 1977 Housing Develop ment and Finance Corporation (HDFC) HDFC provides market rate loans for housing to individuals and cooperatives. 1985 Indira Awas Yojana (IAY) Program for rural housing, implemented as a part of Rural Landless Employment Guarantee Pro gramme. 1986 Urban Basic Services Program for poverty alleviation by providing ba sic urban services. 1988 National Housing Bank NHB promotes, funds, and regulates private Housing Finance Institutions (HFIs); it refinances State level (Apex) housing cooperatives. 7 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Land policies were aimed towards an equitable distribution of land through land acquisition and distribution. Cooperatives were given preferential treatment in terms of allocation of land. Since cooperatives were considered as ideal alternative to the state for equitable distribution of land, they were either exempt from land acquisi tion procedures, or were given preferential treatment in the allocation of acquired land. For example, the Urban Land (Ceiling and Regulation) Act enacted in 1976 sought to place limits on urban property holdings. However, cooperatives were exempt from the provisions of the Act. Table 3.7 outlines the programs and organizations that came up during Phase II. 3.3.3. Phase III (1988-present) The demand for economic liberalization started in the mid 1980s. Government policies started to acknowledge the limitations of state outreach and recognized the significance of private sector. The Planning Commission noted in the Eighth Five Year Plan, “we have to start rolling back the Public Sector investment from those sec tors of the economy where the private sector can move in and step up our investment in the social sector” (Government of India, 1992, p. i). Explicit thrust for economic liberalization came with the New Economic Policy of 1991, which marked a substan tive change in this direction. Under the policy, privatization gained more currency, and public sector investment was rolled back. Public sector investment in housing, for example, dropped from 27 percent during the Third Five Year Plan to 8 percent during the Seventh Five Year Plan (Table 3.5). Public-private partnerships were also increas ingly encouraged with the shift in policy. Further, many of the administrative and fis cal responsibilities were devolved to the local governments with the Constitutional Amendments 73 and 74 passed in 1992. 7 5 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Two events in 1988 were instrumental in changing the course of housing pol icy. First, the National Commission on Urbanisation (NCU) tabled its report, which considered housing in the context of broader urban development. Consequently, many of the policies that emerged were more broadly focused on urban development rather than being sectoral. For example, NCU’s recommendations were used as base for formulating the Mega Cities Program in 1993. More broadly, however, the Integrated Development of Small and Medium Towns (IDSMT) program launched in 1979-80 already gave a thrust to integrated provision of services along with shelter. With re spect to housing cooperatives, NCU (Government of India, 1988, p. 221) recom mended an expanded role of cooperatives to “optimise the use of scarce resources of land, building materials, and finance.” It recognized that “it is a distinctive feature of the cooperative housing movement that members of cooperative societies assist each other in building houses” (p. 221). NCU’s strategies for promoting housing coopera tives included: allocating land to cooperatives of low and middle income households; establishing mutual benefit organizations to serve the poor; providing finance and building materials to augment the pooling of resources in the collective; and providing soft loans to voluntary agencies that organize poor households into such cooperatives (p. 221). Second, the United Nations Centre for Human Settlements (UNCHS) advo cated governments to formulate a National Housing Policy based on enabling ap proach (UNCHS, 1990). The approach had emerged in response to the failure of con ventional public housing strategies, and aimed to enhance the role of the private and Third sectors in the housing process (UNCHS, 1996). Broadly, enabling marked a shift from project to process based approach. On one hand, enabling policies focused on institutional and organizational reforms for increasing the efficiency of housing 76 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. markets (e.g. property rights, land use and zoning, housing finance, rent control, hous- t n mg norms, and so on). On the other hand, the enabling policies aimed to mobilize the potential of various actors in the housing market (e.g. public agencies, private firms, third sector organizations, and individual households). The Habitat II Conference in Istanbul (1996) reinforced the formation of such partnership to “integrate and mutually support objectives of broad-based participation through, inter alia, forming alliances, pooling resources, sharing knowledge, contributing skills and capitalizing on the com parative advantages of collective actions” (UNCHS, 1996, p. 22). In the political economic context of economic liberalization, India also adopted the enabling approach as the basis for National Housing Policy, which was first drafted in 1988 and went through several revisions until 1998. Consistent with the enabling approach, the National Housing and Habitat Policy adopted in 1998 included: removing legal, financial, and administrative barriers for facilitating access to land, finance and technology; forging partnership between private, public and cooperatives to enhance the capacity of the construction industry; and empowering village coopera tives to mobilize credit. According to the policy, corporate, private and cooperative sectors would (i) take the lead role in land assembly, construction of houses and de velopment of amenities; (ii) forge partnership with State Governments and local au thorities for construction of houses; and (iii) collaborate with the State Governments to work out schemes for slum reconstruction on cross subsidization basis (Government of India, 1998b, Section 3.4). In this, the National Housing Policy recognized “group- based or cooperative activity or community associations as the principal form of hous ing activity in m any rural settlem ents and large urban areas facing shortage o f land” 7 This was especially the emphasis o f the World Bank (1993), which advocated "the reform o f government policies, institutions, and regulations to enable housing markets to work more efficiently” (p. 1). 77 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. (Government of India, 1998, p. 17). Policies like Swama Jayanti Shahari Rozgar Yo- jana (SJSRY), which were oriented towards providing employment to the urban un employed, relied on local development through community organizations (e.g. neighborhood groups and Community Development Societies). The National Housing and Habitat Policy also proposed a national scheme for construction of 2 million housing units per year for the poor (Economically Weaker Section and Low Income Groups) during the Ninth Five Year Plan (1997-2002). The scheme was initiated in 1999, with 1.3 million units earmarked for rural areas, and 0.7 million units earmarked for urban areas. Of these, 100,000 units have been targeted for construction by housing cooperatives annually. Table 3.8 summarizes the major programs undertaken during Phase III, along with the role of housing cooperatives. The main thrust of the enabling policies has been, however, on privatization and consequent policy reforms. In this context, housing cooperatives do not have the same level of preferential treatment as they did in the context of democratic socialism. In the changed context, leaders of the cooperative movement became apprehensive of liberalization policies. They feared withdrawal of concessions and started making ap peals for increased state assistance.8 In this context, cooperative leaders also sought reforms in cooperative laws and policies to compete effectively with the private sector. The Brahmaprakash Committee gave a thrust in this direction by prescribing a model act to provide more autonomy to cooperatives (Government of India, 1991). How ever, few states adopted the model act. In 2000, the National Cooperative Housing Federation also drafted a Model Law on Housing Cooperatives. The law, if adopted, 8 For example, the Convention o f Parliamentarian's Forum for Developm ent o f Cooperative M ovem ent argued: “In the wake o f liberalisation, privatisation and globalisation o f Indian econom y and with the emphasis on the opera tion o f market forces, the cooperatives are finding it difficult to compete with the market forces. ... In order to create [such] a conducive environment, the government will have to extend both financial and institutional sup port.” (National Cooperative Union o f India, 1997, p. 7-8) 78 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. would place housing cooperatives as a separate sector, rather than including housing within the broader ambit of cooperatives in general. Table 3.8. Phase III (1988-present) Programs and Role of Housing Cooperatives Year Scheme Purpose and role of cooperatives, if any 1989 Night Shelter Scheme Scheme to provide shelter and sanitary facilities to pavement dwellers 1989 Nehru Rozgar Yo- jana Program for persons living below poverty line in urban areas (Includes SUME; SUWE; SHASU) Scheme of Urban Micro Enterprises (SUME) Loan cum subsidies program for training and in frastructure support to unemployed and underem ployed youth for self-employment. Scheme of Urban Wage Employment (SUWE) Employment program for the poor in smaller towns Scheme of Housing and Shelter Upgra- dation (SHASU) Employment program for persons involved in housing and building activities 1991 Urban Basic Ser vices for the Poor (UBSP) Program integrated earlier EIUS and UBS pro grams 1993 Mega City Program Revolving Fund program for infrastructure provi sion in selected Mega Cities 1996 National Slum De velopment Program Program for implementing physical and social infrastructure in slums 1997 Swama Jayanti Shahari Rozgar Yo- jana Program replaced NRY and UBSP. It consists of 2 schemes: USEP and UWEP Urban Self Em ployment Program (USEP) Program includes: (i) self-employment through Micro-enterprises; (ii) Development of Women and Children in Urban Areas; (iii) Information, Education and Communication Urban Wage Em ployment Program Program for employment of urban poor in con struction of socially and economically useful as sets 1998 Credit cum Subsidy Scheme Program provides credit and subsidy as a part of the Indira Awas Yojana 1999 Two Million Hous ing Program Program for constructing 2 million houses annu ally for the poor (EWS and LIG); the target for cooperatives is 100,000 units annually. 7 9 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3.4 The Role of National Institutional Structure Development experts like Peter Bauer argue that cooperatives grew in India due to direct state support. However, as it is evident from above, although the national institutional framework provided a conducive environment for the growth of housing cooperatives, housing cooperatives played only a small role in implementing state sponsored schemes. Further, direct state support also reduced as India’s political economy transformed from democratic socialism to economic liberalization. Thus, direct state support, by itself, does not explain the phenomenal growth of housing co operatives in India. Rather, I argue that supportive institutional structures were crucial to the development of cooperatives. The structures helped in enhancing cooperatives’ collective strengths, while reducing concurrent weaknesses. They also helped in re ducing cooperatives’ transaction costs of formation and functioning. Three elements of the cooperative institutional structure can be identified in India, which helped expand the role of housing cooperatives. The first element is the institutional structure providing administrative, legal, and other procedural support. While the Registrar’s Office is institutionalized across India, other support mecha nisms have also developed to various degrees in different states. The second ele ment—which is perhaps most significant for growth of housing cooperatives nation wide—is the two-tiered national institutional structure for channeling financial support to cooperatives. The third element, which is much less clearly established than the other two, is the institutional support for accessibility to land. All three elements clearly bear the imprint of state patronage in the Indian context. First, the Registrar’s office continues to be a state organ. Second, the two-tiered national institutional struc ture for finance was initiated by the state. Third, cooperatives’ accessibility to land has also depended on the extent to which state governments have given preferential 8 0 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. treatment in allocating land. Once set-up, however, these the institutional structures provided various degrees of support for growth of housing cooperatives, despite wan ing of direct state support. 3.4.1. Administrative, Legal, and Procedural Support Structures Although administrative, legal, and procedural support structures vary across India, the Registrar’s Office is well established across the country. When the office was initiated under the Cooperative Societies Act, 1904, it was expected that the “fos tering care of the Registrar would be less required until his duties became purely offi cial” (Catanach, 1970, p. 51). However, over the years, the Registrar gained signifi cance as “friend, philosopher, and guide” in the development of housing cooperatives. Indeed, the Cooperative Registrars became so powerful that they are also referred to as “Brahma, Vishnu, and Shiva” rolled into one.9 The Registrar’s office has been historically a state organ, both during the pre- and post-Independence periods. State governments control the functioning of housing cooperatives through the Registrar’s office. Most states have historically had a coop erative ministry, under which the Registrar’s office is located. The Registrars are re sponsible for the administration of the Cooperative Societies Act in each state. They are responsible for the promotion, development, and management of housing coopera tives. The Registrar’s functions generally include granting registration to new coop eratives, which provides legal incorporation for formal cooperatives; approving/ amending a cooperative’s bye-laws; auditing accounts of cooperatives annually; amal gamating, transferring, or winding up cooperatives; settling disputes between coopera tives or between cooperative boards and members. The precise nature of power of the 9 In Hindu m ythology, Brahma, Vishnu, and Shiva are three Gods responsible for the creation, preservation, and destruction o f the universe respectively. 81 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Registrar’s office differs between states, depending on the role accorded by the state government. On one hand, the basic role of the Registrar’s office to comply with Coopera tive Societies Act has been broadly useful in formation and functioning of coopera tives according to established cooperative principles. In case of uncertainty, the Reg istrar’s office steps in to clarify procedures and to provide guidelines. Such assistance helped in overcoming some of the internal collective problems in the administration and management of cooperatives. On the other hand, sweeping powers accorded to the Office also constrained the autonomous development of cooperatives in a few states. Such state governments used the Registrar’s office to politically interfere in the daily functioning of cooperatives; hence, Registrar’s office controlled, rather than pro viding support to cooperatives in these states. In other states, where cooperatives have greater degree of autonomy, other institutional structures supporting the administra tive, legal, and other procedural issues of cooperatives also emerged. Chapters 4 and 5 highlight the role of Registrar’s Office in Maharashtra and Tamil Nadu, which af fected the differential development of housing cooperatives in Mumbai and Chennai. 3.4.2. Institutional Support for Finance The second element—the two-tiered institutional structure for financial support housing cooperatives—was established nationwide based on the recommendations of the Report o f the Working Group on Housing Cooperatives. The structure principally consists of the state level Apex cooperative and primary cooperatives.1 0 The principal function of Apex cooperatives is to channel housing finance to primary cooperatives. They also perform other functions, such as providing administrative, technical and le 1 0 Some states had several State level cooperatives (e.g. Andhra Pradesh, which had seven). In these states, the Apex cooperative amalgamated the State cooperatives into one. 82 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. gal guidance to primary cooperatives; promoting new primary cooperatives; and pro viding assistance in procuring land and building materials to them (Mathur, 1971, p. 497). The Working Group also recommended setting up of National Cooperative Housing Federation (NCHF) with the “responsibility of promoting, developing and coordinating the activities of housing cooperatives in the country” (NCHF, 1996). Consequently, NCHF was established in 1969. It assists Apex cooperatives in finan cial, legal, managerial, and technical matters, and is supported through annual contri butions from Apex cooperatives. Although the NCHF is not directly involved in dis bursing housing finance, it has helped establish Apex cooperatives across the country in several states. In 1969, nine states and union territories had Apex cooperatives; by 1990, almost every state had an Apex cooperative. The development of Apex coop eratives reduced uncertainty of funding sources for primary cooperatives, thus reduc ing the transaction costs of primary cooperatives in obtaining finance. Apex cooperatives borrow money from various sources for on-lending to pri mary cooperatives after adding additional percentage points to the interest rate to cover their overhead and administrative expenses. Life Insurance Corporation (LIC) of India became a major source of finance for Apex cooperatives, mainly because it is required by law to invest at least 25 percent of the net increase in its investible funds in socially oriented schemes, such as housing. In 1994-95, LIC’s investment in the housing sector was Rs. 8,325.2 million, of which Rs. 1,380 million (or 16 percent) was invested through Apex cooperatives. The two-tiered institutional structure aided in providing a stable source of funds for primary cooperatives. Table 3.9 lists the sources of finance for Apex and primary cooperatives between 1957-58 and 1994-95. As the table shows, Apex coop- 83 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. eratives depended largely on direct government support in 1957-58; since then, Apex cooperatives have considerably reduced their dependence on government support. Since the 1990s, state support for Apex cooperatives is principally through loan guar antees. LIC is a major source of finance for Apex cooperatives (over three quarter of Apex cooperatives’ finance was raised through LIC in 1994-95). Other financial insti tutions such as HUDCO form a smaller component; support from HUDCO is nonethe less important since it is usually targeted towards low-income housing projects (espe cially in rural areas). National Housing Bank (NHB), which was established in 1988, also started to refinance loans of Apex cooperatives. In 1995-96, about fifty percent of housing units refinanced by NHB were through Apex cooperatives.1 1 Thus, Apex cooperatives became relatively independent, relying much less on direct governmental financial support. Table 3.9. Sources of Funds for Housing Cooperatives (percent share) State evel (Apex) Cooperatives Primary Housing Cooperatives 1957- 1964- 1974- 1984- 1994- 1957- 1964- 1974- 1984- 1994- 58 65 75 85 95 58 65 75 85 95 Government 91.2 17.9 3.9 1.7 0.2 41.9 24.8 11.1 3.9 3.6 Life Insurance Corporation 87.8 84.1 76.9 1.3 2.4 1.5 Central financ ing agencies 11.6 7.5 12.5 10.2 10.4 18.0 21.3 14.9 11.1 HUDCO 1.1 3.1 1.1 5.1 12.2 Apex Coopera tives 0.3 0.1 0.1 45.0 53.7 56.5 Others 8.8 70.4 0.4 0.5 9.6 47.8 57.2 20.2 20.0 15.1 Total 100 100 100 100 100 100 100 100 100 100 Source: Calculated from Statistical Statements Relating to Co-operative Movement in India (for respective years). 1 1 Calculated from N H B ’s Annual Report (1995-96), pp. 11-12. 84 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. For primary cooperatives, government and “other sources” were the major sources of finance before the two-tiered institutional structure was established nation wide—these two sources provided over 80 percent of finance for cooperatives in 1957-58 and 1964-65. Since the establishment of the two-tiered structure, Apex coop eratives became the major source of finance for primary cooperatives—their contribu tion increased from 45 percent in 1974-75 to 56.5 percent in 1994-95. Thus, in the face of the reduction of direct government support to primary cooperatives, the two- tiered cooperative institutional structure became instrumental in providing financial support to primary cooperatives. 3.4.3. Institutional support for access to land While the two-tiered institutional structure for providing financial support to cooperatives is well established, there is not an equivalent formal structure for chan neling land to cooperatives. Yet, institutional support for access to land has been pro vided to varying degrees through preferential treatment—housing cooperatives have been given preference in the allocation of land acquired by government, or have been exempt from land acquisition policies. Land acquisition and reform policies were first initiated in rural areas for equi table distribution of land in the context of democratic socialism. Similar policies were also implemented in urban areas due to rapid urbanization following Independence. Large cities in particular had acute housing shortage and overcrowding. In 1959, Land Acquisition and Development Scheme was implemented to grant loans to State gov ernments for large scale acquisition and development of urban land for low income households. In the ideological context of democratic socialism, cooperatives became an ideal alternative for state ownership, and were given preferential allocation of such acquired land. 85 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. In Delhi, for example, the Large-scale Land Acquisition policy adopted in 1961 provided fertile ground for the growth of housing cooperatives. As collective organizations, cooperatives were expected to use land optimally and provide collective services; they were considered a suitable alternative to the state for equitable land dis tribution. Housing cooperatives in Delhi were modeled after Bombay, and Tenure Cooperatives in the form of Cooperative Group Housing Societies (CGHS) emerged in Delhi. Since Delhi Development Authority (DDA) (the local public housing author ity) has traditionally owned most of the land in Delhi, the growth of Cooperative Group Housing Societies depended centrally on allocation of land by DDA. Although the allocation has been uneven over the years, DDA has indeed given preferential al location of land to Cooperative Group Housing Societies. Misra (1986, pp. 70) notes that by 1978, “nearly 65% of the societies [were] provided with serviced land by the DDA at rates very significantly lower than that of the open market.” By 1994, over 12 1200 Cooperative Group Housing Societies had been established in New Delhi. Nationally, the Urban Land (Ceiling and Regulation) Act (ULCRA) of 1976 was a major legislative piece in terms of urban land reform. The Act sought to place limits on urban property holdings to: (i) prevent concentration of urban property; (ii) ensure equitable land distribution, (iii) discourage luxury housing; and (iv) promote orderly urbanization.1 3 The Act put limits on the area of land and house that an indi vidual household can own and occupy. Cooperatives were exempt from land acquisition under the Act. Hence, it be came expedient to form cooperatives to evade the Act. Risbud (n.d.) reports that in Jaipur (Rajasthan), of the 800,000 square meters of land notified by Jaipur Develop 1 2 See Appendix 1 for evolution o f housing cooperatives in N ew Delhi. 1 3 From the Preamble o f ULCRA, 1976. 86 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ment Authority under the ULCRA, only 1,00,000 square meters of the land was actu ally acquired by the authority and the rest was acquired by cooperatives (p. 10). Simi larly, in Lucknow, Amitabh (1997, p. 107) observes that most cooperative societies were formed after the ULCRA was enacted: “Before 1977 there were 14 cooperative housing societies in Lucknow city. The figure increased to 403 in the 1990 (of which 82 were declared defunct CHSs).” ULCRA has been much criticized as an institutional barrier hindering the func tioning of housing land market. Further, it hardly achieved its goal—over the two decades of the tenure of ULCRA, only 8.6 percent of the land declared as surplus was acquired. The Act was formally repealed in 1999 after much criticism. With the re peal of the Act, cooperatives became even more dependent on the state government’s housing policies for their access to land. In order to fulfill cooperative’s quota of housing construction under the Two Million Housing Program, the NCHF has been lobbying state governments to allocate 30 percent of land acquired for housing to co operatives; however, land allocation varies between state governments. The institu tional structure for providing land is thus not well defined. Yet, the historical institu tional support for providing land through exemptions in land acquisition and preferen tial allocation of acquired land cannot be neglected. 3.5. Conclusion Although the British introduced cooperatives during the turn of twentieth cen tury, housing cooperatives grew modestly during the pre-Independence period in In dia. Housing cooperatives were mostly concentrated in Bombay and Madras prov inces. After Independence, housing cooperatives grew phenomenally across the coun try. The initial thrust for development of cooperatives came from the state. The ideo logical context of democratic socialism provided a conducive institutional environ- 87 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ment for the growth of housing cooperatives. However, direct state support does not fully explain the growth of housing cooperatives. I argue that the development of supportive institutional structure has been crucial to the phenomenal growth of hous ing cooperatives in post-independent India. These structures helped in reducing some of the cooperatives’ weaknesses in controlling agency problems. Once set-up, these structures enabled growth of housing cooperatives despite reduction of direct state support. First, broad institutional structures providing legal, administrative, and proce dural support helped overcome some of the collective action problems inherent to co operatives. In particular, the Registrar’s office, which is the oldest element of institu tional structure, gained significance over the years, and became an important element in overseeing the growth of housing cooperatives. Second, the national two-tiered institutional structure of Apex and primary co operatives has been useful in providing financial support to cooperatives. In a context when direct financial support from government reduced, the institutional structure for finance was instrumental in providing a stable source of finance to primary coopera tives. Third, preferential treatment for access to land helped in the development of housing cooperatives, although such support differs across states. 8 8 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4. Housing Cooperatives in Mumbai While housing cooperatives were active before Independence in Mumbai,1 their activities grew phenomenally after Independence. In the pre-Independence pe riod, housing cooperatives were predominantly in the form of Tenure Cooperatives for collective ownership and management. In the post-Independence period, private de velopers, public housing agencies, and non-governmental organizations adapted coop eratives as vehicles for forming collectives for fulfilling various housing objectives. Although most formal cooperatives are yet tenure based, they have been adapted for diverse roles in public housing, slum redevelopment, and micro-fmance. I argue that both the local housing market and the local institutional framework aided the growth of housing cooperatives in Mumbai. With respect to the local hous ing market, historical economic and demographic characteristics of Mumbai lead to dense multi-family housing morphology since late 19th century. In this context, de mand for collective organizations like cooperatives grew. With respect to the local institutional framework, rent control policies and flat2 ownership rules since Inde pendence favored the growth of housing cooperatives. While ownership became more optimal due to rent control, flat ownership regulations required formation of housing cooperatives in multi-family units. A rather robust cooperative institutional structure supporting cooperatives evolved in the post-Independence period. The structure helped cooperatives to over come some of their internal weaknesses, and also reduced their transaction costs of 1 Mumbai was called Bombay until 1997. The state in which the city is located bore the eponymous name o f Bom bay State until 1960, when the state was bifurcated into Maharashtra and Gujarat. Mumbai is the capital o f Ma harashtra State. 2 Flats are the same as apartments in Mumbai, and are often interchangeably used. However, flat ownership legally refers to units that are collectively owned in multi-family housing; apartment ownership refers to units that are individually owned (i.e. condominium units). 89 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. formation and functioning. First, due to liberal cooperative policies in Maharashtra, state control has been minimal. Structurally, Registrar’s office could hardly interfere in the internal decision making processes, so that cooperatives were relatively autono mous. Additional structures for administrative, legal, and procedural support also arose. Second, the two tiered institutional structure for finance through Apex co operatives provided a stable source of funding in the initial decades. Third, although there is no well defined structure for providing land, public housing agencies gave preferential treatment to housing cooperatives. Consequently, housing cooperatives grew phenomenally in Mumbai, and took on diverse roles. The chapter is organized in the following way. First, I examine the growth of housing cooperatives before Independence. Second, I outline the growth of housing cooperatives after Independence. Third, I examine the institutional context of growth of housing cooperatives in Mumbai. Fourth, I outline the support provided by the co operative institutional structures in the development of cooperatives. Fifth, I outline the diverse roles of housing cooperatives in Mumbai. Finally, I summarize the condi tions that led to phenomenal growth of housing cooperatives in the city. 4.1. Pre-Independence Period Rapid economic growth of Mumbai during the late 19th century resulted in the modernization of the city, almost erasing its underlying rural character. Urban and housing development that followed was modernistic with an explicit bias towards Western urban form. Housing morphology in the form of multistoried buildings be came the norm. Housing conditions deteriorated as rapid commercial development brought waves of immigrant laborers, who were mostly males. In this context, coop eratives emerged as “modem” organizational vehicles for providing “decent” housing 90 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. to the middle class. Housing cooperatives were principally in the form of Tenure Co operatives. 4.1.1. Mumbai’s Urban Growth and Demography, 1860-1947 A brief background of Mumbai’s urban economic growth and its demography is required to understand the evolution of Mumbai’s local housing market. Mumbai’s economy gained significance as a colonial maritime trade center since the early nine teenth century, and was ranked next only to Calcutta in the Indian subcontinent. Sev eral political and economic factors contributed to the city’s economic growth since 1830s. Politically, Mumbai gained much significance for the British as the adminis trative capital for Western India. Economically, export and import through Mumbai port rose steeply. While cotton exports to Britain and China rose sharply, opium trade with China contributed even more to Mumbai’s economy.3 Although manufactured textile, tea, and metals were imported through the port, there was an export surplus. Infrastructure improvements at both the local and global level furthered Mumbai’s world trade potential. At the local level, advent of railways during 1850s strengthened Mumbai’s linkage with its cotton producing hinterland. At the international level, opening of Suez Canal made Mumbai the closest port in Indian subcontinent to trade with Europe and it became the “Gateway of India.” Table 4.1 shows the five-year average merchandise export and import through Mumbai between 1871 and 1939. As the table shows, there was a spurt in exports during 1861-1865. Indeed, the major thrust for Mumbai’s economic growth came dur ing this period. Due to the American Civil War, British mills faced constraints in the import of raw cotton from America; hence, it became expedient for these mills to im 3 Markovits (1996) reports that while merchandise exports rose six fold between 1830 and 1860, opium sales rose more than ten-fold. Share o f cotton trade grew from 36 percent to 44 percent. 91 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. port cotton from western India through Mumbai. Consequently, there was a boom in cotton exports through Mumbai. Although the boom subsided after the Civil War ended, economic growth continued due to the growth in cotton mills set up by the lo cal entrepreneurial Parsi and Gujarati community. The first indigenous cotton mill was established in 1854 and over the next two decades, fifteen mills employing over 11,000 workers came up (Kosambi, 1986, p. 34). Related manufacturing industries including railway workshops also grew in this period. By 1908, Mumbai had 166 fac tories (of which, about 50 percent were cotton mills) employing nearly 130,000 people (over 13 percent of the population) (Kosambi, 1986, p. 52). Mumbai became “Man chester of India” by the beginning of twentieth century. The second economic boom came after World War I. Further improvements and extension of railway network in Mumbai in 1915 and surge in export trade after the war increased the city’s competi tive position in the world trade market. Table 4.1. Five Year Average Merchandise Export and Import through Mumbai, 1871-1939 (Rupees, million) Exports Change (%) Imports Change (%) 1871-76 221 119 1876-81 225 1.8 154 29.4 1881-86 320 422 210 36.4 1886-91 370 15.6 275 31.0 1891-96 385 4.1 282 2.5 1896-01 326 -15.3 348 23.4 1901-06 438 34.4 332 -4.6 1906-11 510 16.4 428 28.9 1911-16 580 13.7 522 22.0 1916-21 880 51.7 839 60.7 1921-26 1027 16.7 908 8.2 1926-31 705 -31.4 805 -11.3 1934-39 332 -52.9 583 -27.6 Source'. Markovits (1996), p. 32. 9 2 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. The post World War I economic boom subsided in the mid 1920s, and Mum bai’s export based economic growth reduced since then. Labor-management relations in the cotton mills also deteriorated as labor strikes grew. Mills could hardly adopt newer technological improvements to keep their competitive position in world trade, especially with respect to Japanese mills.4 Growth of other inland cotton producing centers also dipped Mumbai’s share of cotton exports. As Markovits (1996, p. 36) notes, After 1922, the Bombay cotton-textile industry ceased to grow and the 1920s and 1930s saw the rise of inland centres such as Ahmedabad and Coimbatore. Between 1920 and 1939 Bombay’s share in India’s cot ton-textile industry fell from 43.9 to 28.3 per cent in spindles, and from 51 to 33.2 per cent for looms. Further, the 1929 Depression had an adverse effect on Bombay’s trade. Between 1921-26 and 1934-39, Mumbai’s export fell from Rs. 1,027 million to Rs. 332 million; its import fell from Rs. 908 million to Rs. 583 million (Table 4.1). Despite the decline in trade, Mumbai’s economy remained stable due to economic diversification towards financial services and manufacturing. Mumbai’s population growth roughly followed the city’s pattern of economic growth. Mumbai’s population exceeded half million by 1849 (Sykes, 1852). There was a spurt in population growth in response to labor demand created by the economic boom of 1860s. The city’s population reached 816,552 by 1864, but soon reduced to 644,405 in 1872 after the boom was over. Over the next two decades, population grew steadily to again reach 821,674 in 1891. Population growth dampened temporar ily with the plague in 1896,5 but it soon resumed in the twentieth century. Mumbai’s 4 See W olcott (1991, 1994, 1999) for M umbai’s world trade position in cotton and textile with respect to Japanese cotton m ills during the inter-war period. See Gordon (1978) for the nexus between internal local politics and busi nessmen. 5 See Klein (1986) for a review o f urban development, disease, and death in Mumbai between 1870-1914. 93 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. population crossed 1 million with the second economic spurt following World War I, and reached nearly 1.49 million in 1941 (Table 4.2). Table 4.2. Population Growth and Sex Ratio in Mumbai, 1849-1941 Population Average annual growth rate Sex Ratio (Females per 1000 males) 1849a 566,119 662 1864t > 816,552 2.9% 1872 644,405 -2.6% 613 1881 773,196 2.2% 662 1891 821,764 0.6% 585 1901 776,006 -0.6% 617 1911 979,445 2,6".. 529 1921 1,175,914 2.0°«. 524 1931 1,161,383 -0.1% 552 1941 1,489,883 2.8% 581 Source: Unless otherwise mentioned, Kosambi (1986, p. 165). Notes: (i) Data (a) obtained from Sykes (1852) (ii) Data (b) obtained from Upadhyay (1990), p. PE-92. Table 4.3. Contribution of migration to population growth of Mumbai, 1901- 1941 Year Net Growth (’000s) Natural Increase (’000s) Net migration (’000s) Percent Natural In crease Percent Net Migration 1901-1911 221 -129 350 -58 158 1911-1921 232 -168 400 -72 172 1921-1931 17 -58 75 -29 129 1931-1941 404 4 400 1 99 Source: Zachariah (1968, p. 15) Much of the population growth in Mumbai was contributed by migration. As Table 4.3 indicates, contribution of migration to population growth was far more sig nificant than natural increase—during the five decades before Independence, contribu tion due to natural growth was either negative or an very small percentage. Kosambi (1986, p. 55) reports, “from 1881 to 1931, only about one-fourth of Bombay’s inhabi tants were bom in the city itself, and in times of prosperity which attracted a larger 9 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. influx of immigrants, this proportion sank even further, as for example, to 16 percent of the total in 1921.” Immigrants came to the city primarily seeking jobs and were usually employed in trade, harbor, and textile industry; some were also employed as domestic servants. Over 80 percent of the immigrants were from within the Bombay Presidency (mostly Konkan and Deccan regions), and a smaller proportion came from other distant states (Kosambi, 1986, p. 167). Labor immigration into Mumbai gave the city a distinctive characteristic with respect to sex ratio. Males were more prominent than females, primarily because male laborers left their families in the village in order to work in the city. Between 1849 and 1941, female-male ratio did not exceed 662:1000, and reached as low as 524:1000 in 1921 (Table 4.2). Preponderance of men in some industries (e.g. harbor, dockyards, mills) further lopsided the ratio towards men in certain pockets of the city. The labor migration to Mumbai was unlike Chennai where migrants came to the city along with their families, so that the sex ratio was more balanced (sex ratio reached over 950 fe males per 1000 males in Chennai). This distinctive demography of Mumbai affected the nature of housing development—especially labor housing, which consisted of sin gle room units serving as sleeping quarters for single men and was usually located in close proximity to the workplace (e.g. mills, dockyards, and so on). Most men lived as cheaply as possible and remitted money to their family back in the village. Accord ing to a survey conducted by the Labor Office in Mumbai in 1921-22, nearly 53 per cent of the single men did not spend anything on house rent and their budget reflected a “comparatively large remittance upcountry, the equivalent of 26.2 percent of their pay” (Shirras, 1923, p. 35-36). 9 5 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.1.2. Mumbai’s Local Housing Market, 1860-1947 Mumbai’s urban pattern followed the typical British colonial city, wherein the residential areas of the British settlers and the Indian natives were clearly segregated. Until the mid-nineteenth century, urban growth was principally restricted to south Mumbai, centering on the port. The British built the Fort and military cantonment ar eas of Colaba for defense. They lived on the south side of Fort and their residential areas were characterized by spacious, single storied houses (“bungalows” with a “ve randah” and garden). By the 1860s, the British expanded their spacious residences towards the west (Malabar Hill and Cumballa Hill) and a few inland suburban pockets towards the north (e.g. Parel) (Kosambi, 1986, p. 43-44). Native Indians lived on the north side of the Fort in dense, overcrowded settlements. A fire broke out in 1803 in the northern area of Fort, in which much of Indian merchants’ property was damaged. After the incident, a separate Native Town was formalized for the Indians. Indian set tlements also grew in response to the colonial commercial and military needs (e.g. Dhobi Talao, where washermen for soldiers lived). Much of Mumbai’s urban development happened after the economic boom of 1860s. First, Fort walls were completely dismantled by 1862 to create space for pub lic buildings (several landmark colonial buildings were built in the space following the demolition). The Fort area evolved into a Central Business District. Elaborate colo nial public buildings set in Gothic style came up between 1860 and 1900. Second, ex tensive land reclamation schemes were also proposed to mitigate congestion due to concentrated urban growth in south Mumbai. Land reclamation was initiated to “pro vide residential accommodation and open spaces in the south of the City” (Reed and Sheppard, 1928, p. 367).6 6 See Banerjee-Guha (1996) for the impact o f land reclamation schemes on Bom bay’s urban development. 96 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The rapid commercial development following the economic boom virtually obliterated the underlying rural character of Mumbai. Backbay reclamation schemes and urban development coalesced the seven islands that originally formed the city (hence this part is also called Island City). Colonial style urban development towards the south, back-bay reclamation along east and west, cotton mill districts (e.g. Byculla, Mazagaon, Parel), and retail market and housing developments transformed Mumbai into a modem city. Evenson (1996) notes that the rise of colonial style paralleled a decline in traditional Indian architecture, which was “abetted by the circumstance that wealthy Indians, instead of encouraging local craft traditions, often lent their patron age to buildings in European style” (p. 171). Only a few settlements with indigenous mral character remained by 1920s. This was unlike Chennai, where internal village structure of the city had been in tact. Housing followed the modernization trend of urban development. Concentra tion of urban growth to the south and limited transportation availability towards the n north contributed to vertical rather than horizontal residential expansion. Flats and single room tenements in multistoried buildings became common. Evenson (1996, p. 176) notes that Mumbai was, ... in many ways comparable to a western industrial metropolis, provid ing its inhabitants with a range of urban building types and a way of life that was attaining a degree of internationalism. As early as 1906, a visitor to Bombay had observed, ‘great block of flats’, and by the 1920s apartment living was becoming common for the middle classes. Three types of housing development emerged in this context. First, in the dense native town areas, five and six storied structures with mixed use (retail shops in the first floor and dwellings in the floors above) came up. Second, in peripheral areas 7 About 37 percent o f the population was concentrated in less than 4 percent o f M umbai’s island area (Klein, 1986, p. 732). 97 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. of the native town, “wadis” or two and three storied buildings clustered around a cen tral space became prevalent; usually, same kinship or subcaste groups occupied these wadis. Third, chawls (single room accommodations with common sanitary facilities) were established as workers’ housing near the cotton mills (Kosambi, 1986). Chawls were generally of three types. First, they were single family houses that were subdi vided so that maximum amount of space could be rented out. Second, they were pri vately built two to four storied structures with a central corridor. These structures had little light and ventilation. Usually shopkeepers selling daily consumer wares to la borers occupied the first floor. Third, they were “public” housing constructed by Bombay Improvement Trust, which were better lit and ventilated (Upadhyay, 1990, p. PE-93). While economic growth induced modem urban and housing development in Mumbai, it was also accompanied by housing deficiency. Real estate costs rose sharply and rental housing was on par with some British cities (Evenson, 1996). Un der such high demand, even poorly constructed and poorly maintained housing be came expensive. Ownership housing was beyond the means of even middle class households, who resorted to rental housing. Talmaki (1920, p. 191), the leading pro ponent of cooperative housing in Mumbai, observed: People pay high rents and yet have insanitary houses. Land values are rising by leaps and bounds, the cost of labour is heavy, and the prices of materials are increasingly high. Construction of houses is an enter prise entirely in the hands of capitalists. The middle class small owners have disappeared or are disappearing, and the vast majority of the peo ple are mere tenants paying rent to capitalist owners. Post World War I economic boom put more pressure on housing as more mi grant laborers arrived in the city. A rent control law (Bombay Rent (War Restriction) Act II and VII) was implemented in 1918 to curb increase of housing rents following 98 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. World War I. Rent control, however, had an adverse effect on the housing market. Landlords “withdrew rooms from the market, preferring to leave them vacant and to create scarcities which enabled them to levy additional cesses, rather than letting them unprofitably at controlled rents” (Chadavarkar, 1994, p. 183-184). Although rent con trol did not inhibit building construction, such construction was directed towards the higher end of the housing market, which was not included within the scope of rent control. In the rental market, pugdee system, or the system of demanding a large cash payment in advance, became common among landlords. Housing conditions were worse for the labor class. Nearly 70 percent of tene ments in the city consisted of single rooms. According to the survey by Labor Office in 1921-22, about 97 percent of the working class families lived in such single rooms, which housed between two and eight families. Over 75 percent of such overcrowded tenements were located in cotton mill areas (e.g. Byculla, Parel, Mahim) (Shirras, 1923, p. 23-25).8 Evenson (1996, p. 167) also notes: Overcrowded dwellings were common, with every inch of space, in cluding verandas, rented out as sleeping space for labourers. Addi tional storeys were hastily added to existing houses, often resulting in collapse of buildings. While the slums of Madras and Calcutta took the form of village-like clusters of huts, the multistorey tenement, or chawl, became typical of Bombay. A single chawl five to seven storeys high might have from five hundred to one thousand inhabitants crowded into dark and unventilated rooms. Housing conditions were worse than chawls for lower working class poor and the “backward” and “depressed” classes. Burnett Hurst (1925) notes three types of “sheds” that accommodated the poor. The first type was the zavli shed, which was made of dry coconut or date leaves where carters and mill workers lived along with 8 Nearly 75.6 percent o f the single room tenements were situated in Byculla, Parel and Mahim, which were pre dominantly cotton mill areas. 99 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. domestic animals. The second type included sheds built of corrugated iron sheets and other waste materials; and the third type consisted of small huts, made of mud walls and palmyra leaves’ roof. Slum dwellings and pavement dwellings of post Independ ence period reflected similar building characteristics. Overcrowding, unsanitary living conditions, poor drainage, infestation, tightly built buildings with little or no light and ventilation became common in the housing among poor households. In 1901, about 36.6 percent of the population lived in single rooms occupied by 6 to 20 or more persons; the situation had not changed perceptibly by 1920 when about 36 percent lived in single rooms occupied by six or more persons. In mill districts, percentage of population living in single rooms was even more: 47.4 percent in 1901 and 42.6 percent in 1921 (Upadhyay, 1990, p. PE-94). The working class laborers of cotton mills “lived packed in chawls, usually with several families, ten to fifteen people, domiciled in one room, their possessions a few deal boxes, blan kets, mats, brass or earth pots, tin lamps, a stool and a trunk for clothing” (Klein, 1986, p. 733). In areas with such worse housing conditions, epidemic diseases like plague and cholera took their toll. These areas were among the worst affected during the 1896 plague. Between 1896 andl910, over 172,500 died of plague and over 10,000 died of cholera (Klein, 1986, p. 740, 743). Infant mortality was also very high in such areas (Ramasubban and Crook, 1996, p. 155). Bombay Improvement Trust was set up in 1898 following the plague in re sponse to the deteriorating housing conditions. The Trust carried out extensive demo litions of densely constructed sheds and hutments. Although the Trust’s objectives included re-housing the displaced persons in n ew ly constructed ch aw ls, disproportion ately more people were displaced than those who were re-housed. The Trust’s con struction activities were also hampered due to the real estate market—economic rents 100 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. that the Trust could charge (to cover its acquisition, construction, and maintenance costs) were higher than the amount that an average millworker could afford (Chanda- varkar, 1994, p. 176-177).9 In addition, the newly constructed chawls were not attrac tive for workers due to their distant location from cotton mills. After the post World War I economic boom subsided, real estate and housing prices also dropped from their previous peaks. Yet, labor housing conditions hardly improved. Problems of overcrowding and unsanitary living conditions persisted espe cially in mill districts, where the majority of labor population lived. Chandavarkar (1998, p. 103) notes: A survey conducted in 1921 discovered that 27 percent of the popula tion in Parel and 33 percent in Umerkhadi lived in rooms containing six or more persons. Another investigation conducted in the mid-1930s found over 35 per cent of families of ‘untouchable’ workers sharing a single room with at least one other family, while over 63 per cent lived in a single room. ‘Every sixth person in the city’, it was reported in 1939, ‘lives in conditions which are prohibited even by the existing an tiquated law. The Report o f the Rent Enquiry Committee, 1939 also noted high degree of over crowding and worsening housing conditions. Another wave of migration to the city since World War II further deteriorated the housing conditions. Mumbai City’s popu lation surged from nearly 1.5 million in 1941 to 2.3 million in 1945 (Government of Bombay, 1946). Suburban growth towards the north along the railway lines also in tensified with the burgeoning population. Suburban population grew from over 300,000 in 1941 to over 465,000 in 1946. The rapid increase in population was, how ever, not accompanied by an equivalent growth in housing stock and social services; consequently, overcrowding increased. According to estimates of Report of the Hous 9 Although chawls were let out by the Trust at sub-market rate rents, their vacancy rates were still high (e.g. in Worli, 78 o f the 121 chawls constructed were vacant) (Reed and Sheppard, 1928, p. 368). 101 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ing Panel in 1946, overall overcrowding of Mumbai City increased from 4.01 persons per room in 1931 to 7.01 persons per room in 1945. 4.1.3. Rise of Housing Cooperatives for Middle Class Housing, 1913-1947 I argue that housing cooperatives grew in the context of deteriorating housing conditions in Mumbai essentially as a mechanism for providing decent housing for the middle class. Housing cooperatives were “modem” organizations that were intro duced by the British under the Cooperative Societies Act in 1904. While the Act originally envisaged cooperatives for rural credit, its amendment in 1912 allowed co operatives to expand their activities and to include other objectives including housing. Following the Amendment, the Bombay Cooperative Housing Association (BCHA) was formed in 1913, which propagated cooperatives for achieving coordination among neighbors for better housing. It “carried on intensive propaganda by holding weekly meetings for discussion, by organising public lectures, and by publishing leaflets on the various aspects of the housing problem, and the possibilities of meeting it by re course to co-operation” (Talmaki, 1931, p. 117). Although the BCHA had a short life span (its functions were taken over by the Provincial Institute in 1919 (Hough, 1932, p. 192)), it provided the initial impetus for growth of housing cooperatives in Mumbai. The BCHA comprised of eminent leaders who spearheaded cooperative movement in housing. For example, J.P. Orr, the Chairman of the Bombay Improve ment Tmst, was also the first chairman of the BCHA. BCHA became a medium for him to further his campaign for improving housing conditions through physical im provements (for betterment of light and ventilation) and sanitary improvements (for better sewage disposal). Orr used BCHA to emphasize the need for cooperation be tween neighbors (landowners, builders) in order to achieve mutual gains while attain ing better housing conditions. In one of his lectures, Orr (1915, p.4) emphasized: 102 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Co-operation has solved difficulties in many fields of enterprise and we may well consider how far it may be used to counteract the intensifica tion and expansion of slum areas in Bombay; and my main object this evening is to suggest how neighbors can assist one another to make the most of any holdings they may wish to build on by arranging to share one another’s open space so that each may have far more open space conveniently situated around his house then he would have if he were to provide the whole of the open space on his own land and could not prevent his neighbours from building up to the edge of it. The chief proponent of housing cooperatives was S. S. Talmaki, the founder of BCHA, who was “a pioneer first in co-operative banking and then in the building and letting of blocks of tenements on a ‘co-partnership’ basis” (Catanach, 1970, p. 132). Talmaki emphasized joint-ownership as being superior to individual ownership, which became central to housing cooperatives in Mumbai. Two types of housing coopera tives, both based on tenure, emerged in Mumbai: (i) tenant co-partnership system and (ii) tenant ownership system. Under the first system, the cooperative owned both land and buildings; under the second, the cooperative owned the land, and members leased housing units from the cooperative. Both systems were adapted as modem organiza tional forms to foster cooperation for a more orderly housing development. Talmaki championed tenant co-partnership system and offered several ration ales for why it was more suitable for Mumbai. First, joint-ownership under the co partnership system would entail cooperation, which was essential to co-ordinate or derly building construction between parcels of land for basic services and natural light and ventilation; such cooperation could not be assured under individual ownership.1 0 Second, co-partnership tenancy offered the twin advantages of both ownership and rental housing, while minimizing attendant risks in both. While perpetual owner ship was assured by co-partnership tenure, it did not tie members to their housing. In 1 0 This argument should also be seen in the context o f planning regulations in Mumbai, which were yet in their infancy. 103 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Mumbai’s context, this element gained significance due to population flux (especially due to labor having its roots elsewhere in the villages, where it migrated). Further, members in the co-partnership system did not have to be concerned about the uncer tainty of rent increases by landlords, which had become increasingly rampant in the aftermath of World War I. Talmaki (1931, p. 198-199) explained: Tenants enjoy fixity of rent and tenure; and no member is tied down to the house, but is at liberty to quit it after giving due notice to the soci ety. A new tenant coming in his place pays his share of the capital, which may be used for repaying that of the outgoing member. Tenants have a choice in the selection of their neighbours, the latter not being forced upon them as in the case of residents in buildings owned by landlords. Thus the tenants practically enjoy all the benefits obtainable by owners, without the risks attached to individual ownership. Third, co-partnership system made an explicit provision for maintenance and upkeep by budgeting for sinking funds and maintenance costs. Such assurance of maintenance could potentially limit deterioration of buildings and keep them from de generating into slums. This was unlike rental housing, where landlords spent little on maintenance in order to realize higher profit margins in the context of Mumbai’s high housing demand conditions. Talmaki (1931, p. 201-202) further explained: In the case of houses of individual ownership, repairs must depend not only on the condition of the owner’s purse, but also on his willingness to meet the necessity as it arises. For the one guiding principle with most owners is to get the utmost from their buildings and to spend as little as possible on them by way of repairs or otherwise. This is one of the main causes why most properties rapidly degenerate into slums. Housing cooperatives in Mumbai were characteristically different from those in Chennai. In Chennai, cooperatives were used predominantly as instruments of fi nance and members constructed housing on land owned by them; hence, coordination between members was not imperative. In Mumbai, cooperatives were not formed as instruments of finance only—they were formed on a communal basis. Housing units 104 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. were developed in close proximity on land acquired by the cooperative, a part of which was earmarked for common facilities (children’s playground, community hall, and so on). Each member leased a house from the cooperative, and paid monthly dues towards amortization of purchase price. Strict provisions were made against sub letting or transfer. Thus, tenant ownership cooperatives in Mumbai were spatially co hesive and cooperation could be achieved since land was collectively owned. Collec tive tenure gained popularity in Mumbai, and it received little resistance in the context of the city’s modem flats and tenements (where common spaces and facilities were indivisible). This was unlike Chennai, where collective tenure played little role in the context of its individual plotted housing. In the context of Mumbai’s steep housing market, cooperatives hardly helped the working class laborers. Two rationales were offered for why cooperatives did not aid in the housing process of lower income laborers. First, cooperatives raised capital from their members or borrowed from other cooperative banks. Laborers could not afford to pay the substantial amount of initial capital. Second, Otto Rothfeld, an influ ential Registrar in Bombay province, observed that forming cooperatives of laborers, although desirable, was inherently problematic since they were mostly single, unedu cated men who migrated in and out of the city: The most desirable object for Co-operative Housing would certainly be the relief of those classes by the provision of cooperative settlements with cheap but healthy tenements suited to working men and their fami lies. So far nothing has been done in this direction. The mill-hands themselves are uneducated and are to a large extent a floating popula tion, consisting in many cases of men alone without their families. To expect them at this stage to assimilate the principles of co-operation or apply them with any unity and consistency seems almost impossible.1 1 1 1 Quoted in Mukherji (1923, pp. 308-309). 105 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Consequently, housing cooperatives were principally directed towards the middle class. Cooperatives had an aura of exclusivity and were formed along caste lines. The Saraswat Cooperative Housing Society started by Talmaki in 1915 is a good illustration of how the traditional (caste) and the modem (housing development) interplayed in the cooperative housing system. Saraswat Brahmins were traditionally an entrepreneurial group and were generally well to do; their homogeneity was impor tant for maintaining their group culture (e.g. religious festivities). The cooperative was financed by Shamrao Vithal Co-operative Bank, which belonged to the same community. Bombay City Improvement Tmst provided land on a long term lease at reduced rates. Modernization came in the form of flats in three storied buildings; five such buildings were constructed over two years. Each flat was self-contained, with one or more bedrooms, kitchen and internal toilet facilities; separate accommodations were also provided for servants. Common facilities included a meeting hall along with library and reading room that were maintained by the cooperative. While mem bers enjoyed modem housing facilities, they also maintained their traditional group identity through the cooperative. Bombay Catholic Cooperative Housing Society is another illustration of a similar Tenant Co-partnership cooperative formed along religious lines for Roman Catholics. However, this cooperative was on a larger scale consisting of 400 members that established a cooperative colony on 45 acres of land acquired by it near Santa Cruz in suburban Mumbai. The cooperative developed land by providing roads, drainage and other public utilities. Unlike the group housing built by Saraswat Coop erative H ousing S ociety, houses w ere m eant for one or tw o fam ilies w ith individual garden areas. The housing scheme included other amenities like a school, children’s playground, church, community hall, cooperative store, dispensary, dairy and poultry 106 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. farm (Mukherji, 1923, p. 299). The entire development was owned collectively by the cooperative and each member paid rent. The cooperative reserved the first right to buy back houses, and it placed restrictions on members’ ability to transfer occupancy rights (for example, members could sell their occupancy rights only to Roman Catho lics). Activities of housing cooperatives grew steadily since the second decade of twentieth century. Talmaki (1931, p. 121) reports that “two societies were registered in 1916, five in 1918, six in 1919, and thereafter societies sprang up not only in Bom bay City and its suburbs, but also in the districts.” Increase in the activities of coop eratives was also prompted by two institutional changes. Both changes were made possible by administrative reforms of 1919, which transferred oversight of coopera tives from central control of Government of India to provincial control. The first insti tutional aspect was that Government of Bombay started to provide government loans to cooperatives since 1922; such government funds were not available when coopera tives were under central control. Housing cooperatives could now obtain loans at cheaper rates from the state. Second, provincial control allowed Government of Bom bay to enact provincial legislation related to cooperatives. Consequently, the Bombay Cooperative Societies Act was passed in 1925. The Act explicitly included housing under the ambit of cooperatives, defining the broad objective of housing cooperatives as provision of dwelling houses. However, the Act did not did not differentiate be tween tenant co-partnership and tenant ownership cooperatives. By 1930, there were 23 housing cooperatives of both tenant co-partnership and tenant ownership types in M um bai and its suburbs (Talmaki, 1931, p. 121). U nlike Mumbai, cooperatives of the tenant co-partnership type were not popular in other dis tricts of the Bombay province. In 1943-44, there were 114 cooperatives in the prov- 107 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ince, of which 101 were of tenant-ownership type, 10 were of tenant co-partnership type, and the rest were of the mixed type (Government of India, 1951, p. 120). The tenant ownership and mixed type cooperatives were primarily in Mumbai City. Cooperatives’ activities slowed after 1930s due to several reasons. First, eco nomic decline coupled with Depression and World War II had dulled the housing market in general. Volume of housing construction reduced considerably since its peak in the 1920s. Second, cooperative housing was limited to middle and higher in come groups, in which market segment, rents and housing prices had already de pressed since the mid 1920s. Rent control measures had also been lifted in 1929. Co operatives were affected in two ways. First, Government of Bombay discontinued lending to cooperatives since 1933. Second, cooperatives did not quickly adapt to the changing housing market; consequently, their rents were higher than prevalent market rate rents. Reserve Bank of India (1950, p. 44) noted: On account of a lull in trade and industry during the period of depres sion, rents came down in the city of Bombay with a startling and alarm ing speed. The [cooperative] societies were, however, charging the agreed schedules and this made them unpopular. Flats were vacated in large numbers and this had an adverse effect on societies based on co partnership system. The demand for cooperatives again rose following World War II, when hous ing demand increased considerably with greater population influx. Between 1941-42 and 1945-46, the housing cooperatives increased from 105 to 136. Yet, the increase in did not signify an equivalent rise in their activities. Although the total working capital of cooperatives recorded a modest increase from 12 million rupees to about 12.8 mil lion rupees, the working capital had dipped during the intervening years (Reserve Bank of India, 1950, p. 43). Further, building industry had been in general slump fol lowing wartime controls on building materials and labor. 108 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.2. Post-Independence Period In the post-Independence period, housing cooperatives flourished in Bombay state in general, and Mumbai in particular. Cooperatives almost tripled in the first two years in Bombay state, from 167 in 1946-47 to 480 in 1948-49, and their membership rose from 10,369 to 24,838 (Reserve Bank of India, 1950, p. 43). The total working capital of cooperatives rose from 14.7 million rupees to over 32.5 million rupees dur ing these two years. Construction activities of housing cooperatives, however, re mained stagnant at around 3,300 housing units. The activities of housing cooperatives grew enormously in the state since then. By 1957-58, there were 2,006 primary hous ing cooperatives in the state, with 90,101 members. The total working capital of hous ing cooperatives was 164.16 million rupees, and they contributed 24,024 units to the housing stock. Table 4.4. Growth of Primary Housing Cooperatives in Maharashtra, 1960-61 to 1999-2000 1960-61 I 1970-71 1980-81 1990-91 1999-00 Number of Cooperatives Membership Working Capital (million rupees, unadjusted) 1,495 76,500 250 7,425 290,900 1,314 15,640 531,700 3,508 33,913 1,131,500 ................ 9,792 50,284 1,819,200 46,440 Houses and Tenements Con structed during the year 3,200 5,400 10,700 31,700 110,500 Source: Government of Maharashtra (2001), Co-operative Movement At a Glance in Maharashtra. The activities of housing cooperatives in Maharashtra grew consistently after it was bifurcated from Bombay state in 1960. Table 4.4 shows the growth in activities of housing cooperatives in Maharashtra between 1960-61 and 1999-2000. As the table shows, the number of primary housing cooperatives grew sharply from 1,495 to 50,284, and their membership grew from 76,500 to over 1.8 million. The growth in financial activities is proxied by the growth in working capital, which grew from 250 109 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. million rupees to 46,440 million rupees. Further, the contribution of cooperatives to the housing stock rose from 3,200 units to 110,500 units. In parallel with the growth of housing cooperatives in the state, housing coop eratives grew phenomenally in Mumbai too. In 1948-49, there were 156 primary co operatives (61 Tenant Ownership Societies, 70 Tenant Co-partnership Societies, and 25 mixed type) with 12,027 members in Mumbai City and the suburbs (Naik, 1953, p. 205). Cooperatives became active in both the formal and informal sectors since then. By 1995-96, there were 14,634 registered housing cooperatives, with about 738,085 members in Greater Mumbai. The total working capital of cooperatives was over 6,749 million rupees, and they had constructed 15,232 tenements.1 2 The role of housing cooperatives evolved with the changes in national and state level housing policies. Similar to the national level policies which emphasized high level of public sector participation in housing until the 1970s, housing policies in Mumbai also emphasized public sector involvement. This emphasis is reflected in Figure 4.1, which indicates the contribution of various sectors to housing stock be tween 1956-57 and 1965-66. The state government intervened directly in the housing process during this period. Indeed, until 1959-60, public sector agencies (comprising Bombay Housing Board; Bombay Municipal Corporation and Bombay Electric Sup ply and Transport (BEST); state and central government agencies; statutory bodies and corporations) contributed over 50 percent of the incremental housing stock. Although housing provided by public agencies reduced since then, public housing still formed a significant part. The share of private sector and cooperatives to the housing stock rose to over 50 percent since 1960-61, as the private sector started to adapt cooperatives for collective housing management. 1 2 Data obtained personally from files o f O ffice o f the Registrar (Greater Mumbai) on 10 June, 1997. 110 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4.2 gives the sector-wise contribution to housing stock between 1973- 74 and 1981-82. As the figure shows, cooperatives provided between 20 to 32 percent of the incremental housing stock in this period (except 1980-81, when it fell to 13 per cent). Private sector housing activities increased during this period, contributing be tween 40 to 68 percent of the incremental housing stock; the share of public housing agency (BHADB) fell to 20 percent or less. Figure 4.3 provides the sector-wise contribution to housing stock between 1982-83 and 1990-91. As the figure shows, cooperatives provided between 15 to 34 percent of the incremental housing stock in this period (except 1988-89, when it fell to 9 percent). Following national trends of increasing private sector participation, private sector housing in Mumbai also maintained the leading edge, contributing between 53 to 79 percent of the incremental housing stock. The share of public housing agency (BHADB) reduced to below 15 percent. Thus, cooperatives have played an increasingly active role in the housing mar ket in Mumbai, contributing 15 to 34 percent of the incremental housing stock on most years between 1973-74 and 1990-91. The share would be more if the formation of cooperatives by the private sector and public agencies are also taken into account. Such formal cooperatives are Tenure cooperatives, oriented towards collective man agement. Formal cooperatives have also been used by public housing agencies for transferring ownership in old buildings in central city areas and for slum redevelop ment schemes. Informal cooperatives are mostly formed by non-government organi zations (NGOs) among slum and pavement dwellers to mobilize them for collective action for housing rights and to define m icro finance groups. I l l R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 4.1. Sector-wise Contribution to Housing Stock in Greater Mumbai, 1956-57 to 1965-66 80% 35,000 - 30.000 - 25,000 - 20,000 a 40% 15,000 10,000 - 5.000 ■ i i — i— i — r 1956-57 1957-58 1958-59 1959-60 1960-61 1961-62 1962-63 1963-64 1964-65 1965-66 Y ea r I P u b lic a g e n c ie s i i P rivate S ector and C o o p eratives •T o ta l U n its Source: Bombay Metropolitan Regional Development Authority (BMRDA), 1995, p. 303. to Total U n its Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 4.2. Sector-wise Contribution to Housing Stock in Greater Mumbai, 1973-74 to 1981-82 70% 60% 50% 40% 30% 20% 10% I Private Sector ■ ■ C o o p e r a tiv e s I, _ .3 B H A D B I I Em ployers ■ Total units 35,000 30,000 2 5,000 20,000 15,000 10,000 5,000 1973-74 1974-75 1975-76 1976-77 1977-78 1978-79 1979-80 1980-81 1981-82 Year Source: BMRDA, 1995, p. 304. U J Total Units Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 4.3. Sector-wise Contribution to Housing Stock in Greater Mumbai, 1982-83 to 1990-91 80% a 40% L 1982-83 1983-84 1984-85 1985-86 1986-87 1987-8 Year 3 5000 - 30000 -- 2 5000 20000 15000 -- 10000 -- 5000 1988-89 1989-90 1990-91 I B H A D B Em ployer provided L I Private sector I J Cooperatives ■ Total units Source: BMRDA, 1995, p. 305. Total Units 4.3. Mumbai’s Institutional Context, 1947-2000 Why did housing cooperatives grow phenomenally in Mumbai during the post- Independence period? I argue that in the context of Mumbai’s particular local histori cal conditions of urban growth and demography, both the local housing market and the institutional framework of Mumbai were conducive to the growth of formal housing cooperatives. In terms of the local housing market, the demand for collective organ izational vehicles like cooperatives grew with the growth of multi-family dwelling units. In terms of institutional framework, cooperative policies were relatively lib eral—state intervention was minimal, so that cooperatives had greater autonomy. In this context, an elaborate cooperative institutional structure supporting cooperatives evolved. The institutional structure reduced the transaction costs of formation and functioning of housing cooperatives. Indeed, the procedures for forming cooperatives became so facile that private developers could form cooperatives of owners almost overnight. The ease of forming groups through cooperatives facilitated the use of these organizations for defining informal collectives among slum and pavement dwellers. 4.3.1. Mumbai’s Urban Growth and Demography While pre-Independence urban growth was mostly limited to Mumbai City, post-Independence urbanization rapidly expanded towards the suburbs in the north of the City. Greater Mumbai was formed in 1950 as such suburban developments were brought under one municipal corporation along with Mumbai City. Mumbai City cov ers a narrow linear strip of about 69 square kilometers, and Greater Mumbai covers about 436 square kilometers. Greater Mumbai comprises three broad administrative divisions: Mumbai City, Suburbs, and Extended Suburbs. Until 1968, most of the 115 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. urban growth was limited to Greater Mumbai’s municipal limits, after which urban growth spilled over to the Metropolitan Region. The Mumbai Metropolitan Region, which was initially defined in 1967, covers a much larger area of 4,330 square kilome ters. The urban built-up area formed about 12 percent of the Metropolitan Region in 1991, extending towards north and east (Thane, Kalyan, Mira-Bhayander, New Bom bay and Vasai-Virar). During the post-Independence period, Mumbai’s economy diversified with growth of several different activities related to port, manufacturing, government, fi nancial institutions, and trade and services. Although Mumbai went through economic ups and downs (e.g. due to recession in 1980s, and decline of cotton mills), its econ omy remained vibrant, surpassing the other three major Indian cities of Delhi, Cal cutta, and Chennai. Nearly 40 percent of Maharashtra’s Domestic Product originates in Mumbai (BMRDA, p. 97). With the decline of Calcutta’s port, Mumbai’s port be came the largest in India, handling the most amount of cargo. The construction of Jawaharlal Nehru Port Trust with modem facilities during the 1990s also provided a boost to Mumbai’s economy. Until the 1970s, Mumbai’s economy was dominated by traditional manufacturing comprising chemical, engineering, automobile, textile and railway industries. Cotton industry, which was the mainstay of Mumbai’s economy during its pre-Independence growth, started to decline in the mid 1970s due to labor problems and technological obsolescence. Many of the cotton mills became prime candidates for urban renewal as they occupied prime locations in Mumbai City, but remained idle for several years.1 3 In the 1990s, Mumbai’s economy transformed to a 1 3 Redevelopment o f Textile M ills land is a highly contentious issue in Mumbai. In 1996, the state government appointed a study group under Charles Correa, a leading architect, for the redevelopment o f the m ill areas. The study group found 58 cotton textile m ills, o f which 26 were sick units under public ownership; the rest 32 were under private ownership, and did not even allow the study group to inspect the properties (Editorial, Econom ic and P olitical Weekly, June 3, 2000). 116 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. service economy, centering on banking, insurance, and other financial services. Mumbai became India’s most global city, with more foreign trade, cargo movement and transnational firms.1 4 Table 4.5 shows the employment distribution in primary, secondary, and terti ary sectors in Greater Mumbai between 1961 and 1991. The secondary sector (of which manufacturing is a major component), reduced marginally from 43.5 percent in 1961 to 41.0 percent. Tertiary sector (which includes services) formed the major em ployment base, increasing from 54.6 percent to 58.1 percent. Mumbai is the financial capital for India, housing many of the headquarters of major national and international banks. It is also the major Indian city for movie productions. Since Mumbai is the capital of Maharashtra, it houses major state administrative functions. Over one-third of the workforce was employed in the public sector in 1990. While Mumbai’s formal sector economy has grown since Independence, informal sector economy has also been traditionally significant for the city. According to BMRDA (1995, p. 112) esti mates, the share of informal sector employment was 49, 55.4, and 65.6 percent respec tively in 1971, 1981, and 1991. Table 4.5. Employment in Greater Mumbai by Sector, 1961-1991. 1961 1971 1981 1991 Primary 1.9% 1.3% 1.4% 0.9% Secondary 43.5% 45 3% 44.^% 41.0% Tertiary 54.6% 53.4% 53.9% 58.1% Source: BMRDA (1995, p. 129). Note: (i) Primary, secondary, and tertiary sectors are census definitions. Primary sector includes agriculture and related activities, and mining. Secondary sector includes manufacturing and construction. Tertiary sector includes trade and commerce, transportation and communication, and other ser vices. 1 4 In 1998, there were nearly 340 transnational companies in Mumbai, as compared to about 135 in Delhi, about 120 in Chennai, and about 35 in Bangalore (Nijman, 2000, p. 578). 117 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. With the growth of Mumbai’s economy, the city’s population surged in the World War II period. Labor migration contributed a major part to the population growth initially. New defense based industrial establishments and commercial ven tures provided employment to these immigrants from rural areas; further a substantial proportion of those displaced from Pakistan due to Partition in 1947 also settled in Delhi or Mumbai. Most of the migrants were also male laborers. Zachariah (1966, p. 382-383) notes in his study of migration to Mumbai: About 43 percent of the migrants were between the ages of 20 and 35, and over 80 percent were in the working ages 15-59. ... The sex com position of migrants favored males, with the excess of males exceeding 800 for every 1,000 females. The ratio of males per 1,000 females was almost normal (1,045) at ages 0-4 but increased to 2,307 at ages 40-44. Migration contributed less to the population growth in the later decades as new indus tries were more capital intensive and therefore created fewer jobs (Zachariah, 1968).1 5 While 80 percent of population growth was due to migration during 1941-51, the share reduced to 52 percent and 49 percent during 1951-61 and 1961-71 respectively (Table 4.6). Table 4.6. Contribution of migration to population growth of Mumbai, 1941- 1981 Year Net Growth (’000s) Natural Increase (’000s) Net migration (’000s) % Natural Increase % Net Mi gration 1941-1951 1193 243 950 20% 80% 1951-1961 1158 558 600 48% 52% 1961-1971 1819 933 886 51% 49% 1971-1981 2230 1203 1027 54% 46% Source: Government of Maharashtra (1981). 1 5 Industrial policies follow ing Independence emphasized large-scale industries, in which several large scale auto motive, heavy-duty, and chemical industries were established in Mumbai. The later industrial location policies, however, encouraged decentralization o f industries to satellite towns and industrial centers away from metropolitan areas; thus the adjoining areas o f Greater Mumbai in Thane and Kalyan became industrial centers and gradually coalesced with Greater Mumbai into the metropolitan region. M echanization o f cotton m ills also freed workers for other industries. 118 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 4.7 indicates the population growth in Greater Mumbai between 1941 and 2001. Greater Mumbai’s population grew at an annual average rate of 6.6 percent between 1941 and 1951, principally due to labor migration, who arrived in the city as a result of wartime inflation in economic opportunities. Although overall population growth rate decreased since then to 1.5 percent during 1991-2001 period, absolute growth in population continued unabated. Mumbai City’s population growth peaked in 1981, after which its population started to decrease. Its annual population growth rate steadily decreased from 5.6 per cent during 1941-51 to -0.6 percent during 1991-2001. Population deconcentrated from Mumbai City for a combination of various reasons. Industrial location policies since the early 1970s sought to relocate heavy industries out of the Bombay City, so that it housed mainly corporate head offices, banks, and government offices. With improvement of transportation facilities, especially mass rail transit, peripheral subur ban development towards the northeast accelerated. Consequently, the population of suburbs grew at an annual rate of over 10 percent until 1971, after which it began to reduce. The extended suburbs grew at an annual rate of 12.1% and 11.3% between 1951-61 and 1961-71 respectively, after which it began to reduce. The growing popu lation of Mumbai was thus absorbed by the suburbs and extended suburbs. While 83 percent of population lived in Mumbai City in 1941, its population share reduced steadily since then to 26 percent in 2001. The percent share of population of suburbs increased from 11 percent to 43 percent, and that of extended suburbs increased from 6 percent to 31 percent during the same period. The percent share of population of Greater Mumbai in the Mumbai Metropolitan region steadily decreased from 76.8 per cent in 1971 to 61.8 percent (Table 4.8). 119 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 4.7. Population Growth of Greater Mumbai, 1941-2001 Population Population Growth Rate (Annual) Population Share City Suburbs Extended Suburbs Total City Suburb Ext. Suburb Total City Suburb Ext. Suburb 1941 1,489,883 205,285 106,188 1,801,356 83% 11% 6% 1951 2,329,020 510,250 155,174 2,994,444 5.6% 14.9% 4.6% 6.6% 78% 17% 5% 1961 2,771,933 1,036,585 343,538 4,152,056 1.9% 10.3% 12.1% 3.9% 67% 25% 8% 1971 3,070,378 2,166,864 733,333 5,970,575 1.1% 10.9% 11.3% 4.4% 51% 36% 12% 1981 3,285,040 3,522,517 1,435,848 8,243,405 0.7% 6.3% 9.6% 3.8% 40% 43% 17% 1991 3,174,889 4,167,681 2,583,321 9,925,891 -0.3% 1.8% 8.0% 2.0% 32% 42% 26% 2001 3,000,000 4,930,000 3,500,000 11,430,000 -0.6% 1.8% 3.5% 1.5% 26% 43% 31% Source'. BMRDA ( 995, p. 92). Figures for 2001 are estimates by BMRDA N > o Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 4.8. Population Growth of Mumbai Metropolitan Region, 1971-2001 Population Population Share in the Region 1971 1981 1991 2001 1971 1981 1991 2001 Greater Mumbai 5,970,575 8,243,405 9,925,891 11,430,000 76.8% 74.4% 68.3% 61.8% Western Region 232,659 332,421 595,868 1,063,092 3.0% 3.0% 4.1% 5.8% North-East Region 1,054,207 1,799,628 2,921,172 4,216,704 13.6% 16.2% 20.1% 22.8% New Bombay Region 160,669 257,913 548,476 1,166,558 2.1% 2.3% 3.8% 6.3% Neral Karjat Region 112,516 132,385 166,021 180,471 1.4% 1.2% 1.1% 1.0% Panvel-Uran Region 108,679 139,333 180,828 227,193 1.4% 1.3% 1.2% 1.2% Pen Region 56,624 73,835 88,297 92,444 0.7% 0.7% 0.6% 0.5% Alibag Region 81,602 99,109 107,811 110,097 1.0% 0.9% 0.7% 0.6% Total Metropolitan Region 7,777,531 11,078,029 14,534,364 18,486,559 100% 100% 100% 100% Source: BMRDA (1995, p. 92). Figures for 2001 are estimates by BMRDA. 4.3.2. Mumbai’s Local Housing Market In spite of Mumbai’s economic growth, Mumbai’s housing market did not sig nificantly improve during the post-Independence period. Housing supply did not keep pace with the population increase. Institutional constraints such as post-World War II rent control policies and urban land ceiling adversely affected housing supply. The reduction in housing supply and increasing population put more pressure on the hous ing market, so that Mumbai’s housing deficit ballooned from 130,652 units in 1961 to 318,000 in 1991. One effect of Mumbai’s superior economy as compared to other Indian cities is that Mumbai’s real estate prices have been historically higher than other cities.1 6 Ac cording to Nijman (2000, p. 576), Mumbai’s peak values of prime office space in 1995-96 were about “one-third above those in Delhi and about six times as high in Chennai and Bangalore.” Indeed, Mumbai was reported to have the world’s highest commercial real estate prices in 1994-95.1 7 Residential property prices in Mumbai have been closely connected with commercial property prices, and have also been higher than other Indian cities. Another feature of Mumbai’s peculiar urban demography is its male predomi nance, especially among migrant laborers. Although the female-male ratio increased from 575:1000 in 1951 to 820:1000 in 1991 (Table 4.9), the ratio is lower than that of Chennai (the ratio remained even lower in Mumbai City). Single room tenements and 1 6 According to Government o f India (1988), land price in Mumbai in 1983 ranged from 150 to 15,000 rupees per square meter; in Delhi, it ranged from 198 to 4500 rupees per square meter, and in Chennai, it ranged from 179 to 2243 rupees per square meter. Aiyar (1997) also reports that although real estate prices slumped in Indian cities since 1994, prices in Mumbai were still much higher than that in Delhi and Chennai. 1 7 The Economist, February 24, 1996, p 112; and August 31, 1996, p. 84 122 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. chawls continued to persist as the predominant mode of accommodation for laborers, which were also adapted as means of housing low-income households. Table 4.9. Sex Ratio in Mumbai, 1951-1991 Year Sex Ratio (Females per 1000 males) Female Workers per 1000 male workers 1951 575 1961 625 95 1971 671 107 1981 735 125 1991a 820 172 Source: Unless otherwise indicated, Kosambi (1986) Note: Data (a) obtained from Deshpande (1996). The fundamental feature of Mumbai’s housing market due to the high demand conditions is that its residential density continued to be high in the post-Independence period. As Table 4.10 shows, number of persons per room hardly reduced since 1961; nearly 70 percent of the households lived in single room units historically. According to a report of the study group for Greater Mumbai (Government of Bombay, 1959, p. 2), 5 to 10 persons could be found in a room of 100 square feet. Inner city areas of Mumbai City faced much higher rates of overcrowding. These areas also have the ma jority of old, dilapidated buildings, many of which were constructed before 1940. Table 4.10. Residential Density of Greater Mumbai 193 l a 1961b 1971 1981 1991 Number of households (’000s) 803 1073 1580 2088 Number of rooms (’000s) 1092 1458 2235 2921 Average size of house hold 5.3 5.4 5.1 4.8 Number of persons/ room 4.01 3.7 4.0 3.7 3.4 Percent households with 1 room 74 72.3 77.4 68.9 72.9 Sources', (a) Data obtainec from Government of E iombay (1946, p. 11) (pertains to Mumbai City); (b) Data obtained from O’Hare, Abbott, and Barke (1998, p. 274). 123 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In the formal sector, the high residential density made single-family housing development a luxury for even the high-income households in Mumbai. In this con text, multi-family housing development in the form of flats and tenements in multi- storied buildings became the norm. The pre-Independence form of vertical housing development was thus reinforced during the post-Independence period. Multi-family housing in Mumbai is typically a five storied building, housing between 12 to 30 units.1 8 According to the national sample survey conducted in 1988-89, only 6.14 per cent of the households lived in independent houses; 30.92 percent lived in flats, and 51.7 percent lived in chawls.1 9 Multi-family housing created the demand for collec tive organizational forms for management of common properties and facilities. Poli cies for flat ownership since the 1960s, for example, emphasized formation of associa tions, societies, companies, or cooperatives for collective tenant management. Coop eratives gained more popularity than other forms of collective organizations partly due to historical precedence, and mainly due to the development of support institutional structures that lowered the transaction costs of formation and functioning of coopera tives. In the informal sector, slum population burgeoned during the post- Independence period. According to the Report o f the Study Group for Greater Bom bay (Government of Bombay, 1959, p. 102), nearly 13.4 percent of the population lived in slums in 1959 (the densities of some of which reached up to 1,175 persons per acre). The percentage of slum population increased since then to 30.8 percent in 1981 and 43.2 percent in 1991; by 2001, about 58.5 percent were estimated to live in slums 1 8 According to M umbai’s planning regulations, buildings over 5 stories require placement o f elevators. Private developers typically tend to econom ize on construction costs by limiting upto 5 stories. Buildings over 5 stories are usually high-incom e housing and are more recent phenomena (Tiwari, Parikh, and Parikh, 1999). 1 9 Sarvekshana, 1992, p. S-268. 124 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. (Ministry of Urban Affairs, 1996). Indeed, Dharavi, the largest slum in Asia is located in Mumbai. The post-Independence slums are generally of three types: (a) old, dilapi dated buildings;2 0 (b) semi-permanent structures (e.g. patra chawls), which could be authorized or unauthorized construction; and (c) zopadpattis or hutments (Desai, 1988, p. 69). Landlords have little incentive to renovate old, dilapidated buildings, due to rent control regulations, and there are routine reports of building collapses every year. The other two types of slums are usually located on illegally occupied land or mar ginal land (e.g., along railway lines, sidewalks, marshes, water mains) and have little or no services. With the increasing popularity of cooperatives in the formal sector, public housing agencies and non-governmental organizations started to use them as vehicles for achieving collective action in the informal sector. Tenant cooperatives were formed for mobilizing tenants for repairs of old and dilapidated buildings. In slums, cooperatives were formed for redevelopment and for securing tenure. Informal cooperatives were formed among pavement dwellers for microfmance schemes. 4.3.3. Mumbai’s Institutional Framework While Mumbai’s local housing market favored the development of coopera tives by creating a demand for collective organizations, the institutional framework also favored their development. I will highlight two legal frameworks that assisted in the development of cooperatives. The first is the framework of rent control policies, and the second is the framework of flat ownership regulations. While rent control policies favored the development of cooperatives indirectly, flat ownership regulations favored cooperatives in a rather direct manner. 20 Although “old, dilapidated buildings” are classified as slums since they may be below city’s norms o f inhabita tion, they are not slums in the sense that they are not constructed on illegally occupied land. Also, they are not strictly in the informal sector. 125 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. a. Rent Control Policies Rent control forms one of the core elements of the institutional framework af fecting housing market in Mumbai. Except for a ten-year period between 1929 and 1938, rent control has existed in one form or the other in Mumbai since 1918. The Bombay Rents, Hotel Rates, and Lodging House Rates Control Act (1947), which con solidated earlier wartime rent control laws, was intended as a temporary initiative to control rapidly rising rents. Under the law, existing rents were capped at 1940 levels with little or no annual increase; in newly constructed rental housing, courts were em powered to determine the standard rent. Tenants could be legally evicted only under exceptional circumstances when a landlord required the house for personal (or family) use or a tenant used the property for purposes other than that for which it was let; evic tion could not be carried out in order to rent to another tenant. The 1947 Act has been revised several times, extending its life and scope. Government instituted rent inquiry committees never seriously challenged the basis of rent control, although they acknowledged negative effects of the Act on housing mar ket. For example, it was extended in 1959 to legalize existing sub-tenancy and to pro tect such sub-tenants against eviction. The Rent Enquiry Committee constituted in 1975 recommended converting the 1947 Act to a permanent measure (Government of Maharashtra, 1977, p. 82). Legislative changes in 1986, however, provided some re prieve for landlords. The changes provided for increase in rents for different types of repairs and renovations, besides allowing tenants to make rent-deductible expenses for structural preservations and improved services in case of default by landlords. New buildings were given a five-year holiday, after which rent control came into effect. Rent control legislations went fairly unchallenged by landlords until 1993. With the spurt in real estate market and economic liberalization since late 1980s, landlords mo- 126 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. bilized political support against the Act. Consequently, the new Maharashtra Rent Control Act covering the whole state was enacted in 1999, which accommodated land lords’ demands to some extent. The new Act allowed a 5 percent annual increase in rents, but stringent conditions against eviction remained. Rent control had several side effects on the housing market in Mumbai. First, landlords had little incentive to maintain and repair the buildings, leading to neglect and dilapidation of older buildings in central city areas. Second, in spite of official sanctions against premium and additional charges other than rent, the system of pug- dee again proliferated. Third, rent control distorted property taxes—taxes were based on Annual Rateable Value, which in turn was based on rents determined by rent con trol legislation. Taxes in central, more expensive areas were less than those in subur ban areas. Fourth, rent control constrained the rental housing market. While it pro vided a disincentive for new rental housing construction, landlords were also unwill ing to rent their property. In 1998, 40,000 apartments were estimated to be sitting 0 1 empty due to rent control laws (Nijman, 2000, p. 580). Rent control measures gave rise to other tenure forms such as “leave and license” in Mumbai, which were essen tially meant for letting for temporary use (typically not exceeding 11 months), so that it did not fall within the scope of rent control. Rent control indirectly facilitated growth of cooperatives by creating incentives for ownership.2 2 First, cooperatives offered an alternative to the traditional landlord- tenant relationship—they internalized such relationship through joint ownership. Cer tain provisions of rent control were not applicable to housing cooperatives since mem 2 1 See National Building Organisation (1968) for relationship between rent control and housing production in Mumbai. 22 According to sample survey conducted by Operations Research Group in 1987/88, about 62 percent were owners (quoted in Tiwari, Parikh, and Parikh, 1999, p. 1785). Ownership rates in Mumbai and Delhi, the two cities where rent controls are in effect, are higher than other major Indian cities. 127 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ber-cooperative relationships were different from landlord-tenant relations. Disputes that arose between members and cooperatives came under the purview of Maharashtra Cooperative Societies Act, rather than under that of rent control laws.2 3 The institu tional structure of separate Cooperative Courts (described later) provided an alterna tive mechanism for settling such disputes. Thus, the cooperative provided an organ izational alternative for ownership. Indeed, housing policies sought to utilize coopera tives as antidotes to the problem of dilapidation in older buildings created by rent con trol measures (for example, the Maharashtra Act No. 21, 1986, see Section 4.5.3). Second, since returns from rental housing were low,2 4 the formal private sector became oriented towards constructing housing for sale to owners. The Rent Enquiry Committee (Government of Maharashtra, 1977) for example, reported, “the position is that rental housing in private sector has virtually come to a halt, particularly in Bom bay. Almost all new residential flats are being built for sale” (p. 27). The private sec tor adapted cooperatives as a means of transferring ownership of flats in the face of flat ownership regulations. b. Flat Ownership Regulations While multi-family housing proliferated in Mumbai’s environment of high demand, it was also accompanied by several abuses in the sale and transfer of owner ship of flats. The Ownership Flats Enquiry Committee constituted in 1960 noted sev eral such abuses of builders taking advantage of eager purchasers, many of which re lated to transfer of ownership rights to individuals. While the Committee recom 23 Disputes with non-members in housing cooperatives (i.e. renters or license holders) could potentially fall within rent control regulations. However, the cases are too com plex for a simple delineation between what falls under the scope o f rent control legislations and what falls under cooperative legislations, 24 Returns from rental buildings typically ranged between 5 and 7 percent, while savings instruments such as Na tional Savings Certificates offered 10 percent returns. 128 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. mended procedural changes in the sale of flats, it also highlighted the need for a col lective organization of flat purchasers to resolve disputes and other problems that could otherwise arise in multi-family housing due to individual ownership rights. The Committee argued that the land and building, of which the flat is a part, are jointly owned, while an individual’s claim to ownership is restricted to the flat. It proposed a collective organization to resolve problems arising due to the tension between individ ual and collective ownership: Most of the malpractices could be avoided if there is a single legal en tity, representative of all the flat owners and having effective control over them, for looking after the payment of taxes, water charges, lease rent, providing common services such as conservancy, lift, garden and compund, watchmen, etc. and attending to repairs and maintenance of the building. This peculiar position is generally due to the defective agreement entered into between the flat purchaser and the builder. (Government of Maharashtra, 1962, p. 23) The Report, however, did not recommend the cooperative as a sole means of collective organization—it considered the cooperative along with other collective legal entities (including associations and companies). It made several recommendations regarding the formation of cooperatives for collective ownership and maintenance in the private sector. Based on recommendations of the Ownership Flats Enquiry Committee, the Maharashtra Ownership Flats Act was passed in 1963 as a consequence of “sundry abuses, malpractices and difficulties relating to the promotion of the construction of, and the sale and management and transfer of flats taken on ownership basis.” The Act essentially formalized the need for a collective organization (cooperative or company) as a mechanism for builders to transfer ownership of housing to prospective buyers of flats. Under the Act, builders were required to form cooperatives or companies of prospective buyers before transferring the ownership of flat. 129 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Although the Act did not specify cooperatives as the only mode of collective organization for ownership and management, they became more popular than other organizational forms as the institutional structure supporting the functioning of coop eratives strengthened. First, Mumbai already had a history of Tenure Cooperatives since the pre-Independence period; hence, procedures relating to cooperatives were well established. Second, the 1925 legislation relating to cooperatives was revised in 1960 (Maharashtra Cooperative Societies Act, 1960), which accommodated the spe cialized needs of housing. The Act defined a housing cooperative as “a society, the object of which is to provide its members with open plots for housing, dwelling houses or flats; or if open plots, the dwelling houses or flats are already acquired, to provide its members common amenities and services” (my emphasis). Although ten ant co-partnership and tenant ownership types of housing cooperatives had long been in existence, the Maharashtra Cooperative Societies Rules, 1961 explicitly defined them, and recognized other types of cooperatives (such as House Mortgage Societies 9 S and House Construction Societies). In contrast to the Maharashtra Ownership Flats Act supporting collective own ership, another piece of legislation—the Maharashtra Apartment Ownership Act, 1970—provided legal support for individual ownership in condominiums. The Apartment Ownership Act was intended to “provide for the ownership of an individual apartment in a building and to make such apartment heritable and transferable prop erty.” The aim was to provide a marketable title that can be used as a security by apartment owners to raise loans. The Act did not require apartment owners to form cooperatives; rather, it stipulated the formation of Association of Apartment Owners, 25 Tenant Ownership Housing Society is defined as a cooperative where “land is held either on lease-hold or free hold basis by Societies and houses are owned or are to be owned by members”; Tenant Co-partnership Society is a society “which holds both land and buildings on lease-hold or free-hold and allot them to their members.” 130 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. of which every owner had to become a member. Although condominiums increased in popularity in other cities of Maharashtra, cooperatives remained the predominant mode of organization in multi-family housing in Mumbai. 4.4. Cooperative Institutional Structure in Mumbai As the last two sections highlight, the institutional framework of Mumbai was conducive to the growth of housing cooperatives. While the demand for collective organizational forms like cooperatives arose in response to the predominantly multi family housing market, rent control and flat ownership regulations also encouraged the formation of cooperatives. Yet, in the context of Mumbai’s institutional framework, other collective organizational forms like condominiums, associations, companies could also have grown. I argue that the cooperative institutional structure was instru mental in reinforcing the development of cooperatives. At the organizational level, the institutional structure provided support to cooperatives in terms of overcoming their internal weaknesses in the formation and functioning, and strengthened their in ternal capacity for collective action. At the institutional level, the structure reduced the transaction costs of formation and functioning of cooperatives. Similar to the institutional structure at the national level, there are three ele ments of the support structure in Mumbai. The first element is the structure for ad ministrative, legal, and other procedural support, which is rather well established. The structure of Registrar’s office is such that it does not interfere in the internal affairs of cooperatives. Besides this office, other institutional support systems for managing the internal affairs of cooperatives also emerged. The second element is the two-tiered cooperative institutional structure for finance, which provided an important source of finance when other housing finance mechanisms were yet undeveloped. The third 131 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. element is the institutional support for preferential access to land. These three ele ments provide various degrees of support for housing cooperatives, as noted below. 4.4.1. Administrative, legal, and other procedural support Cooperative policies in Maharashtra have been more liberal than other states in providing autonomy to cooperatives, while providing support to their functioning. This is reflected in the institutional support structure for administrative, legal, and other procedures in Mumbai, which is more robust than other cities. The first feature of the cooperative institutional structure is the nature of Registrar’s Office, which regulates the development of cooperatives. The Registrar’s Office has been histori cally such that it has had little interference in the internal decision making processes of cooperatives. This is partly due to the strength of cooperative intelligentsia in Bom bay State from the beginning, who vigorously fought for cooperatives’ autonomy. While Cooperative Societies Acts (1925, 1960 and amendments in 1986) provide broad powers to Registrar for registration, audit, liquidation, and other responsibilities, these powers are limited to overseeing that cooperatives follow the basic legal frame work. Chapter VII (Management of Societies) of the Cooperative Societies Act, 1960 explicitly states that “the final authority of every society shall vest in the general body of members in general meeting, summoned in such a manner as may be specified in the bye-laws.” Thus, the laws ensured that the Registrar’s Office does not interfere in the internal decision making processes of cooperatives. For example, although the Registrar has powers to convene general body meetings of cooperatives, s/he (includ ing any external nominee) has no right to vote in the meeting. This is unlike other states like Tamil Nadu, where the State government routinely interferes in the daily functioning of cooperatives through the Registrar’s Office and other government nominees (e.g. Special Officers). 132 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The second feature of cooperative institutional structure in Mumbai is the sepa ration between regulatory and legal functioning of the Registrar’s Office. Separate Cooperative Courts constituted in 1982 under the Maharashtra Act No. 18 have the powers to provide legally binding decisions. These Cooperative Courts, and not the Registrar’s Office, handle disputes between members and cooperatives. These dis putes typically relate to property, recovery of dues from members, differences between members and cooperative management boards, and so on. For example, the Coopera tive Court handles cooperatives’ foreclosure on members due to defaults in payment. On one hand, the creation of separate Cooperative Courts ensured limitation of power of Registrar’s Office to basic regulatory functions, so that the Registrar’s Office has more of an advisory role. On the other hand, Cooperative Courts provided a separate legal channel for settling disputes within the scope of Cooperative Societies Act in an efficient manner, without having to take recourse to the already beleaguered Civil Courts for settling disputes. The third feature of the cooperative institutional structure is the Mumbai Dis trict Cooperative Housing Federation (MDCHF), which emerged endogenously within the cooperative system to provide support to the internal functioning of cooperatives. MDCHF was organized as a central body in 1947-48 to assist primary housing coop eratives by providing them with financial, legal, and administrative advice. With rapid growth of Tenure cooperatives in the private sector, the Federation has since helped in evolving common policies and by-laws, and in raising consciousness with respect to various aspects of functioning of cooperatives. The agency conducts seminars on various aspects of cooperative housing (management, finance, auditing, taxation, reg istration of property, payment of stamp duty, and so on), publishes a monthly journal relating to current legal and procedural changes, and advises both cooperative boards 133 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and members. It has also standardized by-laws for Tenure Cooperatives. Private de velopers, public agencies, and non-governmental organizations use these services of MDCHF to form cooperatives. Cooperative board members obtain tax, legal, and ad ministrative advice. Members of cooperatives obtain legal and other advice. Such advisory support has been crucial to lowering the transaction costs of formation and functioning of cooperatives. Thus, the institutional structures strengthened the internal collective features of cooperatives, while overcoming some of the internal difficulties. 4.4.2. Institutional Structure for Financial Support Financial support to housing cooperatives had been available from the Bombay provincial government and cooperative banks during the pre-Independence period. Government financial support was, however, inconsistent and cooperatives had to de pend on shares raised through members and other cooperative banks. After Independ ence, Bombay State was one of the earliest states where the two-tiered institutional structure for providing financial support to cooperatives was established. The Apex Cooperative, originally called the Bombay Cooperative Housing Finance Society (BCHFS), was established in 1952 to mobilize finance for primary housing coopera tives. It was renamed Maharashtra Cooperative Housing Finance Society in 1960, when Maharashtra was carved out of Bombay State, and it was renamed more recently as Maharashtra State Cooperative Housing Finance Corporation Limited (MSCHFC). MSCHFC borrows from other sources in order to on-lend to primary coopera tives. Primary cooperatives could borrow from MSCHFC by becoming its member. MSCHFC covers its administrative and operational costs by adding certain percentage points to the interest rate at which it borrows from other sources. In an environment when there were few other formal sources of housing finance, MSCHFC provided a stable source of funding for primary cooperatives. Unlike commercial banks, 134 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. MSCHFC specialized in providing long-term loans for housing. Typically, these loans are secured through immovable property deeds. MSCHFC provides loans to primary cooperatives for constructing houses for the cooperatives’ members and for acquiring buildings for dwelling by members. Thus, although some funding is pro vided to Cooperative House Mortgage Societies, funding for on-lending by coopera tives to members is very little. Since 1991-92, MSCHFC started to include individu als as members and began to provide finance to such members. Until the 1970s, Life Insurance Corporation was the main source of funding to MSCHFC. Although the funding sources diversified since then to include other finan cial organizations (National Housing Bank, Housing and Urban Development Corpo ration), LIC is still the major source for finance. Table 4.11 lists the major sources from which MSCHFC obtained funding in 1995-96. As the table shows, LIC provided over 60 percent of the loan amount; cooperative banks provided 23 percent; and NHB provided 12.7 percent. Loans given to primary cooperatives using LIC funds are also secured through LIC’s Group Insurance Scheme. Table 4.11. Maharashtra State Cooperative Housing Finance Corporation’s sources of funding, 1995-96 Source Amount (million rupees) Share Life Insurance Corporation 2,001.28 60.7% Cooperative Banks 758.82 23.0% Housing and Urban Development Corpora tion 6.19 0.2% Unit Trust of India 112.50 3.4% National Housing Bank 418.82 12.7% Total 3,297.61 100% Source: Maharashtra State Cooperative Housing Finance Corporation, Annual Report 1995-96. MSCHFC’s operations were mostly concentrated in Greater Mumbai in its ini tial years. However, its role in Mumbai’s housing market declined since then for two 135 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. reasons. First, it increased its efforts to reach underserved areas of the state, which reduced Mumbai’s share. Second, housing prices in Mumbai reached beyond loan ceilings under MSCHFC’s stipulations. Although disaggregate data is not available for the number of housing units constructed in Mumbai with financing from MSCHFC, the amount of loans advanced give an indication of its operations in Mum bai. Table 4.12 gives a summary of MSCHFC’s financial operations in Mumbai be tween 1960-61 and 1995-96. While 95 percent of its loan amounts were disbursed in Greater Mumbai in 1960-61, its relative share steadily reduced and formed about 1 percent in 1995-96. Table 4.12. Loans Disbursed by Maharashtra State Cooperative Housing Fi nance Corporation, 1960-61 to 1995-96. Year Total mem bers Loans Disbursed (million rupees, unad justed) Share of Loan Amount advanced in Greater Mumbai 1960-61a 216 3.5 95% 1970-71a 2,128 72.4° 59%c 1980-81a 5,144 75.0 N.A. 1992-93b 10,111 483.5 1% 1995-96b 10,269 192.4 1% Sources: (a) Naik (1985), p. 80-81 and p. 163; (b) Calculated from Annual Reports of Maharashtra State Cooperative Housing Finance Corporation, 1990-92 to 1995-96. Note: Data (c) are for 1969-70. Since 1988, with the establishment of National Housing Bank, private Housing Finance Institutions (HFIs) and commercial banks became active in the city for pro viding housing finance. Housing Development Finance Corporation (HDFC) became the largest provider of housing finance. In the face of national economic liberalization policies and Mumbai’s local housing market, the share of HFIs in the formal private sector housing market considerably increased during the 1990s. Unlike MSCHFC, HFIs lend at competitive market rates, and have larger pool of funds for their loan op erations. Also, unlike MSCHFC, where the major share of lending is towards primary 136 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cooperatives, HFIs lend to individuals. The HFIs recognize cooperative ownership as 'y/r a security, so that members of housing cooperatives could borrow from them. Lend ing to cooperative members represents a marked shift from traditional lending to co operatives. In case of default in loan repayments, HFIs could potentially evict a mem ber and attach property, even though the cooperative (and not the member) owns such property. The liability for repayments thus rests with the member, and not the group. These institutional innovations have broadened the scope of lending to members in the formal private housing market; such members hardly borrow from MSCHFC.2 7 4.4.3. Institutional Support for Access to Land Unlike the previous two institutional structures, which are well defined, there is no formal institutional structure for mobilizing land for housing cooperatives. Yet, preferential treatment of cooperatives for allocation of public land cannot be ne glected. Housing policies during the post-Independence period in Mumbai consis tently supported such preferential treatment based on the rationale that cooperatives provide an efficient method for usage of land through multi-family housing. Land policies in Mumbai roughly parallel the national level policies on land acquisition. Thus, until the 1970s, when the state intervened in the housing process through public housing schemes, large scale land acquisition policies were in effect. In this context, housing cooperatives were considered as alternative to state ownership, and were either exempt from such acquisition or were given preferential treatment in allocating such land. Housing cooperatives also had a role in implementing Central 26 Typically, the HFIs require an allotment letter from the cooperative and cost/ sale/ resale deeds. Additional stan dard documentation include floor plans, monthly income receipts (for salaried em ployees), income tax returns (for self-employed). 27 Although MSCHFC started to lend to individual members in the 1990s, such lending forms a m iniscule portion o f its overall lending. 137 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and State government schemes. Central government projects like Subsidized Indus trial Housing Schemes included providing loans and grants to cooperatives of indus trial workers for their housing (Chapter 3). At the State level, the Post War Recovery Scheme (Backward Class Co operative Housing Scheme, PWR 219) initiated in 1948 gave a prominent place to co operatives formed of backward castes. Under the scheme, land was provided free of cost to co-operatives of backward classes where government land was available; where such land was not available, government subsidy equivalent to cost of acquisi tion or purchase of private land was provided. Long term (25 years generally) interest free loans were also provided upto the extent of 75% of the cost of the construction depending on the category of the backward class. To promote more assimilation of backward classes into the society and to remove “untouchablility,” non-backward classes were included up to 10 percent of the membership with similar concessions. MSCHFC continues to provide loans to Backward Class housing cooperatives based on State government guarantees. A smaller margin (0.25 percent) is added to the in terest rates on loans to Backward Class housing cooperatives, as compared to other types of cooperatives. The Urban Land (Ceiling and Regulation) Act enacted in 1976 sought to bring excess private land into state control, in which cooperatives were exempt from such land acquisition. However, as explained in Chapter 3, the Act adversely affected the land market by reducing the supply of land, and very little land was acquired by the state. According to BMRDA (1995, p. 256), of the 4,836 hectares of land declared for acquisition under the law, only 243 hectares was actually acquired. Housing coopera tives also benefited little from the Act. 138 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Public land for cooperatives is provided mainly through public housing agen cies (Maharashtra Housing and Area Development Authority, MHADA) and metro politan development agencies (Bombay Metropolitan Regional Development Agency, BMRDA; City and Industrial Development Corporation, CIDCO). These agencies, however, do not directly allocate land to cooperatives; rather, cooperatives are used as vehicles for providing ownership in public housing and other schemes. In Navi Mum bai (New Bombay), where CIDCO owns most of the land and is responsible for its development, CIDCO allocates land to cooperatives for housing development. Ac cording to a survey conducted by CIDCO (1995, p. 90), while CIDCO constructed the major bulk of housing (nearly 84 percent), cooperatives accounted for 13 percent of the housing construction. Further, slum development policies in Mumbai also empha size cooperatives for provision of secure tenure to slum-dwellers. Slum Redevelop ment schemes and major metropolitan projects like Bombay Urban Development Pro ject and Mumbai Urban Transportation Plan utilize cooperatives to form groups of slum-dwellers for providing tenure. 4.5. Diverse Roles of Housing Cooperatives I have argued that while the institutional framework facilitated the develop ment of cooperatives, the institutional structure reinforced the development by provid ing them crucial support. Hence, the activities of housing cooperatives grew phe nomenally in the post-Independence period. In this section, I examine the relationship between institutional context and the role of housing cooperatives during the post- Independence period in Mumbai. Cooperatives are used as vehicles for fulfdling housing objectives in both the formal and the informal sector, and they play diverse roles in the housing process. First, in the private sector, housing cooperatives have been adapted as mechanisms for collective ownership and management. Second, ma- 139 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. jor public programs (such as public housing and metropolitan development schemes) have utilized housing cooperatives as organizational vehicles. Third, cooperatives have been used as vehicles for mobilizing collective action among tenants for dilapi dated buildings in inner city areas that went into neglect due to rent-control laws. Fourth, cooperatives have been used as vehicles for implementing slum redevelopment programs. Fifth, non-government organizations use cooperatives in the informal sec tor to organize slum-dwellers, pavement dwellers, and indigent women for fighting for their housing rights, for self-help, and to organize them for microfmance. 4.5.1. Housing Cooperatives in the Private sector The Maharashtra Ownership Flats Act, 1963 legally formalized the need for private developers to form a collective organization like a cooperative or a company of prospective buyers before transferring the property. Another piece of legislation, the Maharashtra Apartment Ownership Act, 1970, legally formalized individual owner ship (equivalent to condominiums). The two pieces of legislation, thus, did not explic itly promote cooperatives per se; rather, they laid the legal need for forming collective organizations. Yet, in the context of Mumbai, housing cooperatives become more popular than other forms of collective organizations. As I highlighted earlier, part of the rationale for this lies in the historical prece dence of housing cooperatives over other forms of organizations. In the pre- Independence period, cooperatives were adapted as a modem organizational form for collective ownership and management for obtaining decent housing by the middle class. The internal organizational functions of housing cooperatives were thus already established to a certain degree. Tenure Cooperatives in the form of Tenant Co partnership Societies and Tenant Ownership Societies had already been functioning. 140 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In a way, to paraphrase Elster, the presence of cooperatives induced the preference for them. More importantly, however, the cooperative institutional support structures, which had been established to a certain degree in the pre-Independence period, be came more robust during the post-Independence period. While most observers of co operative development have recognized the need for financial support mechanisms, they overlook the need for administrative, legal, and other procedural advice (Horvat, 1982). Such support systems are required to overcome the internal collective prob lems of the cooperative. Other organizational forms hardly had similar support mechanisms. In this context, housing cooperatives became more popular than other forms of collective organizations. Indeed, the procedures for forming cooperatives are so facile that promoters can form them almost overnight. I am not arguing that there is a one-way causation relationship supportive be tween institutional structures and the development of cooperatives in the private sec tor. Rather, I am suggesting that there is a mutually reinforcing relationship between the internal functioning of cooperatives and the institutional structures. Tenure coop eratives are inherently stable due to perpetual ownership; hence relations between members are long-term. Consequent problems that arise between members are often solved within the collective decision making system of cooperatives. However, more enduring problems persist if members do not agree with the collective decisions, and disputes arise between cooperatives and members. The mechanisms to resolve such disputes are indeed important for the development of cooperatives; if difficulties per sist, cooperatives would loose popularity as a vehicle for collective ownership and management. On one hand, as cooperatives manifest the need for institutional struc ture to resolve such internal disputes and other collective problems, they reinforce the 141 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. development of the institutional structure. On the other hand, as the institutional struc ture becomes more refined, the formation and functioning of cooperatives become smoother. Thus there is a two-way relationship between cooperatives and the institu tional structure. A last feature of formation of housing cooperatives in the private sector is that they are not limited to any particular income group. Cooperatives range across all in come groups. Some traditional cooperators deride the private sector housing coopera tives in Mumbai, arguing that these cooperatives do not reflect the “cooperative” spirit. Rather, it is an opportunistic set-up for the wealthy, with little sense of mutual ity between members. Such arguments overlook the fact that the cooperatives in the private sector have also to follow the Maharashtra Cooperative Societies Act; hence, the cooperatives abide by the fundamental guidelines of collective ownership and de mocratic decision-making. Moreover, proliferation of cooperatives in the private sec tor may have its advantages. First, cooperatives are neither symbols of exclusivity in Mumbai, nor or they stigmatized as housing low-income groups only. Second, some of the costs of establishing institutional structures are borne by cooperatives in the pri vate sector. Such costs may be exorbitant for cooperatives to bear if they were limited to low-income. Once the institutional structures are established, positive institutional externalities accrue, which are useful for furthering the development of cooperatives. 4.5.2. Housing Cooperatives in Public programs Although there is no explicit institutional structure for providing land to coop eratives, public housing agencies supported development of cooperatives through preferential allocation of land. In this, the public agencies used housing cooperatives as vehicles for public housing programs and metropolitan development schemes. 142 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Housing cooperatives were considered as alternatives to the state for equitable distri bution. Until the 1970s, when public housing programs were in effect, the State gov ernment constructed rental housing. Under the Subsidized Industrial Housing Schemes, for example, the Bombay Housing Board constructed rental housing for in dustrial employees; some of these schemes were also implemented through coopera tives. Low-income and middle-income group housing were also constructed on a rental basis. Rental housing in the public sector suffered from similar problems as the private sector. Low rent recovery and lack of resources for maintenance became prob lematic, resulting in neglect. Consequently, since the 1980s, the Board started to trans fer ownership through tenant cooperatives, which became responsible for maintenance and management. Tenants could purchase housing units on a hire purchase basis. Later public housing schemes have also been based on ownership, rather than rental. Under the Mass Housing Program of CIDCO and MHADA, for example, houses have been built for sale. Multi-family housing constructed under the scheme have been largely allocated through cooperatives. Cooperatives have been also used as vehicles for land allocation in the Land Infrastructure Servicing Program (LISP), a sites and services scheme carried out under the World Bank funded Bombay Urban Development Project (BUDP). While plots for individual housing (for mainly middle and higher income groups) were leased to individual households, plots developed for cluster housing (for low-income households) were leased to cooperatives.2 8 Besides public housing programs, the public agencies also provide preferential treatment in the allocation o f public land. CIDCO, for exam ple, provides preferential 28 Cluster housing principally consists o f 25-30 single room units (mainly ground floor) clustered around a small open space. Households share common toilet facilities. 143 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. treatment in land allocation to cooperatives among even high-income households. In deed, the bulk of housing constructed in New Bombay are through cooperatives. Ac cording to Bhattacharya (1995, p.5), nearly 80 per cent of tenements constructed in the private sector have been through cooperatives. 4.5.3. Housing Cooperatives and Inner City Housing In the inner city areas, housing cooperatives were promoted as means for re pairs and reconstruction of older dilapidated buildings, which had been neglected by landlords essentially due to rent control. If rent control measures had not been there, it is doubtful the need for formation of housing cooperatives would have arisen. In Mumbai’s political context of strong tenant groups, rent control measures received little resistance, and were reinforced over time. Rent control reduced incentives for landlords to spend on maintenance, in which older buildings in inner city were particu larly affected. The Report o f the Housing and Housing Finance Panel noted in 1959 (Government of Bombay, 1959, p. 20), “There are 18,000 old buildings containing about 1,61,300 tenements in Greater Bombay. 6000 of such buildings containing 57,000 tenements will finish their useful life within the next five years.” Initially, the government assumed the task of rehabilitating old buildings through Bombay Building Repairs and Reconstruction Board Act, 1969. Under the Act, Bombay Repairs and Reconstruction Board (later Bombay Housing and Area De velopment Board) was formed for undertaking structural repairs to buildings that could be repaired. The Board also acquired those buildings that were beyond repairs and reconstructed them to re-house residents who had been staying in them. The re pairs and reconstruction were principally financed through Building Repairs and Re construction Cess imposed on buildings identified by the Board. The Board identified 19,642 buildings for cess in 1971, of which over 84 percent were constructed before 144 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1940. Obsolescence of housing stock increased over years as indicated by subsequent surveys conducted in 1981 by Bombay Municipal Corporation, which identified 30,237 buildings in various stages of dilapidation. The pace of repairs was thus not commensurate with the obsolescence. The Rent Acts Enquiry Committee prescribed that “old buildings should be taken over or acquired by Government at a pre-determined amount and transferred to the co-operative housing societies or associations of apartment owners formed by the tenants of such buildings” (Government of Maharashtra, 1977, p. 74). Subsequent housing committees in Mumbai expanded Rent Acts Enquiry Committee’s recom mendations relating to the role of tenants’ cooperatives. The High Powered Steering Group for Slums and Dilapidated Houses (Government of Maharashtra, 1981), for example, recommended formation of cooperatives of tenants as a pre-condition for receiving technical and financial government assistance for both repairs and recon struction. Under the Steering Group’s recommendations, building and land would be compulsorily acquired by the Board and leased to tenants’ cooperatives. Tenant coop eratives could undertake structural repairs themselves by taking a “No Objection Cer tificate” from the Board, which reimbursed a part of the repair costs upto a preset ceil ing. The Maharashtra Act 21, 1986 legalized the requirement for forming coopera tives of tenants to undertake repairs and reconstruction of buildings constructed before 1940. According to the law, at least 70 percent of the tenants needed to consent for forming a cooperative. If the quorum were met, others unwilling to join the coopera tive would become a tenant of the cooperative and continue living on the same terms and conditions as before. BHADB acted as an intermediary in the transfer of owner ship by acquiring the property. The costs of acquisition were to be covered through a 145 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. down payment by cooperatives to the Board, which, in turn, would compensate the landlord. Since the law was essentially aimed to benefit sitting tenants, the Act pro hibited cooperatives and members to sell or transfer the property. Tenants favored obtaining ownership rights, but there are contradictory reports on their attitude towards forming cooperatives. In the sample survey of households in cessed buildings conducted by the Tata Institute of Social Sciences (Sharma, Raju, and Reddy, 1986, p. 35), about 88.7 percent of the respondents favored the transfer of ownership to tenants. Similarly, in another survey conducted by Indian Human Set tlements Program (IHSP), nearly two-thirds were willing to own reconstructed flats through hire-purchase schemes.2 9 With respect to formation of cooperatives, accord ing to the Tata Institute of Social Sciences survey (conducted before the 1986 Act), about 73.5 percent of the respondents did not view tenant cooperatives as a feasible mechanism to undertake repairs and maintenance. Respondents found factors like “differences of opinion among their neighbours and their non-cooperative attitude as the main hurdles in the direction” (p. 43). However, in a later survey conducted by Dua (1991), nearly 81 percent and 96 percent of the respondents preferred cooperative 30 as a vehicle for undertaking repair and reconstruction respectively. From 1970 to 1992, about 20,172 buildings were put on the priority list for re pairs, of which the Board repaired 10,697 and Tenant cooperatives repaired 3,562 buildings. During the same period, 1820 buildings were declared beyond repairs, of which about 273 buildings (with nearly 20,000 tenements) were reconstructed (MHADA, 1993, p. 27-28). The reconstruction process has been slower due to legal delays in the acquisition o f properties b y the Board. 29 Based on Dua (1991), Table 26, p. 44. 30 Calculated from Dua (1991), Tables 29(A) and 29(B), p. 47. 146 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.5.4. Slum Redevelopment Programs The usage of housing cooperatives in the slum redevelopment programs per haps best represents the broad acceptance of cooperatives as vehicles for achieving housing objectives. The Slum Upgradation Program (SUP) implemented under BUDP explicitly envisaged slum regularization on a cooperative tenure basis. Under the pro gram, 200 slum areas (housing about 100,000 households) were to be upgraded for transforming slum housing to “Environmentally Acceptable Legal Shelter.” The prin cipal argument for providing cooperative tenure was that speculation could be poten tially contained. As land was collectively held, individuals could not sell their prop erty after obtaining tenure (which had been a typical problem in slum upgradation pro grams, resulting in “upward filtering”). By 1993, about 15,475 households in 140 cooperatives had received collective tenure (Panwalkar, 1996, p. 138). While the main focus of SUP was on providing collective tenure, the larger goal was to improve the living environment. Cooperatives became ideal vehicles for community development. They were used to identify community groups and to de marcate clear boundaries of slums for regularization. They were also used to provide common services like roads, drainage, water pipes, balwadis (children’s centers), and so on. Panwalkar (1996, p. 132) reports that the concept of cooperative tenure was so • " J 1 well institutionalized that it did not meet with much resistance from slum-dwellers: Since most of Bombay’s slum dwellers are by origin rural and the co operative network in the rural Maharashtra is well established, dissemi nation of information on cooperation as a concept that must be institu tionalized did not meet with any resistance; in fact there was an enthu siastic response. While residents of slums covered under the scheme were thus able to internalize the general concept, what needed more 3 1 See Panwalkar (1996) for an excellent description o f ground realities in orchestrating housing cooperatives for the SUP program. 147 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. persuasion was to obtain acceptance of the specific nature of a housing cooperative within the context of the SUP. Subsequent slum redevelopment schemes have also been based on the coopera tive as an organizational vehicle for collective tenure and for achieving collective ac tion. The Prime Minister’s Grant Program (PMGP), which was funded in 1985 mainly for upgradation and redevelopment of Dharavi, also utilized cooperatives as organizational vehicles. While upgradation under PMGP was similar to SUP, rede velopment aimed at rehousing the slum-dwellers in five-storied housing complexes. Similar to the SUP program, co-operatives were to be provided 30-year, renewable leases to the redeveloped land. Slum redevelopment programs gained more support than slum upgradation programs partly due to the possibility of creation of a real asset, T9 and became the thrust of slum policies during the 1990s. The Slum Redevelopment program (SRD) program initiated in 1991 combined formation of slum-dwellers’ cooperatives and market mechanisms for slum redevel opment. Under the program, 70 percent of the slum-dwellers in a slum pocket needed to consent and form a cooperative before undertaking the redevelopment program. The cooperative would hold collective tenure to the land, and private builders would carry out the construction. Private developers were given an incentive to build by stipulating a higher Floor Space Index (FSI) of 2.5 for slum redevelopments. How ever, SRD program did not met with much success, partly because it was not yet fi nancially lucrative for private developers, and partly due to procedural problems. The Afzalpurkar Committee appointed in 1995 sought to overcome some of constraints in the profitability of private developers by relaxing some of the Develop ment Controls. Based on its recommendations, SRD was replaced with SRS (Slum 32 See Mukhija (2001b, 2002) for some explanations on the preference o f redevelopment over upgradation. 148 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Rehabilitation Scheme). SRS provided abatements in municipal property taxes, and allowed for transfer of development rights to make the projects more commercially lucrative. Yet, the need for 70 percent consent by cooperatives of slum-dwellers re mained, which has been criticized by some observers as a hurdle for redevelopment (see, for example, O’Hare, Abbott and Barke, 1998, p. 281). By 1998, about 446 pro posals were received, of which 367 were approved for implementation (Mukhija, 2001a, p. 801). 4.5.5. Non Government Organizations and Housing Cooperatives Non-Governmental Organizations (NGOs) became active in Mumbai since early 1980s, when a broad civil society movement emerged to protest large-scale evic tion from slums. According to Desai (1996, p. 237), NGOs’ “efforts culminated in a Supreme Court judgement in 1984, which established the possibility of security of tenure for many slum-dwellers.” NGOs have been largely involved in providing ser vices and advocating for housing rights of slum and pavement dwellers. They adapted cooperatives as community organization mechanisms in the context of emphasis of formal policies on housing cooperatives for slum upgradation. Thus, most NGOs use the cooperative as a vehicle for mobilizing slum-dwellers and pavement dwellers into groups for community action. NGOs have also used cooperatives for organizing indi gent women into collectives for micro-finance. These cooperatives are usually infor mal (i.e. they are not formally registered under Maharashtra Cooperative Societies Act), but the broader interest is to establish formal cooperatives that can participate in the slum redevelopment programs. One illustration of usage of cooperatives is that by the Alliance of three NGOs: Society for Promotion of Area Resources Centre (SPARC), National Slum Dwellers Federation (NSDF), and Mahila Milan. SPARC, a well-known NGO in Mumbai, was 149 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. established in 1984 and is mainly involved in training, educating, advocating and lob bying activities. NSDF was established in 1974, and advocates housing rights for slum-dwellers across Mumbai and other Indian cities. Mahila Milan, which literally means “women together,” was established in 1987, and works with women pavement dwellers in Mumbai City. The Alliance of three organizations has been instrumental in organizing four projects for slum rehabilitation in Mumbai. As NSDF had been utilizing informal co operatives as a means for community action already, the members were organized into a formal cooperative to obtain legal tenure. However, unlike the redevelopment schemes undertaken through SRD or SRS, housing undertaken through the NGOs in volves substantial community participation in planning and design and sweat equity in construction. Mahila Milan assists collectives of women in managing their own resources. It initiated a network of women’s groups among pavement dwellers to pool their sav ings. There are three kinds of savings schemes: the crisis savings and credit scheme, the income generation scheme, and the housing savings schemes. Each women’s net work consists of about 10 to 15 women from within the community, who are also re sponsible for managing the accounts and for taking decisions. The Crisis Credit Scheme is essentially a rotating savings and credit scheme for emergency or medical uses. By 1996, nearly 10,000 women were involved in the scheme (People’s Dialogue Report, 1996). The Income Generation Scheme is funded through a government credit line called Rashtriya Mahila Kosh to SPARC. The funding is provided to a women’s group at 8 percent interest rate, which is on-lent to m em bers at 12 percent interest for business purposes. In Housing Savings schemes, savings are pooled into a coopera tive so that the group becomes more credit-worthy for borrowing from other sources. 150 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.6. Conclusion In the pre-Independence period, housing cooperatives emerged in Mumbai in the context of the city’s historical economic and demographic characteristics. Rapid economic growth of Mumbai during the late 19th century due to specific local and global historical factors had resulted in the modernization of the city, which almost obliterated the city’s underlying rural character. The economic growth attracted sig nificant labor migration into the city, most of which was composed of single male la borers. With the growth in housing demand, housing morphology in the form of mod em multistoried buildings became the norm. In this context, demand for collective organizations like cooperatives grew. Housing cooperatives were adapted as “mod em” organizational vehicles for providing “decent” middle class housing. These co operatives were mainly Tenure Cooperatives, which were organized for collective ownership and management of multi-family housing. Provision of finance and land by the provincial government of Bombay provided a boost to cooperatives’ activities. In the post-Independence period, activities of housing cooperatives grew phe nomenally in Mumbai. Both the local housing market and the institutional framework of Mumbai were conducive to the growth of formal housing cooperatives. In terms of the local housing market, the demand for collective organizations grew with the growth of multi-family dwelling units. In terms of institutional framework, flat own ership became more optimal due to rent control policies since Independence. Al though flat ownership regulations enabled cooperative as well as other types of owner ship (condominiums, associations, companies) in multi-family units, Tenure Coopera tives became more prominent than other organizational forms. While part of the rea son for prevalence of cooperatives is historical precedence, I argue that the develop 151 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ment of supportive cooperative institutional structures enabled growth of housing co operatives as compared to other organizational forms. The cooperative institutional structure supporting cooperatives aided coopera tives to overcome some of their internal weaknesses, and reduced their transaction costs of formation and functioning. First, due to liberal cooperative policies in Ma harashtra, state controls on functioning of cooperatives were minimal. Structurally, Registrar’s office could hardly interfere in the internal decision making processes of cooperatives, so that they were relatively autonomous. Additional structures for ad ministrative, legal, and procedural support also arose. Second, the two tiered institu tional structure for finance through Apex cooperatives provided a stable source of funding in the initial decades. Third, although there is no well defined structure for providing land, public housing agencies gave preferential treatment to housing coop eratives. The ease of forming groups through cooperatives facilitated cooperatives to diversify and to undertake multiple roles. First, in the private sector, housing coopera tives have been adapted as mechanisms for collective ownership and management. Second, major public programs (such as public housing and metropolitan development schemes) have utilized housing cooperatives as organizational vehicles. Third, coop eratives have been used as vehicles for mobilizing collective action among tenants for dilapidated buildings in inner city areas that went into neglect due to rent-control laws. Fourth, cooperatives have been used as vehicles for implementing slum redevelopment programs. Fifth, non-government organizations use cooperatives in the informal sec tor to organize slum-dwellers, pavement dwellers, and indigent women for fighting for their housing rights, for self-help, and to organize them for microfmance. 152 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 5. Housing Cooperatives in Chennai Similar to Maharashtra, housing cooperatives had been active in Tamil Nadu before Independence. Although housing cooperatives in Tamil Nadu are yet more ac tive than many other Indian states, their growth is quite modest in Chennai. Housing cooperatives are mainly Finance Cooperatives, and hardly diversified their activities since Independence. Unlike Mumbai, the institutional context in Chennai did not fa vor the development of cooperatives. In the context of Chennai’s modest urban growth during pre-Independence pe riod, multi-family housing in modem structures hardly came up. Housing develop ment was horizontal (a phenomenon that continued till the 1970s), rather than vertical, wherein individual plotted developments became the norm. Hence, there was little demand for collective organizations like housing cooperatives for collective ownership and management. Tenure Cooperatives hardly emerged in Chennai. Housing coop eratives were adapted for financing and constructing housing in individual plotted de velopments. In the post-Independence period, government controls implied that coopera tives had little autonomy. Consequently, cooperatives developed in the direction em phasized by the government, and have been hardly active in roles beyond that pre scribed by the state. As the state emphasized use of cooperatives as vehicles for fi nance in rural areas, their role in urban areas decreased. In this context, the coopera tive institutional structure for financial support to housing cooperatives became robust. Besides the financial support structure, there are little support mechanisms in Chennai. Consequently, transaction costs of formation and functioning of housing cooperatives are high, and there has been little growth in the activities of housing cooperatives. 153 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. This chapter is organized in the following way. First, I analyze the growth of housing cooperatives in Chennai during the pre-Independence period. Second, I out line the growth of housing cooperatives during the post-Independence period. Third, I examine the institutional context of Chennai. Fourth, I examine the development (or the lack of it) of cooperative institutional support structures. Fifth, I conclude with identifying the conditions that restrained the growth of cooperatives in Chennai. 5.1. Pre-Independence Period Since its inception, Chennai was distinctive from Mumbai in its economic and demographic characteristics, which influenced its housing morphology. Chennai de veloped as a “city of villages” so that its internal rural structure was preserved. It de veloped as a network of settlements with relatively low-rise housing structures on in dividual plots. Housing cooperatives that emerged were adapted as credit mechanisms for housing development on individual ownership basis. 5.1.1. Chennai’s Urban Growth and Demography The evolution of housing morphology of Chennai is set in its colonial histori cal growth as a city o f villages. For nearly 150 years since its inception as a British colonial city in 1639, a “segmentary urban pattern evolved consisting of a series of loosely linked self-contained villages, depending only in a marginal way on the com mercial center of Madras” (Lewandowski, 1975, p. 347). The growth of the city as a group of village-like settlements with distinctive concentrations of caste and occupa tional skills contributed to its spatial and social fragmentation in a predominantly rural atmosphere. Unlike Mumbai during pre-Independence period, Chennai did not grow as a compact city around a colonial center. The original colonial settlement consisted of 154 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Fort St. George, surrounded by the White Town (consisting of the British) and the Black Town and its suburbs (consisting of native Indians). Although the colonial population had outgrown the White Town by later 1700s, it did not expand contigu ously since the land adjoining the Fort was reserved for military (cantonment) areas. The Black Town was also congested and had poor sanitation and ventilation. Lewandowski (1975, p. 347) adds that “each of the other villages that came together to form the city of Madras had its own functions, service castes, bazaars, temples, mosques and churches.” Each village was self-contained, which distinguished Chen nai’s settlement pattern. By late eighteenth century, the dual structure of White Town and Black Town started to dissolve and was accompanied by leap-frog spatial expansion of both the British and the Indians to suburban areas. Suburban growth started by 1760s and be came intense by the end of the century—almost half the land shifted from village pro prietors to private ownership by the end of the century. British residents impelled suburban growth towards the south (e.g., Nungambakkam), and west of Cooum river (e.g., Egmore, Chetpet, Kilpauk) in their search for spacious garden houses and es tates. Native Indians also contributed to the growth, but their settlements were more dense and towards the north (e.g. Tiruvottiyur, Velachery). Older native village set tlements like Mylapore and Triplicane also predated the colonial city. Unlike Mum bai’s linear geography, Chennai’s flat plain was conducive for suburban expansion, limited only by the ocean (Bay of Bengal) to the east. By 1800s, Chennai acquired some urban centers with the growth of mercantile, financial, retail, and administrative establishments in the Fort area and Black Town. British residents continued to live in suburban areas segregated from indigenous set tlements, and they commuted to work in the urban center. The colonial urban form of 155 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. spatial differentiation of social and functional areas was in contrast with the native In dian settlements. Native settlements were spatially coherent but socially heterogenous with “higher” castes towards the center of the village around a temple, and the “lower” castes towards the periphery (chief examples being Mylapore and Tiruvottiyur). New settlements that had come up in response to the trade opportunities generated by the East India Company also became dense localities with a wide range of castes (e.g. Chintadripet for weavers, Washermanpet for washermen, Royapuram for boaters). With more urbanization during nineteenth century, Chennai became a city with several densely populated settlements, which had distinctive identifiable characteristics re flecting the local rural social order. In analyzing the growth of Chennai during eighteenth and nineteenth century, Neild (1979, p. 218) argues that the city consisted of three strands of separate, yet overlapping societies. The first strand consisted of suburban villages, which belonged to pre-colonial agrarian society. The second strand comprised of town centers, which had links in indigenous urban and rural society, but grew principally in response to colonial settlement. The third strand was the colonial urban and suburban society, which reflected the British colonial interests and policies. By the 1830s, at least forty identifiable untouchable settlements existed; the East India Company also set up sev eral colonies of specialized castes and occupations for textile production (Neild, 1979, p. 227-228). In spite of the few commercial developments in the city, Chennai did not have a large scale industrial growth during the nineteenth century. Textiles manufactured in the city w ere exported through M umbai. C hennai’s port had not yet gained significant volume of trade. A few smaller industries like tanneries, railway workshops (the ma jor railway station, Central station was built in 1872), cotton mills, and timber saw 156 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. mills had also come up. A large segment of the population was, however, employed in government, professional, and personal services (by 1900, about 25 percent of the city’s population was engaged in such activities) (Lewandowski, 1975, p. 342). Chennai was thus predominantly an administrative center and, to a lesser extent, a trading center. Low commercial and industrial development was conducive in pre serving the internal village structure of settlements in Chennai. Many of the distin guishing characteristics of the rural settlements persisted. This was unlike Mumbai and Calcutta where rapid commercial growth virtually obliterated their internal village structure. Urban growth was slow in Madras province in the early decades of twentieth century since rural agricultural lending was quite lucrative for local moneylenders (sowcars), so that capital was directed towards villages.1 Moneylenders’ activities were quite ubiquitous since they were primary sources of funding for a wide range of agricultural activities. To relieve peasants’ dependence on moneylenders, Sir Freder ick Nicholson, Collector of Madras Presidency, proposed setting up of Raiffeisen type cooperatives that could be built on existing indigenous credit associations (Nicholson, 1895). Nidhis, chit funds, and other mutual benefit organizations that functioned simi lar to rotating savings and credit associations had already been functioning in the prov ince. Nicholson’s report was one of the influential pieces for bringing cooperatives to India, culminating in the Cooperative Societies Act, 1904. Credit cooperatives grew quickly as additional vehicles for credit besides other traditional mechanisms. Baker (1981, p. 587) claims that the “notion of the coopera tive became a panacea which government applied to all internal economic problems... 1 The Industrial Commission o f 1918 and the Labour Commission o f 1931 blamed rural lending for lack o f capital in industrial development. 157 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Such problems were blamed on heinous ‘middlemen’, and some form of cooperative was prescribed as a cure.” Besides credit, cooperatives were promoted in a broad range of fields, including marketing, consumer stores, weaving, and so on. Housing cooperatives also emerged as Ordinary Building Societies, which principally provided credit for housing construction. Although Chennai’s economic growth was slow in the beginning of twentieth century, it accelerated since the 1930s. By 1931, Chennai’s link to the national rail way system had been completed, including electrification of suburban railway lines (MMDA, 1980, p. 118). Newer enterprises came up, while older ones expanded. Ma jor enterprises included two cotton mills, European banks, and several joint-stock companies; other enterprises included tanneries, beedi rolling, and weaving industries. Export trade through Chennai’s port also increased. Yet, the city was primarily an administrative and commercial center and some of the major industries were those needed by the government—mint, armory, and railway networks. While much of the activities were concentrated in the Fort area, other village settlements also had secon dary facilities in terms of market areas, temples and so on. The growth of other set tlements was not completely haphazard though. Indeed, the Madras Town Planning Act was passed by 1920, a few years after Patrick Geddes had highlighted the need for planning in Madras province. Essentially modeled on the British system, the Act categorized land use into four uses—residential, shopping and business, open space, and industry. Most residential neighborhoods that came up consisted of housing con structed on open developed plots. Consistent with the economic pattern of Madras province, population growth in urban areas was nominal in the beginning of the century. Chennai’s population grew from 397,552 to 578,550 between 1872 and 1921, representing an average an- 158 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. nual increase of under one percent. Since the depression of 1929, Chennai’s annual population growth rate doubled to about two percent—it reached 713,394 in 1931 and 865,334 in 1941 (Table 5.1). Table 5.1. Population Growth of Chennai, 1872-1946 Year Area (sq. km.) Population Annual growth rate (av., %) Gross Density persons/hectare 1872a 68.17 397,552 - 58.3 1881a 68.17 405,848 0.21% 59 5 1891a 68.17 452,518 1.15% 66.4 190? 68.17 541,167 1.96% 79.4 191? 68.17 555,620 0.27% 81.5 192? 68.17 578,550 0.41% 84 9 193? 68.17 713,394 2.33% 104.6 194? 77.21 865,334 2.13% 112.1 1946c 77.21 1,235,000 7.12% 159.9 Sources: (a) Lewandowski (1975, p. 343); (b) MMDA (1995); (c) Report o f the Ma dras Provincial Housing Committee (1948), p. 12. At least two factors contributed to the spurt in population growth since the 1930s. The first was push factor from rural areas since agricultural prices dropped (following the 1929 depression). Consequently, rural lending went down and agricul- 'y tural indebtedness increased several folds. The second was pull factor—the reduction in agricultural prices was accompanied by a favorable shift in terms of trade towards industry. The number of joint stock companies rose rapidly in Chennai and other ur ban areas of south India during the 1930s. As Baker (1978, p. 238) describes: Moneylending had been by far the safest and most profitable outlet for surplus capital in the countryside. In the prosperous years of 1920s there had been a big build-up of capital in the villages, and a massive over-investment in land and moneylending in the most fertile regions. In the 1930s the rural capitalists began to look for other avenues for in- 2 According to Baker (1978, p. 237), “The government, in an effort to prevent a complete seizure, set aside Rs. 10 lakhs for rural credit and distributed it through cooperative societies. But within months the cooperatives had joined the scramble to realize debts. B y late 1931, almost half o f the ten thousand cooperative societies in the prov ince had collected less than 40 percent o f repayments due, and their total arrears ran to Rs. 311 lakhs. Nine hun dred societies were dissolved in the next two years.” [One lakh is equivalent to 100,000.] 159 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. vestment. Not only capital but also people began to leave the land for towns. In 1935-36, the land market in the province began to pick up. Although the city sprawled over 68.17 square kilometers, population distribu tion was uneven, with dense concentrations in a few settlements. The city limit was extended in 1941 to include the suburban area of Mambalam in the south, and the area increased to 77.21 square kilometers. Following World War II, Chennai’s population increased more rapidly. By 1946, the population had increased to over 1.2 million (Government of Madras, 1948), when several villages of Chingleput district were in corporated into Chennai city and the area was extended up to 128.83 square kilometers. 4.1.2. Chennai’s Local Housing Market In the context of Chennai’s growth as a city o f villages, the housing morphol ogy of Chennai followed the village housing pattern, with relatively low rise houses made of mud and tiles on individual plots. Many of the larger houses were originally constructed as single family dwellings or for the extended family, but with population growth, multiple families occupied a single house. The large garden houses of the British with expansive tracts of land were laid out in cultivated fields that were later integrated with the city. The pattern of large garden houses was still maintained when Indians employed in gainful professions bought them later. With more densification, there was more intensive use of land with horizontal expansion, rather than vertical expansion. Poorer rural migrants settled in mud and thatch houses at the periphery of the city. Lewandowski (1975, p. 348) relates: The development of land use patterns in eighteenth and nineteenth cen tury Madras followed the model of the old South Indian village, whose streets were lined with shops and low-lying residential buildings made of mud and tile. As population expanded in Madras (and even in eight eenth century there were high density areas of settlement such as Black town), houses were constructed on a horizontal rather than a vertical plane. Although upper floors were added onto houses in Bombay and 160 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Calcutta during this period, Madras remained a single-storied city. In Calcutta, for example, there was a steady evolution of brick built houses in the old city; by 1901, 42.5 percent of houses were multi- storied. In Madras, only the wealthiest families chose to build multi- storied terraced or tiled houses. Alternatively, as the need for private space increased in the densely populated neighborhoods, extensions were added to the fronts and backs of houses. Thus a larger portion of surface area of the city became occupied by buildings. The demographic characteristics of Chennai were also distinctive from that of Mumbai. Much of the population growth due to migration was from rural areas within the Madras presidency. Migrants belonging to a few castes (e.g. Vellalars) had close relationship with the colonial authorities, and could wield much of the same influence and status as that in the countryside. The rural-urban relationship was also fluid with other rural labor migrants who maintained their rural base. Individual house designs also reflected the rural pattern (e.g. courtyard houses), whereby similar activities could be carried out. Nield argues (1979, p. 239), “the society and culture of Madras, as a result, reflected the traditions and social order of its rural hinterland to a greater extent than either of Calcutta or Bombay.” Female to male ratio was also comparatively high in Chennai due to migration of families working in the service sector; in Mumbai there were more single men who sought employment in factories. By 1900, there were 984 females per 1000 males in Chennai; in Mumbai, there were 617 females per 1000 males (Lewandowski, 1975). Housing that developed in Chennai had to accommodate families, while that in Mumbai developed to accommodate working men (e.g. in single room chawls). Population growth in Chennai since the 1930s was not accompanied by an equivalent increase in housing construction to accommodate it. Instances of multiple families occupying single house had already started since 1870, when there were 51,741 houses accommodating 87,075 separate families. Subsequent increase in 161 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. population outstripped growth in housing construction—while population grew by 25 percent, the number of houses increased by 15 percent between 1880 and 1900 (Lewandowski, 1975, p. 349). The slower pace in the growth of number of houses increased crowding. In 1921, there were 64,621 houses, with an average of 8 persons per house. In 1931 and 1941, there were 73,845 and 87,888 houses respectively, and residential density increased to about 9 persons per house (Government of Madras, 1948). The overall gross density of the city doubled from 58 persons per hectare in 1872 to about 112 persons per hectare in 1941 (Table 5.1). The density increased more rapidly following World War II, when war time controls on building materials constrained housing construction. There were 90,000 houses by 1946, with an average of 14 persons per house (Government of Madras, 1948, p. II).3 According to the Government of Madras (1948, p. 12) report: It is reported that in one house thirteen families live, eight in the front yard and five in the back yard. Such instances are by no means rare. On an average in the houses occupied by the lower middle-class peo ple, at least four to six families reside in each house while the accom modation can at best be regarded as sufficient only for one or two fami lies. In some cases, it is stated, each family occupies one room which serves as a kitchen, living room and bed room all in one. Slum housing increased with urbanization. Most of the lower castes and poorer segments were accommodated in cheries. Whereas one-eighth of the popula tion lived in cheries by 1871, the proportion rose to a third in the 1930s. In 1934, there were 189 identifiable slums, consisting of 15,942 huts (Government of Madras, 1948, p. 14). Over 200,000 people lived in such slums, and another 10,000 were homeless. Many of these included daily wage earners (coolies, rickshaw pullers, etc,) 3 The number o f houses included both permanent structures (bungalows, mansions, tenements) and temporary structures that were in legal conformity. 162 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and the indigent (including beggars). By 1946, about 200 cheries and slums could be identified in Chennai (Government of Madras, 1948, p. 14). In spite of the increase in housing demand, multi-family dwellings were still rare; much of the housing was in the form of single storied individual plotted devel opment. In many parts of the erstwhile Blacktown and in other settlements (Chintad- ripet, Choolai, and so on where weavers, artisans and workers lived), higher popula tion densities meant more densely constructed housing. Dattatri and Anand (1991, p. 20) observe that “Madras prior to 1939 could be described as a single storey city. 68% of all houses were tiled and a substantial number were Madras-terraced.”4 5.1.2. Rise of Housing Cooperatives as Financial Mechanisms In the context of Chennai’s local housing market, housing cooperatives were adapted as financial mechanisms. Informal group credit systems like chit funds had already been functioning in the Madras province, so that credit cooperatives were not entirely alien. Finance Cooperatives emerged principally in the form of Ordinary Building Societies, which were “confined mainly to the provision of capital for build ing schemes and only occasionally extend[ed] to the joint purchase of land or of build ing materials” (Reed and Sheppard, 1928, p. 420). These cooperatives principally provided long-term credit to members for house construction on an individual owner ship basis. The site and building acted as mortgage security for loans. Some coopera tives also acquired land and divided it into plots for allocating it to members. Growth of Ordinary Building Societies was initially limited due to lack of availability of long-term loans. There were eight such cooperatives in Madras in 1914-15, and increased to eighteen by 1922-23.5 Activities of Ordinary Building So 4 Madras terrace is a form o f tile, typically used for roofing. 5 Government o f Madras (1928, p. 15). 163 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cieties accelerated since 1922-23 when the Madras provincial government started pro viding funds for onward lending. By June 1947, there were 130 cooperatives in the province, of which 128 were Ordinary Building Societies.6 Cooperatives in Chennai followed a similar trend as that of Madras province, i.e. they were principally Ordi nary Building Societies. As the Report of Madras Provincial Housing Committee (Government of Madras, 1948, p. 16) observed: In the Madras city, there are 12 co-operative building societies. All of them have confined their activities to acquiring building land collec tively and parceling it out to their members as building sites, and fi nancing them for the construction of buildings and arranging to recover the loans in equated monthly instalments spread over a period of 20 years. Financing has, however, been the principal activity of these so cieties. They do not even develop the land by laying roads and provid ing other amenities as a co-operative venture. According to the report, four of the twelve cooperatives were composed of industrial employees, one by migrants from Maharashtra, and the rest were open to general pub lic. By June 1946, members of primary cooperatives had constructed 576 houses in Chennai. The first successful lending partnership between the state and cooperatives was exhibited by the Coimbatore cooperative society in 1918-19. It setup an early model for the two-tiered lending, which was later adopted on a national scale in the post- Independence period. First, the government lent to primary cooperatives at a nominal interest rate; second, the primary cooperative added suitable percentage points to the interest rate (for covering overhead and intermediary charges) and on-lent the funds to members. These loans were typically for twenty years, although they could be lent for thirty years w ith the C ooperative Registrar’s perm ission. 66 O f the other two, one was a Cooperative Township (started in Katpadi) and the other was a House Construction Society (started in Madras city) (Government o f Madras, 1948, p. 62). 164 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Some false starts were made with other types of housing cooperatives. For ex ample, cooperatives of industrial workers—especially those in cotton mills—had started in 1939, whereby workers obtained housing on a hire purchase basis. Workers formed cooperatives and raised finance by pooling their share capital. Employers con structed houses on behalf of the workers, and recovered the costs through the coopera tives. Workers owned the house after full payments were made. However, some em ployers did not find it beneficial to provide housing through this arrangement. The Buckingham and Carnatic Mills in Chennai, for example, disapproved the scheme “for the reason, that as retiring employees keep the houses, new employees have to be pro vided with new houses adding a constant burden on the mills besides the possibility of a mill colony developing into a mixed colony” (Government of Madras, 1948, p. 32). Tenure Cooperatives did not become popular in Chennai in the face of its housing morphology. In the context of its rural social order, single family housing (including English bungalows) emerged; in spite of densification, single room tene ments and multi-family housing did not emerge. Much of the housing was individu ally owned. Housing expansion was horizontal, rather than vertical. Overcrowding implied horizontal densification (i.e. more ground level built up area)—more than one household occupied single-family houses, as portions of the house were rented to other households. Thus, collective ownership of housing did not emerge. The Report of Madras Provincial Housing Committee (Government of Madras, 1948, p. 61) ob served: Co-partnership system has never been popular in this presidency. The following advantages are absent in this type of society: (1) individual ownership, (2) an independent house, (3) the satisfaction o f building one’s own house according to one’s own personal interest and (4) the personal interest in, and supervision over, the actual construction and maintenance. 165 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Instead, cooperatives were adapted as vehicles for housing finance, which could be used for constructing housing on an individual ownership basis. According to the above Report: For these reasons, the existing building societies in this province are formed on the individual ownership system. Under this type, the soci ety ordinarily obtains funds from shares, loans from Government and deposits from members. The society is not allowed loans to borrow from anyone except from the Government. These funds are utilized in giving loans to members for the specific purpose of building houses and are given on the security of site or sites belonging to the borrower together with the proposed building and appurtenances thereon. The construction of houses is ordinarily undertaken by the members them selves. (Government of Madras, 1948, p. 61) 5.2. Post-Independence Period Cooperatives’ activities in Madras state7 grew quickly following Independ ence. State government encouraged cooperatives “by according preferential treatment to them in the matter of procurement of materials and land and by making available to societies their expert advice and guidance in the fulfilment of their objects” (Reserve Bank of India, 1950, p. 57). Although building materials (iron, timber, and cement) were in short supply in the post War period, cooperatives were given preferential allo cation in certain government quotas. In the years immediately following Independ ence, cooperatives received direct state support and were promoted as mechanisms for housing finance and construction. In terms of housing finance, Cooperative Building Societies8 got a boost between 1945-46 and 1949-50 (Table 5.2). Their number in creased from 113 to 193 between 1945-46 and 1949-50. During the same period, their 7 Following Independence, Madras province was called Madras State. In 1956, the Telugu speaking region o f Ma dras state was separated into the state o f Andhra Pradesh. In 1968, the remainder o f Madras State was named Tamil Nadu. Chennai is the capital o f Tamil Nadu State.. 8 Ordinary Building Societies were later called Cooperative Building Societies. 166 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. membership almost tripled from 4,027 to 11,697, and their share capital quadrupled from Rs. 0.67 million to Rs. 2.77 million (Reserve Bank of India, 1950, p. 51). Gov ernment loans outstanding with these cooperatives increased nearly eightfold, from Rs. 0.52 million in 1945-46, to Rs. 4.21 million in 1949-50. Cooperative Building So cieties usually on-lent government loans to members after adding a percentage fee for overhead charges. Table 5.2. Cooperative Building Societies in Madras State, 1945-46 to 1949-50 1945-46“ 1946-47“ 1947-48“ 1948-49“ 1949-50b Number of Coopera tives 113 128 139 174 193 Number of Members 4,027 6,657 11,313 11,697 Paid-up Share Capital (’000 rupees) 673 1,624 788 1,904 2,772 Government Loans Outstanding (’000 rupees) 517 443 1,020 2,559 4,211 Sources: (a) Reserve Bank of India (1950, p. 51); (b) Government of Madras (1951). Until the 1970s, the state government supported activities of housing coopera tives in two ways. First, it continued the direct financial support to existing primary housing cooperatives. Second, it included cooperatives as a mechanism for imple menting some of the government housing schemes. In this context, a range of new types of housing cooperatives emerged along with the older ones. Four schemes were implemented that included cooperatives as a part of the implementing mechanism: State Urban Housing Scheme, Rural Housing Scheme, Low Income Group Housing Scheme, and Middle income Group Housing Scheme. The State Government spon sored the first two schemes. The Central Government sponsored the last two schemes, but the State Government implemented them. 167 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Under the State Urban Housing Scheme, loans were made available to coop eratives to the extent of 80 percent of costs of housing construction.9 It was imple mented through five types of cooperatives: Cooperative Building Societies, Coopera tive House Building Societies, Cooperative House Construction Societies, the Coop erative Township, and the Cooperative Tenancy Housing Societies. Similar to Urban Housing Scheme, the Rural Housing Scheme envisaged loans upto 80 percent of costs of housing construction to rural cooperatives; however, loans were extended upto 87.5 percent if agricultural land were used for additional collateral.1 0 Rural Housing Scheme was carried out through Cooperative Rural Housing Societies and Taluk Co operative Housing Societies. Harijan Cooperative Housing Societies were introduced for specifically catering to the “backward class.” Thus a range of types of coopera tives emerged since the 1950s. Table 5.3 lists the types of housing cooperatives that are functioning in Tamil Nadu. The table also provides the main activities in which the cooperatives are involved, and their reclassification according to the three types of cooperatives (i.e. Finance, Building, and Tenure Cooperatives) identified in Chapter 2. A brief description of each type of cooperative follows below. 9 A loan ceiling was also prescribed, based on the schemes. Thus Low Income Group Housing Schemes prescribed a maximum o f 8,000 rupees per house, and were directed towards persons whose annual income did not exceed 6,000 rupees. Middle Income Group Housing Schemes prescribed a maximum o f 16,000 rupees per house, and were directed towards persons whose annual income was between 6,000 rupees and 12,000 rupees. Interest rates for M iddle Income Group Housing Schemes were also higher than the Low-Income Group Housing Scheme. 1 0 Like Urban Housing Schemes, ceiling was also placed on loans availed through Rural Housing Scheme. The ceiling amount was lower (5,000 rupees per house). 168 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 5.3. Types of Housing Cooperatives in Tamil Nadu and their Principal Activities Co-op Type: Building Co-ops Finance Co-ops Tenure Co-ops Principal Principal activities House Con struction Site Deve lopment Loan provision Collective Mainte nance Hire Purchase Co-op Ac tivity Cooperative Building Societies (Ordinary Building Societies) Low Low High Finance Cooperative House Building Societies High High Low Building Cooperative House Construction Societies High High Low High Building Industrial Workers Housing Societies Medium High Finance House Mortgage Banks (Cooperative Housing Societies) Very High Finance Cooperative House Sites Societies Very High Building Rural Housing Societies Medium High Finance Taluk Cooperative Housing Societies Medium High Finance Harijan Cooperative Housing Societies High Finance Cooperative House Services Societies Very High Tenure Cooperative Township High Tenure Cooperative Tenancy Housing Societies Very High Tenure Notes: (i) The meaning of Building, Finance, and Tenure Cooperatives is given in Chapter 2. (ii) I have provided the grading of High, Medium, and Low according to the main activities in which the cooperatives are involved. The last column indicates the principal activity according to my classification, i.e. Building, Finance, or Tenure Cooperative (see Chapter 2). A blank cell indicates that the cooperative does not perform the activity. Os SO Cooperative House Building Societies are similar to Cooperative Building So cieties in extending long term finance and purchasing house sites on behalf of their members. However, unlike Cooperative Building Societies, Cooperative House Building Societies do not pass on loans to members—loans are mainly used by coop eratives themselves to undertake construction on behalf of their members. As Reserve Bank of India (1950, p. 55) noted, in Cooperative House Building Societies, “the ob ject is not only to effect economies in the cost of construction and bulk purchase of materials but also to secure the services of technical staff for planning and building of houses which will be beyond the easy reach of individual members.” Thus, they are mainly Building Cooperatives. By 1950, there were 68 such cooperatives in Madras state. Cooperative House Construction Societies acquire sites, construct houses and rent them to members on hire purchase basis (i.e. each member pays the purchase price in installments). The Cooperative owns the house until all installments are paid; after full payment, the member becomes owner of the house.1 1 Thus, they are Build ing Cooperatives. By 1950, there were 13 Cooperative House Construction Societies in the Madras State. Cooperative Industrial Workers Housing Societies were essentially introduced to undertake the central government sponsored Subsidized Industrial Housing Scheme of 1952. Under the scheme, cooperatives of industrial workers obtain government loans to the extent of 65 percent of construction; they are also given subsidy to the ex tent of 25 percent, and the rest 10 percent is borne by workers. The scheme did not 1 1 Reports o f Registrars classify these cooperatives as “co-partnership” type, since land and housing are collectively owned by the cooperative until repayments are made. I however argue that they are not strictly “co-partnership” type Tenure Cooperatives since the cooperative does not have perennial ownership o f the house. Further, since their primary objective is to develop land and construct housing, they are better classified as Building Cooperatives. 170 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. gain popularity in the initial years, since even the investment of 10 percent was be yond the means of workers. I have classified Industrial Workers Housing Societies as Finance Cooperatives since they mainly provide funding to workers in Tamil Nadu; however, they could also be involved in housing construction. House Mortgage Banks were essentially started as secondary housing finance mechanisms, funded through the state level Apex Cooperative (i.e. the state govern ment did not fund them directly). Now called Cooperative Housing Societies, they on- lend money to their members for house construction and repairs. Unlike Cooperative Building Societies, they do not concern themselves with procurement of land on be half of their members. Four House Mortgage Banks came up in the Madras state by 1950, two of which operated in Chennai. Coonerative House Sites Societies obtain land, develop it, and allot it to mem bers after subdivision. Their main object is to develop land for housing, and they are not directly involved in housing construction. Members construct housing on their own. Thus, they are classified as Building Cooperatives. Cooperative House Sites Societies were introduced essentially as one of the vehicles for implementing State Urban Housing Schemes. Cooperative Rural Housing Societies are similar to Cooperative Building So cieties (in terms of advancing loans, and procuring land on behalf of their members), but their operations are limited to rural areas. Thus they are mainly Finance Coopera tives. They borrow money from the State Government, and on-lend to members for housing construction. Housing constructed by members are usually on individual ownership basis. Taluk Cooperative Rural Housing Societies were initiated in 1976 to imple ment the Village Housing Project Scheme sponsored by the Central Government. Un- 171 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. der the scheme, Taluk cooperatives assist villagers in constructing or improving their houses, and in providing house sites to landless agricultural workers. Taluk Coopera tives place special emphasis on including Adi-Dravidars (a “backward” caste in Tamil Nadu) by providing them with interest free loans for share capital. The activities of Taluk Cooperatives increased since 1976, partly due to increase in the state govern ment’s emphasis on using them as vehicles for rural housing. Harijan Housing Societies were initiated by the State Government to imple ment the Central Government sponsored Harijan Housing Scheme of 1956-57. These cooperatives primarily cater to “lower” caste households, who are provided with sub sidies and loans to construct houses.1 2 Thus, they are mainly Finance Cooperatives. Although Harijan Housing Societies have been active principally in rural areas, a few such cooperatives have also been formed in urban areas (including Chennai). Cooperative House Services Societies were initiated in early 1970s as collec tive organizations for maintenance and management of common properties in multi- storied housing complexes (tenements) constructed by Tamil Nadu Housing Board (TNHB). Thus, they are mainly Tenure Cooperatives. They were formed as per the lease-cum-sale agreement of TNHB with the allottees. They were, however, unsuc cessful, and the Registrar liquidated all such cooperatives in 1986. Their assets and liabilities were transferred to associations of the allottees (Thirunarayan, 1990, p. 26). Cooperative Townships have been in operation since pre-Independence. This type of cooperative usually consists of a large colony, and requires large parcels of land. The cooperative constructs and allots houses to members on hire purchase basis; it also grants loans to members for individual house construction. Unlike other types 1 2 Harijans were to be provided houses at an estimated cost o f 750 rupees each under the scheme. The Central gov ernment provided a subsidy o f 550 rupees per house, and the State Government provided loans to the extent o f 150 rupees per house. 172 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. of cooperatives, it undertakes municipal functions, like provision and maintenance of civic amenities and public utilities. Thus, Cooperative Townships are mainly Tenure Cooperatives. The number of this type of cooperatives is small since it requires large parcels of land (two such cooperative townships exist in Tamil Nadu, which are Kat- padi and Bhavanisagar). Cooperative Tenancy Housing Societies were initiated in 1954. They construct housing units and rent them to members. Ownership of housing rests permanently with the cooperative, and members have occupancy rights to their house. Thus, this type of cooperative is a true Tenure Cooperative. Loans availed under the Low In come Group Housing Scheme were made available to this type of cooperatives for housing construction. However, the activities of this type of cooperative did not in crease since their inception, and no such cooperative emerged in Chennai. Table 5.4 indicates the growth of the various types of housing cooperatives be tween 1949-50 and 1993-94 in Tamil Nadu. As the table shows, the total number of housing cooperatives approximately doubled every decade until the mid 1960s; since the 1970s, there was little growth in number of cooperatives. Indeed, the number of cooperatives even started to reduce since the mid 1980s. By 1999-2000, there were 1,033 housing cooperatives in the state. Figure 5.1 shows the urban-rural distribution of housing cooperatives. It is clear from this figure that the urban cooperatives out numbered rural cooperatives. However, the reduction in the number of primary coop eratives does not necessarily imply that there has been a parallel reduction in coopera tives’ activities in the state; also fewer rural cooperatives does not necessarily imply low rural activity. 173 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 5.4. Growth of Housing Cooperatives in Tamil Nadu, 1949-50 to 1993-94 1949 -50a 1954 -55a 1959 -60a 1965 -66a 1973 -74b 1975 -76b 1979 -80b 1984 -85b 1989 -90b 1993 -94c Cooperative Building Societies 193 124 174 314 415 378 363 442 373 342 Cooperative House Building Societies 68 38 44 133 86 119 151 183 205 173 Cooperative House Construction Societies 13 11 13 30 32 47 30 41 43 23 Industrial Workers Housing Societies 2 22 41 35 39 40 39 31 28 Cooperative Housing Societies 4 5 | 26 45 118 126 125 134 181 218 Cooperative House Sites Societies 66 80 84 78 93 108 72 Rural Housing Societies 29 209 364 380 350 398 257 Taluk Cooperative Housing Societies 100 104 165 165 Harijan Cooperative Housing Societies 21 21 27 29 15 9 4 2 Cooperative House Services Societies 19 113 109 Co-operative Township 2 1 1 1 2 2 2 2 1 1 Cooperative Tenancy Housing Societies 1 4 3 5 4 4 4 4 1 Others 1 1 11 34 Total 280 211 515 1,019 1,210 1,212 1,306 1,421 1,224 1,025 Sources', (a) Report on the working o f cooperative societies in the State o f Madras (respective years); (b) Tamil Nadu State Administration Report (respective years); (c) Registrar of Cooperative Societies (Housing). Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.1. Urban/ Rural Distribution of Housing Cooperatives in Tamil Nadu, 1949-50 to 1993-94. 1949-50 1954-55 1959-60 1 9 6 5 -6 6 1973-74 1975-76 197 9 -8 0 1984-85 1989-90 1993-94 Year I Urban I I Rural ■ ■ ■ Total Source: Same as Table 5.4. Number o f Cooperatives The growth of membership does not parallel the growth of housing coopera tives. Figure 5.2 shows the growth in cooperatives’ membership and urban/ rural dis tribution of the membership between 1949-50 and 1993-94. Clearly, the total mem bership of the cooperatives increased considerably from nearly 250,000 in 1975-76 to over 2.2 million in 1993-94. Membership of rural cooperatives also surpassed urban cooperatives since the mid 1980s. Indeed, the enormous growth in cooperative mem bership is due to increase in membership of rural cooperatives. The growth in number of cooperatives has been necessarily so due to direct government intervention in the functioning of cooperatives since the 1970s. Geographical constraints of non-overlap on cooperatives activities meant that members could join only such cooperative that is in their jurisdiction. Moreover, uniform spread of housing cooperatives, especially in the rural areas through Taluk Cooperative Housing Societies in each district, and gov ernment emphasis on use of cooperatives for rural housing implied a significant growth in rural membership of housing cooperatives. Emphasis on rural housing is clear from Figure 5.3, which shows the distribution of rural and urban housing con structed by primary cooperatives. As the chart shows, major part of the housing con structed by housing cooperatives has been in the rural areas since the mid 1970s. As Table 5.5 shows, Taluk Cooperative Housing Societies constructed the bulk of rural housing (or even housing built by cooperatives in general) since the mid 1980s. Table 5.5. Activities of Taluk Cooperative Housing Societies, 1979-80 to 1993-94 1979-80 1984-85 1989-90 1993-94 Number of Cooperatives 100 104 165 165 Membership 7,068 490,117 847,100 1,513,466 Number of Houses 0 29,116 30,000 34,283 Taluk cooperatives’ Contribu tion to Cooperatives’ Housing Stock 71.6% 58.4% 73.5% Source: Annual Report on the Working o f Co-operative Societies in Tamil Nadu (re spective years). 176 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.2. Growth of Membership in Housing Cooperatives, 1949-50 to 1993-94 100% ■ — -------------------------------------------------------------------------------------------------------------------------- ■ M l 1949-50 195 4 -5 5 1959-60 1965-66 1 9 7 3 -7 4 1975-76 1979-80 1984-85 1990-91 1993-94 Year < u 60% 6 1) c t f C 2 ,5 0 0 ,0 0 0 2,000,000 1,500,000 1,000,000 5 0 0 ,0 0 0 I Urban I I Rural ■ Total Source: Same as Table 5.4. Number o f Members Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.3. Houses constructed by Housing Cooperatives, 1954-55 to 1993-94 100% 80% < o 60% 0 0 c 3 a 0 ) o t-H 40% 20% 6 0 ,0 0 0 4 8 ,0 0 0 3 6 ,0 0 0 2 4 ,0 0 0 12,000 1 9 5 4 -5 5 1 9 5 9 -6 0 1 9 6 5 -6 6 1 9 7 3 -7 4 1 9 7 5 -7 6 1 9 7 9 -8 0 1 9 8 4 -8 5 1 9 9 0 -9 1 1 9 9 3 -9 4 Year I Urban I I Rural ■ Total Source: Same as Table 5.4. oo Number o f Houses Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.4. Distribution of Housing Cooperatives by Type in Tamil Nadu, 1949-50 to 1993-94 100% o 6 0 /o ^ 40% 1 9 4 9 -5 0 1 9 5 4 -5 5 1 9 5 9 -6 0 1 9 6 5 -6 6 1 9 7 3 -7 4 1 9 7 5 -7 6 1 9 7 9 -8 0 1 9 8 4 -8 5 1 9 8 9 -9 0 1 9 9 3 -9 4 Year 1 ,5 0 0 1,200 9 0 0 6 0 0 3 0 0 I Finance Cooperatives ) ) Building Cooperatives i i Tenure Cooperatives ■ Total Source: Same as Table 5.4. Note: Classification of Finance, Building, and Tenure Cooperatives is as per Table 5.3. SO Number o f Cooperatives Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.5. Distribution of Membership by Type of Housing Cooperatives in Tamil Nadu, 1949-50 to 1993-94 2 ,5 0 0 ,0 0 0 100% 2,000,000 80% 1 ,5 0 0 ,0 0 0 60% • - 0 ) o i -H £ 40% • - 1,000,000 5 0 0 ,0 0 0 20% - - 0% 1 9 4 9 -5 0 1 9 5 4 -5 5 1 9 5 9 -6 0 1 9 6 5 -6 6 1 9 7 3 -7 4 1 9 7 5 -7 6 1 9 7 9 -8 0 1 9 8 4 -8 5 1 9 9 0 -9 1 1 9 9 3 -9 4 Year ■ ■ Finance Cooperatives I I Building Cooperatives V /J w / A Tenure Cooperatives EH H Others Source'. Same as Table 5.4. Note: Classification of Finance, Building, and Tenure Cooperatives is as per Table 5.3. 00 o Number o f Members Although many different types of housing cooperatives were introduced in the State, the main activity of housing cooperatives has been to provide housing finance; their secondary activity has been to construct housing (including land development). Figure 5.4 indicates activities of housing cooperatives after reclassifying into Finance, Building, or Tenure Cooperatives (based on Table 5.3). As the figure shows, between 1949-50 and 1993-94, over 60 percent of the cooperatives have been functioning as Finance Cooperatives. Less than 30 percent of the cooperatives have been functioning as Building Cooperatives. Tenure Cooperatives hardly emerged. A similar picture emerges if we examine the distribution of membership by type of Housing Coopera tives (Figure 5.5). The percentage of members of Finance Cooperatives increased on average since 1949-50, and the percentage of members in Building Cooperatives re duced since then. Thus, housing cooperatives in Tamil Nadu have been mainly Fi nance Cooperatives under various forms. Similar to Tamil Nadu, housing cooperatives in Chennai have been mainly been functioning as Finance Cooperatives, and secondarily as Building Cooperatives. Finance Cooperatives are mainly in the form of Cooperative Building Societies and Cooperative Housing Societies. Building Cooperatives have been mainly in the form of Cooperative House Sites Societies, Cooperative House Building Societies, and Co operative House Construction Societies. Unlike the phenomenal growth of housing cooperatives in Mumbai, there has been modest growth of cooperatives’ activities in Chennai in the post-Independence period. The total number of housing cooperatives in 1990-91 was 159 (Table 5.6). Of these, 67 were Finance Cooperatives and 92 were Building Cooperatives. Although Building Cooperatives out-number Finance Cooperatives, Finance Cooperatives are more active than Building Cooperatives. Finance Cooperatives are more active due to 181 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. their lending activities for both house construction and for repairs. With the assistance of the Apex cooperative (Tamil Nadu Cooperative Housing Federation, TNCHF), co operatives financed construction of an average of about 789 houses annually between 1 T • 1991 and 1996. According to the survey of non-slum households in 1991 conducted by Economist Group, percentage of households borrowing from cooperatives was higher than other formal financial institutions for housing construction (21.4 percent in Chennai City; 18.1 percent in Metropolitan area) (Economist Group, 1991). Table 5.6. Housing Cooperatives in Chennai, 1990-91 1990-91 Cooperative Building Societies 47 Cooperative House Building Societies 30 Cooperative House Construction Societies 15 Cooperative Housing Societies 17 Cooperative House Sites Societies 47 Cooperative Industrial Housing Societies 3 Total 159 Source: Registrar of Cooperatives (Housing), Chennai. The activities of Building Cooperatives reduced in Chennai after the initial spurt in their growth during the early years after Independence. For example, one such cooperative—the Madras Cooperative House Construction Society—was started in Chennai in 1947-48, which was instrumental in establishing several large housing colonies covering nearly 400 acres in the city and suburban areas (e.g. in Adayar (Gandhinagar), Raja Annamalaipuram, Mylapore, Ayanavaram, and so on) (Govern ment of Madras, 1951). Large scale land development through cooperatives had re duced drastically in the 1990s (MMDA, 1995, p. 80). Many of the Building Coopera tives have been dormant due to lack of availability of land in the metropolitan area. As Table 5.7 shows, housing cooperatives delivered about 2,489 plots during a ten 1 3 Based on Annual Reports o f TNCHF. 182 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. year period between 1981 and 1991, which averages less than 250 plots per year. Also housing cooperatives hardly constructed new housing (although they financed it). Table 5.7. Sector wise delivery of Housing Units in Chennai Metropolitan Area, 1981-1991 Number of Plots Number of Houses Number of Flats Private Sector 1,43,974 1,00,163 28,863 Public Sector Tamil Nadu Housing Board 30,331 5,347 7,278 Tamil Nadu Slum Clerance Board 4,519 656 12,402 Others 405 10,150 Cooperatives 2,489 - - Total 1,81,313 1,06,571 58,693 Source: School of Architecture and Planning (1993), p. 48. 5.3. Chennai’s Institutional Context, 1947-2000 Why did the activities of housing cooperative in Chennai grow modestly dur ing the post-Independence period? I argue that in the context of Chennai’s particular local historical conditions of urban growth and demography, both the local housing market and the institutional framework of Mumbai restrained the activities of housing cooperatives. In terms of the local housing market, the demand for collective organ izational vehicles like cooperatives hardly emerged in the context of individual plotted development. Multi-family housing is a relatively new phenomenon (starting with public housing in 1970s, and private housing since the late 1980s), and is owned in the form of condominiums, rather than cooperatives. In terms of institutional framework, state intervention restrained autonomy of cooperatives. In this context, cooperatives evolved distinctively in the directions prescribed by the state, viz as financial mecha nisms for rural housing. Cooperatives in Chennai are also mainly used as financial mechanisms for construction. Although cooperatives have been utilized for imple menting state schemes, few such schemes have been carried out through cooperatives 183 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. in Chennai. Further, the private sector hardly adapted cooperatives, and NGOs hardly used cooperatives for collective action. 5.3.1. Chennai’s Urban Growth and Demography Indian census distinguishes three areas in Chennai: Chennai City, Chennai Ur ban Agglomeration, and Chennai Metropolitan Area. Chennai city is the city proper, which also contains the historical core. Chennai City progressively grew from 128.83 square kilometers at the time of Independence to 170 square kilometers in 1978. The Chennai Urban Agglomeration consists of Chennai City and patches of urbanized ar eas in the suburbs. The Chennai Metropolitan Area is a much larger planning jurisdic tion under Chennai Metropolitan Development Authority. The Chennai Metropolitan Area covers about 1177 square kilometers, of which about 51 percent is rural. Although Chennai is a major metropolitan city for South India, its economy during the post-Independence period did not experience a dramatic growth like that of Mumbai. According to Alan Turner, who was the principal author of the structure plan of Madras Metropolitan Area in 1980, “the growth of the economy of Madras is characterized by the growth of the small and marginal sectors, with a relatively low contribution to income generation” (Turner, 1980, p. 41). Until quite recently in the 1990s, large-scale industrial development in the city had been quite limited. Industries at the time of Independence included textile mills, railway factories, electrical machin ery, and tanneries; since then, automobile (including bicycle) and oil refinery indus tries grew in the Chennai Metropolitan area. Small scale industries in metal, rubber, and plastic goods also grew due to state government industrial policies, which empha sized such industries. Informal sector employment rose from 22 percent in 1979 to about 56.5 percent in 1986, and is projected to reach nearly 70 percent by 2011 (MMDA, 1995, pp. 45 and 46). 184 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Chennai’s industrial base hardly diversified as compared to Mumbai, and per capita industrial investment in Chennai remained low (almost one-sixth of Mumbai). Per capita income in Chennai has also been historically the lowest among the four ma jor metropolitan cities (per capita income of Chennai in 1970-71 was around half of that of Mumbai) (Turner, 1980). The major strength of Chennai has been in financial services (especially banking). Also, globalization during the 1990s essentially meant growth of automobile industries (especially South Korean car manufacturing plants) and information technology services. Major infrastructure improvements (construc tion of bus terminal, relocation of wholesale markets, and construction of Mass Rapid Transport System (MRTS)) were also carried out over the last decade. Further, al though Chennai seaport and airport are major gateways for South India, the volume of cargo and international passenger transit through Chennai ports are much lower than that through Mumbai’s ports. In the context of Chennai’s sluggish economic growth, its population has been smaller than the other three major metropolitan cities of Mumbai, Delhi, and Calcutta (some population estimates even place Bangalore ahead of Chennai for 2001). The population of Chennai surged following World War II, at an annual average rate of 6.5 percent (Table 5.8). The post-Independence peak growth period was between 1961 and 1971, when the growth rate reached 4.71 percent for Chennai City, and over 6 percent for the Urban Agglomeration and the Metropolitan Area. The growth rate de creased since 1971, and the population of Chennai Metropolitan Area for 2001 was estimated at 7.5 million. 185 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 5.8. Post-Independence Growth of Chennai’s Population, 1951 to 2001 Chennai City Chennai Urban Agglomeration Chennai Metropolitan Area Population Growth Rate (%) Area (sq. km.) Population Growth Rate (%) Area (sq. km.) Population Growth Rate (%) Area (sq. km.) 1951 1,427,420 6.50 128.83 1961 1,749,600 2.26 128.83 1,944,502 226.07 2,134,202 1177.0 1971 2,572,967 4.71 128.83 3,169,930 6.30 530.77 3,419,295 6.02 1177.0 1981 3,276,622 2.73 170.00 4,289,347 3.53 571.93 4,628,879 3.54 1177.0 1991 3,795,028 1.58 170.00 5,361,468 2.50 571.93 5,917,000* 2.78 1177.0 2001* 4,947,000 3.04 170.00 571.93 7,522,000 2.71 1177.0 Source: MMDA (1995), p. 9. Note: represents estimates by MMDA (1995). Table 5.9. Post-Independence Growth of Households and Housing Stock in Chennai, 1961 to 1991 Households Housing Stock Households per House Chennai City Chennai U.A Chennai Metro Area Chennai City Chennai U.A Chennai Metro Area Chennai City Chennai U.A Chennai Metro Area 1961 350,043 147,694 2.37 1971 444,788 142,041 649,499 338,414 108,196 506,456 1.31 1.31 1.28 1981 629,697 204,591 904,192 620,241 201,975 891,214 1.02 1.01 1.01 1991 748,245 1,057,092 1,166,623 794,322 0.94 oo a s Source: MMDA (1991). Similar to the pre-Independence period, much of the population growth in the post-Independence period has been due to migration from nearby rural districts within the state (Turner, 1980, p.35). The proximity allowed residents to have intensive ru ral-urban exchange.1 4 Migration from nearby rural areas of the state reinforced the traditional rural social order of the Chennai. Further, similar to pre-Independence pe riod, families, rather than single men, migrated to the city. Proximity to home districts allowed women to have nexus with the hinterland. The female to male ratio has been traditionally high in Chennai as compared to Mumbai where migration consisted of male laborers. Thus the demand for housing in Chennai has been oriented towards accommodating families, rather than accommodating single men, which is characteris tically different from that of Mumbai. 5.3.2. Chennai’s Local Housing Market Following war time controls on supply of building materials, building con struction pace had slowed even as the population continued to increase. Between 1941 and 1946, only about 2100 houses were added to the housing stock. Thus, there was an increasing pressure on existing housing stock, with a corresponding increase in house rents and landlord-tenant disputes. Similar to Mumbai, rent control was adopted in 1946 as temporary measures, but it has been extended periodically and is in effect in central city areas.1 5 However, unlike Mumbai, rent control does not form the core of housing policies affecting Chennai’s local housing market. 1 4 According to 1961, 1971 and 1981 census data, in-migration contributed 37.2 percent, 31.5 percent, and 30.6 percent o f the total population o f Chennai City. During each o f these census years, in-migration from Tamil Nadu formed nearly 70 percent o f the total migration. Further, migrants from nearby districts o f Chingleput, North Arcot and South Arcot formed nearly 52.8 percent and 45.8 percent o f the total migration during 1961-71 and 1971-81 respectively. Similar pattern can also be observed for migration into Chennai Urban Agglomeration. 1 5 Madras Law Journal Office (1997), The Tamil Nadu Buildings (Lease and Rent Control) Act, 1960. 187 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Chennai’s local housing market characteristically reflects its economic growth and demography. In the context of slow economic growth, the city did not become as modernized as Mumbai. Unlike Mumbai, where the underlying rural character was erased, the underlying rural character continued to be maintained during the post- Independence period. This implied that the preference for individual housing re mained strong. In spite of the higher population density with increase in population, the increasing demand for housing did not result in vertical expansion. Housing de velopment remained essentially horizontal until the 1970s. This is partly explained by the housing supply—the supply of housing stock almost caught up with the new household formation between 1971 and 1981. As Table 5.9 shows, the gross residen tial density of Chennai dropped from about 1.3 to about 1.01 across Chennai.1 6 Much of the housing development remained horizontal until the mid 1970s and houses were individually owned. Public, private, and cooperative investments in housing were also directed towards individual plotted development. For example, the City Improvement Trust (CIT), which was formed in 1946, acquired land on the pe ripheral areas of the city, and developed it as open developed plots for allocation. Ac cording to Schenk, Holder, and Mulders (1989), about 28 Open Developed Plot Schemes were developed in Chennai between late 1940s and the early 1960s. Several residential neighborhoods including the CIT colony, Mandaveli, and Pushpanagar were developed on this basis. CIT’s activities were taken over by the Madras Housing Board (later renamed Tamil Nadu Housing Board) in 1961. Private and cooperative housing development were also on an individual house/plot ownership basis. 1 6 This statistic should be view ed with some caution, since most other reports from M M DA highlight that housing supply falls short o f demand in Chennai. However, M M DA makes an interpretation mistake in arriving at the con clusion. Figures in Table (1) M M DA (1995, p. 68) do not coincide with the interpretation in Paragraph 3 o f the same page. Growth rate o f housing units is more than that o f households as per the table. 188 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Public housing schemes were also initially oriented towards individual plotted developments. For example, under the World Bank assisted Madras Urban Develop ment Project (I and II), sites and services schemes were essentially based on individual plotted developments for all income groups. Subsequent sites and services schemes carried out under the Tamil Nadu Urban Development Project, were also carried out in a similar fashion. Together, these projects constituted the major part of public agency’s involvement in land development between 1977 and 1989—about 13,457 residential plots were developed through these projects.1 7 Ownership of plots was given on an individual basis to households. The sites and services schemes in Chennai were unlike those in Mumbai, where plots were developed for cluster housing, and tenure was provided on a collective basis through cooperatives. Consideration of co operative ownership hardly arose in the case of Chennai.1 8 In spite of the Chennai’s overall low residential density and predominantly plotted development, it did not imply that households lived in single-family housing (i.e. each household in a separate house on an individual plot), and that overcrowding was not a problem. Overcrowding continued especially for low-income households. First, in the formal sector, house owners typically rented one floor or a part of the house to other households to either supplement their income or to recoup a part of their expenses on housing construction and maintenance. In Chennai, such sharing or renting of premises is called living in a “portion” of the house. According to the sur vey conducted by the Economist Group (1991), about 68 percent and 59.9 percent of 1 7 Calculated from National Institute o f Urban Affairs (1993), pp.62-64. 1 8 Several studies have been conducted on the sites and services schemes in Chennai. These studies examine af fordability, implementation (National Institute o f Urban Affairs, 1993), replicability, and gentrification. Few stud ies, however, examine w hy collective self-help was not even considered for implementing the schemes. The excep tion is Robben and Stuijvenberg (1986), who argue that m ost households did not engage in self-help; rather, they employed contractors for housing construction. 189 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the households surveyed in Chennai City and Chennai Agglomeration Area respec tively lived in a portion of the house. Thus, rental housing adapted to the individual plotted development.1 9 Second, informal housing (i.e. slums) supplemented the for mal housing supply. Slum population has constituted around 30 percent of the Chen nai’s population since 1981 (albeit perceptibly less than that of Mumbai, where slum population exceeds 50 percent). Multi-family housing is a relatively new phenomenon for Chennai, which started with public housing provided by Tamil Nadu Housing Board (TNHB) and slum resettlement schemes of Tamil Nadu Slum Clearance Board (TNSCB) since the mid 1970s. While TNHB provided multi-family housing in the form of flats (apart ments), TNSCB relocated slum-dwellers in tenements. Multi-family housing in multi- storied buildings came to be increasingly viewed as more efficient use of land. Be tween 1981-91, TNHB constructed 5,347 houses and 7,278 flats; TNSCB constructed 12,402 tenements (typically in four storied buildings) and 656 houses for relocating slum-dwellers or to re-house them. While houses and flats constructed by TNHB were predominantly for middle and high income households (nearly 59 percent of 20 houses and 86 percent of the flats constructed were for these income groups), tene ments constructed by TNSCB were primarily for low-income and economically weaker groups. Although TNHB experimented with introducing cooperatives (Coop erative House Services Societies) for collective management, it did not succeed, and all such cooperatives were liquidated in 1986. 1 9 Overall, about 60 percent o f the households lived in rental housing in Chennai in 1981. About 68.06 percent and 62.45 percent o f households in Chennai City and Chennai Urban Agglomeration respectively lived in rented hous ing (in rural areas, about 14.15 percent live in rented housing) (M M DA, 1995, p. 69). 20 Calculated from School o f Architecture and Planning (1993, p. 50), Table 3.3. 190 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Private agencies did not immediately follow the lead of public agencies in con structing multi-family housing. Individual plotted development continued to be the preferred mode (especially among middle and higher income households). Institu tional constraints, such as Development Control Rules (which govern the density of population and built-up area) were also oriented towards traditional housing in the form of individual plotted development. According to the survey conducted by Economist Group (1991), 13.2 percent of the households surveyed in Chennai city lived in a flat; only about 0.1 percent of the households surveyed in Chennai Urban Agglomeration lived in flats. Households living in individual houses accounted for 18.7 percent and 40.0 percent of those surveyed in Chennai City and Chennai Urban Agglomeration respectively. Thus living in multi-family dwelling units had not yet become popular in suburban areas. Private construction of flats by developers increased considerably since 1988 (Figure 5.6). This was principally due to the increasing cost of land in Chennai City. Multi-storied buildings replaced older buildings and bungalows in highly dense areas of the city like Triplicane, Adayar, Mylapore, Nungambakkam and so on (most of the buildings were constructed on two or more contiguous plots obtained after demolish ing the bungalows) (Dattatri and Anand, 1991, p. 110). The growth in apartment con struction is also indicated by the steady increase in number of permissions for con structing them—in 1988, 1992, and 1995, permissions for apartments numbered 561, 827, and 987 respectively. The demand for such housing in the private sector rose es pecially for the salaried class since there were more venues for availing loans through various private sector housing finance institutions. 191 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.6. Construction of Flats/ Tenements in Chennai, 1981-82 to 1990-91 7000 6000 5000 4000 u X > E 3 z 3000 2000 1000 0 Source: School of Architecture and Planning (1993). I I ■ 1981-82 1982-83 1983-84 1984-85 1985-86 1986-87 1987- 1988-89 1989-90 1990-91 Year I T N H B (Flats) I 1 TN SC B (T enem ents) ■ Private (Approved Flats) As the construction of apartments proliferated, formal institutions for apart ment ownership also came into effect. First, Development Control Rules were relaxed to suit construction of multi-family housing. MMDA’s master plan for the city (1995) also explicitly favored group housing as a means for efficient use of land. Private lob bies like Tamil Nadu Flat Promoters Association, an independent organization com prising mostly of developers, also became more active in streamlining procedures, re laxing certain development constraints, and so on. Second, The Tamil Nadu Apartment Ownership Act was passed in 1995 to provide for ownership of an individual apartment in a building and to make such apartments heritable and transferable immovable property. Apartment owners were also required to form an association for management and maintenance; it could be reg istered either under the Tamil Nadu Cooperative Societies Act (1983) or the Tamil Nadu Societies Registration Act (1975). However, unlike Mumbai, few such collec tives have been registered as cooperatives. Support institutional structure for the func tioning of such associations (e.g. administrative support, dispute resolution, etc.) also hardly evolved. 5.3.3. Chennai’s Local Institutional Framework While there has been little demand for collective organizations like coopera tives in the context of Chennai’s local housing market, the institutional framework also inhibited the growth of housing cooperatives. Government control over the func tioning of cooperatives, which was already pronounced due to the broad range of powers accorded to the Registrar’s Office, was further increased with the passage of Tamil Nadu Cooperative Societies (Special Officers) Act in 1976. Cooperatives effec tively lost their autonomy, and were co-opted by the state. 193 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In Tamil Nadu, Office of the Registrar has been traditionally strong since the cooperatives were started in the state. The Registrar is responsible for registering new cooperatives, and verifying their financial viability. The Registrar could audit their accounts, oversee their elections (and conduct them periodically), arbitrate disputes (between members and cooperative boards), execute member petitions, and amalga mate cooperatives. The Registrar could also supersede the cooperative board and its decisions by appointing a government nominee (Special Officer) to manage the coop erative. The Registrar could prosecute board and ordinary members in case of misap- 91 propriation, liquidate cooperatives, and cancel their registration. Besides the sweep ing powers vested with the Registrar, funds available through state and central gov ernment schemes were also routed through the Registrar’s office. The state govern ment could thus interfere in the daily activities of cooperatives through the immense powers vested with the Registrar’s office. Cooperatives have had little autonomy in their decision-making and daily operations. Consequently, cooperatives developed in such areas where the state government laid emphasis through its schemes. Few coop eratives developed outside the ambit of the state government. Dependence of cooperatives on the state government became more critical with the Tamil Nadu Cooperative Societies (Appointment of Special Officers) Act. The Act was passed in 1976 “for a limited period in the public interest for the purpose of safeguarding the interests of the members or depositors, for ensuring the proper utilisation of the amounts guaranteed by the Government, and for improving the effi ciency of the administration.” It prohibited legal proceedings against any government actions taken in “good faith” under the Act. The A ct provided for governm ent control 2 1 The decision to terminate cooperatives is usually based on audit results, or on the application o f more than 75 percent o f the members to wind-up the cooperative. 194 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. over cooperatives by rescinding the posts of elected board members, presidents and chairpersons of cooperative societies. A government nominee (called the Special Of ficer) took their place. The Special Officer was vested with enormous power in con trolling the functioning of the cooperative: [The Special Officer] ... shall subject to the control of the Registrar and to such directions as he may, from time to time, give, have power to exercise all or any of the functions of the committee or board, as the case may be, or of any officer of the scheduled co-operative society and to take such action as may be required in the interest of such scheduled co-operative society.2 2 Since the scope of the Act was expected to be for a limited period, the Special Officer was also required to constitute a new committee or board who could take over the management after s/he left. The Act was implemented as a measure for limited time when the state was under Central government rule. However, successive state governments found it expedient to continue with the Act for political control over co operatives. Hence the Act was periodically extended and Special Officers continued to manage cooperatives with directions from the Registrar. The Act was in effect for 20 years until 1996. During this period, cooperative board elections were hardly held. Cooperatives thus had little autonomy in their operations. Liberalization of economy at the national level did have an effect in mitigating some of the government controls. The state government came under increasing criti cism for its overt interference in the daily affairs of cooperatives. A frequent point of contention was the restrictions on autonomy of cooperatives. While Special Officers intervened in the daily functioning of cooperatives, they had no direct accountability to cooperatives’ members. Their appointment had been frequently made on political considerations. In response to such criticisms, the state government rescinded the law 22 Clause 4.2 o f the Act. 195 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. in 1996 and held elections for most cooperatives. Yet, the government’s intransigence to give up control over cooperatives persisted due principally to political exigencies. Although there have been moves to liberalize cooperative laws, the state government still retained substantial control over the functioning of cooperatives. In May 2001, the Special Officer’s Act was re-invoked, after which control over cooperatives was transferred again to Special Officers. One redeeming feature of the administrative structure of Registrar’s office is the recognition of housing as a specialized activity. Until 1955, all types of coopera tives were under the control of the Registrar of Cooperative Societies, who was re sponsible for cooperatives in all sectors (including housing).2 4 Since 1956-57, coop eratives were slowly transferred to the control of other departments and statutory bod ies (which specialized in such sectors), conferring the powers of Registrar on the Heads of Departments or Boards. In the case of housing cooperatives, control was shifted to Tamil Nadu Housing Board (TNHB) in 1966. From 1966 to 1979, the Chairperson of TNHB acted as the ex-officio Registrar of Cooperatives (Housing), and was assisted by deputy registrars from the cooperative department. Cooperatives could now be used as an integral mechanism for implementing TNHB schemes. Since 1979, however, the administrative structure changed again, and housing cooperatives were separated from TNHB and placed under a functional Registrar of Cooperatives (Housing). In spite of the separation of housing from other types of cooperatives, the potential advantage that could have been gained from the separation has been hardly realized. Housing cooperatives come under the purview of the Cooperative depart 23 Government o f Tamil Nadu, Housing D em and No. .32, P olicy N ote 2001-2002. 24 Information in this paragraph has been culled from Annual Reports o f the Registrars and Tamil Nadu State Ad ministration Reports. 196 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. ment, which is highly bureaucratized with few personnel specialized to deal with technical issues of housing. Direct state control on cooperatives implied that the state could steer coopera tives in the directions where the state had an emphasis. First, cooperatives were em phasized primarily as financing mechanisms. In this emphasis, the role of coopera tives for land development did not get much priority. According to the Dattatri and Anand (1991, p. 126), Presently, co-operative efforts for land development for housing have come to a virtual halt, because of the ban on registration of housing co operative societies. The co-operative effort in housing is today con fined only to financing of the purchase of plot and construction of house by individuals or group of individuals. The boost given to Cooperative House Construction Societies and Cooperative House Building Societies in the initial years after Independence subsided. Land de velopment activities were carried out through other public housing and metropolitan agencies, such as Tamil Nadu Housing Board and Madras Metropolitan Development Authority. Second, cooperatives came to be increasingly used as mechanisms for rural housing, and less so for urban housing. Funding availed by TNCHF through central and state government sponsored schemes (e.g. Nehru Rozgar Yojana), including HUDCO came to be increasingly used to subsidize rural cooperatives. Taluk Housing Cooperatives, which were established in each district, became the prime vehicles for such rural housing. In the last decade, the activities of Taluk Housing Cooperatives have been more pronounced than other types of cooperatives. 1 9 7 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 5.4. Cooperative Institutional Structure in Chennai While the nature of local housing market precluded the demand for collective organizations like housing cooperatives, the local institutional framework also con strained the autonomous development of housing cooperatives. Housing cooperatives emerged mainly as Financial Cooperatives, and secondarily as Building Cooperatives. Tenure Cooperatives hardly emerged. The cooperative institutional structures that emerged—or for that matter, did not emerge—were also in response to such institu tional conditions. Institutional structure for financial support was strengthened in this context. Other institutional structures for providing administrative, legal, and proce dural support to cooperatives hardly emerged. Institutional support for land also faded over the years. The two-tiered institutional structure for finance reinforced the role of cooperatives as financial mechanisms. 5.4.1. Institutional Structure for Finance Until the 1970s, the state government financed various schemes carried out by housing cooperatives through the Registrar’s office. Almost all types of cooperatives received some form of government assistance (either as subsidies or as loans) since they were used as mechanisms of implementing state or central government sponsored housing schemes. House Mortgage Banks were an exception: in the initial years, they raised finances through deposits and savings from members, and through other sources—mainly from South India Cooperative Insurance Society Limited (a life in surance cooperative). The House Mortgage Banks came to play a significant role in the evolution of the Apex cooperative in Tamil Nadu. Although their operations had not increased during 1950s (only one more was established between 1950 and 1958), their member- 198 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ship and loan amount increased to a good degree.2 5 When life insurance companies were nationalized in the late 1950s, South India Cooperative Insurance Society ceased to function, and House Mortgage Banks faced a resource crunch. In this environment, the Madras Central House Mortgage Bank was established in 1959 as a secondary cooperative for mobilizing funds for primary House Mortgage Banks. Life Insurance Corporation (LIC) became the major lender for Madras Central House Mortgage Bank (with the state government acting as a guarantor for such loans). Since then, activities of primary House Mortgage Banks proliferated. In 1959 itself 29 new primary House Mortgage Banks came up, and there were 45 such banks by 1965-66 in Madras State. Until the early 1970s, there were thus two principal sources of housing finance for primary cooperatives: the state government (through various state sponsored schemes); and the Madras Central House Mortgage Bank (where the state was again involved as a guarantor). Thus, the state government was directly or indirectly in volved in financing primary cooperatives. Since 1970s, when the state retreated from direct involvement in housing and finance provision, budgetary allocations were insuf ficient to meet the loan requirements of primary cooperatives. By this time, the Re port o f the Working Group on Housing Cooperatives had also recommended the set ting up of an Apex cooperative to route housing finance. In this scenario, the Madras Central House Mortgage Bank came to be viewed as an appropriate organization to act as an Apex cooperative for mobilizing funding. Hence, it was designated as the Apex cooperative in 1972, and was renamed as Tamil Nadu Cooperative Housing Society Limited (TNCHS). Its scope of lending was ex panded from that of primary House Mortgage Banks to all types of primary housing 25 Between 1950 and 1958, membership o f the banks (including the two in Madras City) increased from 4612 to 7047; share capital increased from 0.62 to 0.93 m illion rupees; their loan amount increased from 0.46 m illion to 0.89 m illion rupees (unadjusted). 199 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cooperatives. Primary House Mortgage Banks were renamed as Cooperative Housing Societies. All types of primary cooperatives were required to be affiliated as members with the TNCHS in order to borrow. Government subsidies and finances were hence forth routed through the TNCHS. Although LIC became the major source of funding, TNCHS also started to raise money from other sources (General Insurance Corpora tion, Housing and Urban Development Corporation, cooperative banks, nationalized banks). State Government participated by contributing to TNCHS’s share capital, and by providing a guarantee for loans from sources like LIC. In 1988, TNCHS was re named as Tamil Nadu Cooperative Housing Federation (TNCHF). TNCHF and pri mary cooperatives became a significant source for housing finance in an environment where there were limited other sources (private housing finance institutions were yet in their infancy). With state government’s emphasis on rural housing, TNCHF’s lending pattern also changed. While housing constructed by TNCHF was predominantly in urban ar eas before 1977-78, the proportion reduced to nearly 24 percent in 1995-96 (Figure 5.7). In Chennai, the proportion of TNCHF loans made available to cooperatives re duced steadily from 13% to nearly 4% between 1984-85 and 1995-96 (Figure 5.8). The institutional structure for finance in Tamil Nadu is perhaps more devel oped than other states in India. The volume and amount of finance disbursed by TNCHF during the late 1990s was more than that disbursed by apex cooperatives in other states. It received several awards during the 1990s, including the “Best Man aged Apex Cooperative Housing Federation” in the country from National Coopera tive Housing Federation, and the “HUDCO National Award” for All Round Excellent Performance from HUDCO. 2 0 0 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.7. Proportion of Urban Housing Constructed with TNCHF funding 90% 80% 70% s 60% o g , 50% O 9 40% C Q -e p 30% 20% 10% 0% CN I C O ON N O ON l I T ) ON o ON I ON <N ON 00 i m oo t" - oo I NO OO OO oo ON O O I oo oo < N 00 On i 0 4 Os ON r- i oo o 00 O n OO ON 00 O n OO O s O O o o Year Proportion of Urban Housing Source: Annual Reports of TNCHF. to o Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 5.8. Proportion and Loan Amount disbursed by TNCHF to Primary Housing Cooperatives in Chennai 8 4 0 14% 7 2 0 12% 6 0 0 10% I 8% I 6% 4 8 0 3 6 0 2 4 0 4% 120 2% 0% 82 -8 3 8 3 -8 4 8 4 -8 5 8 5 -8 6 8 6 -8 7 8 7 -8 8 8 8 -8 9 8 9 -9 0 9 0 -9 1 9 1 -9 2 9 2 -9 3 9 3 -9 4 9 4 -9 5 9 5 -9 6 Year — Proportion - « - Amount Source: Annual Reports of TNCHF. Amount (Million Rupees, unadjusted) In spite of such strength, the institutional structure for finance faces some in herent structural weaknesses. This primarily stems from the Apex cooperative’s de pendence on other sources for finance. First, flow of funds from other sources is er ratic. There is little certainty when funds will come through, which makes on-lending to primary cooperatives a problem. Primary cooperatives may have to wait six months or more to receive funding in certain cases. Second, since TNCHF on-lends the money it borrows, and adds percentage points over the interest rate from the original lender, the effective interest rate of loans from TNCHF could be potentially more than the market interest rates. In an environment of limited opportunities for formal hous ing finance, higher interest rates may not have been problematic. However, with eco nomic liberalization during the 1990s, many private Housing Finance Institutions emerged and started to provide competitive interest rates. Hence, competitiveness of cooperatives as financial mechanisms could lessen in urban areas where such HFIs have emerged. However, cooperatives may remain a major source of finance in rural areas, where alternative financial sources are limited. 5.5. Conclusion The distinctive economic and demographic characteristics of Chennai influ enced its housing morphology. In the pre-Independence period, Chennai’s economic growth was modest. Consequently, unlike Mumbai, Chennai did not undergo modern istic urban development. Migration to the city was predominantly composed of fami lies from neighboring villages, who maintained rural-urban connections. The rural social order was thus reproduced in Chennai. In this context, Chennai developed as a “city of villages,” wherein the city’s internal rural structure was preserved. Chennai’s housing morphology followed the village housing pattern. It developed as a network of settlements with relatively low-rise housing structures on individual plots. Many of 203 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the larger houses were originally constructed as single family dwellings or for the ex tended family, but with population growth, multiple families occupied a single house. Housing development in Chennai was horizontal (a phenomenon that contin ued till the 1970s), rather than vertical, wherein individual plotted developments be came the norm. Hence, there was little demand for collective organizations like hous ing cooperatives for collective ownership and management. Tenure Cooperatives hardly emerged in Chennai. Rather, housing cooperatives were adapted for financing and constructing housing in individual plotted developments. In the post-Independence context, housing cooperatives hardly diversified their activities since Independence. Unlike Mumbai, the institutional context in Chennai did not favor the development of cooperatives. Government controls through the Tamil Nadu Special Officers Act, 1976 reduced the autonomy of cooperatives. Under the Act, government nominees controlled the decision making processes of coopera tives. Consequently, cooperatives developed in the direction emphasized by the gov ernment. As the state emphasized use of cooperatives as vehicles for finance in rural areas, their role in urban areas decreased. The cooperative institutional structures that emerged—or did not emerge— were also in response to such institutional conditions. Institutional structure for finan cial support was strengthened, with the Apex cooperative playing a pivotal role in the disbursement of housing finance. The institutional structure for financial support rein forced the role of cooperatives as financial mechanisms. Other institutional structures for providing administrative, legal, and procedural support to cooperatives hardly emerged. Institutional support for land also hardly existed. Consequently, coopera tives grew modestly in Chennai and hardly diversified. Multi-family housing, which emerged in 1970s, and became more popular during the 1990s have been owned in the 204 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. form of condominiums, rather than cooperatives. Thus, the private sector hardly adapted cooperatives. Public agencies and NGOs have also hardly used cooperatives for collective action. 2 0 5 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 6. Differential Evolution of Cooperatives in Mumbai and Chennai During pre-Independence period, housing cooperatives were more active in Bombay and Madras provinces as compared to other provinces. In this period, hous ing cooperatives had emerged as Tenure Cooperatives in Mumbai, and as Finance Co operatives in Chennai. In the post-Independence period, while housing cooperatives’ activities expanded phenomenally in Mumbai, their activities grew modestly in Chen nai. In Mumbai, housing cooperatives play diverse roles; in Chennai, cooperatives are mainly financial mechanisms, and their activities in land development reduced since Independence. In this chapter, I make a comparative analysis of the differential evolu tion of housing cooperatives to draw broader policy lessons on the institutional fea tures that expand or constrain the development of housing cooperatives. In Chapters 4 and 5 ,1 examined the institutional context of development of co operatives in Mumbai and Chennai respectively. In this, I highlighted how the histori cal conditions of the local housing market and the institutional framework influenced the development of housing cooperatives. The cooperative institutional structures that emerged were also in response to the specific local institutional conditions, but helped reinforce the activities of cooperatives. I propose that there are three policy lessons to be drawn from comparison of the two cities. First, housing policies need to reconcile with the local housing market conditions, since housing cooperatives are quite sensi tive to such conditions. This implies determining the housing market’s readiness and the nature of demand for such organizations. Second, at a broad political level, while state involvement may be helpful, cooperatives require a certain degree of autonomy to expand their role in the housing market and to diversify their activities. State con trols would direct cooperatives in the directions where the state lays an emphasis. Al though this may be useful, it may also inhibit the otherwise broader potential that co- 206 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. operatives have for the housing process. Third, the state could play a catalytic role in establishing the cooperative institutional structures, which in turn could support the formation and functioning of housing cooperatives. This chapter is structured in four sections. First, I make an endogenous level comparison of housing cooperatives between the two cities. Second, I make an ex ogenous level comparison of the local housing market and the institutional context of development of housing cooperatives. Third, I highlight the role of cooperative insti tutional structures in enabling the expansion of housing cooperatives. Fourth, I con clude with the policy implications. 6.1. Endogenous Level Analysis Housing cooperatives formally emerged in Bombay and Madras presidencies in the second decade of the twentieth century, with the amendment of Cooperative So cieties Act in 1912. Although the Act was originally intended to serve the credit needs of rural peasantry, cooperatives could be used for non-credit purposes under the 1912 Act. In Mumbai, cooperatives emerged primarily as Tenure Cooperatives, in the form of Tenant Co-partnership or Tenant Ownership Societies (Chapter 4). The coopera tives constructed as well as collectively owned the housing. In Chennai, cooperatives emerged primarily as Finance Cooperatives, in the form of Ordinary Building Socie ties. The cooperatives financed members for construction of their housing. Members constructed and owned housing on an individual basis. Members had little relation ship with the cooperative once they re-paid their loans. In this, Finance cooperatives were similar to banks specializing in housing. Any individual could potentially be come a member of the Finance Cooperative and borrow using property as a collateral. The organizational differences between housing cooperatives in Mumbai and Chennai persisted during the post-Independence period. In Mumbai, although housing 207 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cooperatives remained primarily as Tenure Cooperatives, they expanded their activi ties to play diverse roles in the housing market. The private sector adapted housing cooperatives for collective ownership and management; the public sector implemented public housing schemes through the cooperatives; and NGOs began to use coopera tives as vehicles for mobilizing collective action. In Chennai, housing cooperatives remained primarily as Finance Cooperatives, and hardly expanded their roles. Various types of Building Cooperatives that were introduced after Independence for housing construction or land development did not sustain their growth. Indeed the activities of Building Cooperatives even reduced since Independence. The internal organizational differences between housing cooperatives in Mum bai and Chennai are striking, and it is useful to consider these differences for the inter nal organizational factors that contributed to the differential evolution during the post- Independence period in the two cities. Tenure Cooperatives in Mumbai were involved in both the production and consumption processes of housing. In Chennai, Finance Cooperatives were mainly involved in the production process; they were not substan tively involved in the consumption process. Building Cooperatives, by the very nature of their housing objectives, were also involved mainly in the production process. Consideration of cooperatives’ role in the production and consumption process raises some fundamental theoretical issues with respect to how intrinsic characteristics of the cooperatives affected the evolution of cooperatives in the two cities. First, the nature of relationships between members and between members and cooperatives are distinctive between Tenure and Finance Cooperatives.1 Tenure coop eratives involve perpetual relationships between members and between members and 1 In this chapter, I am concerned with formal Tenure, Finance, and Building Cooperatives as they function in Mumbai and Chennai. Hence, this excludes cooperatives as microfinance organizations, self-help groups, and other collectives. I consider the other types o f organizations in the next chapter. 208 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cooperatives due to the perpetual nature of ownership in the collective. Members have to deal with each other in making collective decisions. Relationship between members and Finance Cooperatives are strong as long as the member expects to borrow, or is required to re-pay the loans. These cooperatives have tenuous relationships between members in the collective decision-making. Second, Tenure Cooperatives tend to be stable due, again, to perpetuity of ownership. Finance Cooperatives tend to be stable as long as their capital base is sta ble. However, tenuous relationships between Finance Cooperatives and their mem bers could result in Finance Cooperatives having a higher turnover rate. That is, there is a flux of older members and new members: older members exit when they complete their financial obligations, and new members enter in order to borrow. Unlike Finance Cooperatives, Tenure Cooperatives tend to have stable membership. Third, Tenure Cooperatives are highly site specific, i.e. their operations are limited to the real property that is commonly owned. Operations of Finance Coopera tives are not site-specific, i.e. money is fungible and could be utilized for financing housing activities (construction/ development/ repairs) for any site that falls within the geographical purview of the cooperative. Building Cooperatives are characteristically different from Tenure and Finance Cooperatives. They are mainly involved in the production process. Once they fulfill their objectives of housing construction or land development, Building Cooperatives have little role to play (unless such property is owned in the collective through Tenure Cooperatives). Thus, relationships between members and cooperatives exist for the period of production process. After this, Building Cooperatives cease to exist. As Bonus (1986) argues, cooperatives are characterized by a delicate balance between centrifugal forces of being in the collective, and the centripetal forces of in- 209 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. dependent operations. In this, asset specificity matters for the degree to which the scale tips towards the collective or independent operations (Williamson, 1985). In Tenure Cooperatives, which have a high degree of site-specificity, the balance tips to wards centrifugal forces. In Finance and Building Cooperatives, centrifugal forces act as long as members perceive benefits accruing to them from the cooperative. Mem bers exit the cooperative once such benefits stop. Consequently, unlike Tenure Coop eratives, Finance and Building Cooperatives in Chennai hardly have had binding rela tionships between members. Tenure Cooperatives in Mumbai average about 20 to 30 households (sometimes even less).2 Membership in Finance cooperatives in Chennai could range anywhere between 200 and 3000 (the Madras City Cooperative Housing -3 Society had about 3500 members in 1995-96). Building Cooperatives, especially those involved in land development, also tend to have large membership, between 200 and 800 members. Stability raises both prospects and problems for administration in Tenure Co operatives as the high degree of binding relationships between members entails both collective action strengths and problems. While the cooperative provides an organiza tional mechanism for internalizing problems between members due to indivisibility of common properties, it also faces problems of free-riding (such problems typically in clude non-payment of dues by members, non-participation of members in collective decision-making, and so on). The larger prospect of stability of Tenure Cooperatives from a game-theoretic perspective is that relationships between members are not one- shot games; rather they are iterative games, where cooperation between members is an 2 Tenure Cooperatives in Delhi range between 50 and 150 members. The minimum number o f members according to Cooperative Societies Act in most states is 11. 3 Madras City Cooperative Housing Society, Annual Report (1995-96). 210 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. optimal outcome.4 Thus, disputes between members could be potentially addressed through collective decision-making within the cooperative.5 Yet, problems arise when members’ preferences and collective decisions do not align with each other, i.e. there are disputes between members and cooperatives. Resolving such problems requires a secondary mechanism external to the cooperatives, i.e. an institutional structure for administrative support. In Finance Cooperatives, exit by members is not directly a problem for the or ganization. Loans are secured through property mortgage to ensure credible commit ment to repay. Although Finance Cooperatives act like a bank specializing in housing finance, the ideal of cooperative is maintained—in that, loans obtained by the coopera tive is a collective liability.6 In Building Cooperatives too, exit by members does not affect the organization directly, as the credible commitment to repay is ensured through various means (e.g. hire-purchase of houses, providing ownership title deeds after full payments, and so on). The main weakness of Finance and Building Cooperatives (including Tenure Cooperatives where the cooperative is also involved in construction) is access to fi nance and land, which are the basic elements of housing. Cooperatives entail pooling of members’ resources, which is often in the form of share capital. However, since investments in housing are lumpy, finance raised through members needs to be lever aged with external funding for adequate investment in housing (Chapter 2). Coopera 4 See Axelrod (1984). 5 Typical problems in Tenure Cooperatives in Mumbai, for example, include neighborly disputes relating to pets, noise, maintenance o f common areas, and so on, which are sorted out through rules established within the coopera tive. 6 However, more recent organizational innovations cut through the collective liability. Private sector Housing Fi nance Institutions (HFIs) and some Apex Cooperatives (notably, Delhi Cooperative Housing Finance Corporation) provide financing to individuals; members need to produce an allotment letter from the cooperative. Individual members are liable to repay in such cases. As o f this study, the Tamil Nadu Cooperative Housing Federation lends only to primary cooperatives. 211 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. tives face obstacles in raising such finance in the open market. First, under coopera tive rules, Finance Cooperatives can lend mainly to members in India; second, bank procedures for lending are usually geared towards individual short-term lending. In such an environment, the three types of housing cooperatives could develop to the ex tent the funds raised through members are adequate, or they have access to external finance. In the pre-Independence period, such external source was typically either the Government or cooperative banks. In terms of land, Tenure and Building Coopera tives require large parcels of land for contiguous housing development. As collec tives, cooperatives had little access to land in the open market, partly again due to organizational constraints in access to finance. The types of housing cooperatives that emerged in Mumbai and Chennai thus required specific types of institutional support to overcome their internal organiza tional weaknesses for them to expand. On one hand, housing cooperatives provided an impetus for such support; on the other hand, such support came in the specific con text of the cities’ local housing market and local institutional framework. In Mumbai, stability of Tenure Cooperatives gave rise to secondary structures for overcoming some of the internal organizational problems. Mumbai District Cooperative Housing Federation, for example, emerged in this context to provide administrative, legal and procedural support to the Tenure Cooperatives. In Chennai, support mechanisms for mobilizing finance to housing cooperatives was crystallized in the context of the state’s support to cooperatives for finance. In the absence of strong state support for access to land, the activities of Building Cooperatives, which are inherently transient by their nature, reduced. 2 1 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 6.2. Exogenous Level Analysis At the exogenous level, both the local housing market and the local institu tional framework influenced the differential development of housing cooperatives in Mumbai and Chennai. The local housing market characteristics—particularly, the housing morphology—of the two cities differed due to the historical context in which their urban growth took place. Consequently, the demand for collective organizations like housing cooperatives differed between the two cities. In terms of the local institu tional framework, the housing and cooperative policies differed between the two states. While these policies generally favored the development of cooperatives in Mumbai, such policies were more restrictive in Chennai. 6.2.1. Local Housing Market In Chapters 4 and 5, I outlined in detail the historical circumstances of urban growth and demography in Mumbai and Chennai that influenced the evolution of the local housing market. In particular, the housing morphology of the two cities evolved differently, so that the demand for collective organizations like housing cooperatives also varied. Indeed, the local housing market fundamentally influenced the emergence of Tenure Cooperatives in Mumbai and Finance Cooperatives in Chennai. With Mumbai’s economic boom in the later nineteenth century, a “modem” urban form emerged, in which the underlying rural character of the city was erased. Labor immigration into Mumbai due to the economic boom was also largely male dominated. The city’s modernization and rapid increase in population resulted in dense housing conditions, in which multi-family dwelling units in multi-storied build ings and chawls became the norm. In this context, housing cooperatives were adapted as “modem” organizations for providing decent housing for the middle class. 213 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Although Chennai was a major gateway to South India, it did not experience an economic boom like Chennai. Its economic growth was sluggish, and grew essen tially as a “city of villages.” Unlike the largely male dominated labor immigration into Mumbai, immigration into Chennai was largely comprised of families migrating from nearby rural areas. Further, while Mumbai’s urban growth was constrained by its linear geography, Chennai’s urban growth was not constrained by geography. In this context, the underlying rural base of the city remained, including the preferences of households for individual housing. While Mumbai grew vertically to accommodate the increasing population, Chennai became more horizontally dense. Chennai’s hous ing morphology was thus mainly low-rise housing in plotted housing. In this context there was little demand for collective organizations like cooperatives. Consequently Tenure Cooperatives hardly emerged. Rather, housing cooperatives were adapted as means to finance housing construction. Similar to the pre-Independence period, Mumbai kept a leading edge in eco nomic growth as an industrial and financial center, including becoming India’s most globalized city. In the context of the city’s population growth and other institutional constraints highlighted in the next section, the residential density increased so that in dividual plotted development became a luxury even for the rich. Indeed, the housing conditions became worse, to the extent that over 50 percent of the population lives in slums. In this context, formal private as well as public housing were mainly multi family dwelling units, wherein housing cooperatives were adapted as mechanisms for collective ownership and management. Unlike Mumbai, Chennai’s economic growth remained sluggish until recently in the 1990s, when it also experienced growth due to globalization (particularly in in formation technology and automobile industries). Chennai remained a low rise city 214 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. until the mid 1970s, after which public housing agencies (Tamil Nau Housing Board and Tamil Nadu Slum Clearance Board) began to construct multi-family dwelling units for conserving land. Private agencies, however, did not follow suit immediately, and the preference for individual plotted housing remained even during the mid 1980s. Thus, the need for collective organizations for ownership and management hardly arose until the 1990s. Yet, housing cooperatives were not adapted as the means for forming such collectives. In spite of introduction of Tenure Cooperatives for mainte nance in TNHB flats, such cooperatives failed and had to be liquidated. In the context of individual plotted development in Chennai, housing cooperatives were adapted as mechanisms for housing finance. Different types of Building Cooperatives that emerged after Independence were also involved in developing individual plots and housing for hire purchase. Clearly, the type of housing cooperatives that emerged in the two cities were embedded in the history of local housing market, particularly the factors that affected the demand for collective organizations like housing cooperatives. From a policy per spective, this implies that housing policies need to reconcile with the local housing market conditions, in determining the housing market’s readiness and the nature of demand for collective organizations. 6.2.2. Local Institutional Framework In terms of the local institutional framework, rent control policies and apart ment ownership regulations favored the development of Tenure Cooperatives in Mumbai during the post-Independence period. Coupled with the housing policies, policies in Mumbai provided cooperatives greater autonomy in their operations. Unlike Mumbai, rent control did not form the core of housing policies in Chennai, and 215 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. apartment ownership regulations were hardly in vogue in the context of plotted devel opments. Apartment ownership regulations were instituted much later in Chennai. Rent control policies have been fundamental to Mumbai’s institutional frame work. They have been in effect throughout the post-Independence period. They fa vored growth of Tenure Cooperatives indirectly by creating incentives for housing ownership. In the face of rent control, incentives for rental housing construction re duced. Landlords had little incentive to maintain buildings since the returns were low. Cooperatives offered an alternative to landlord-tenant relationships by internalizing n the relationship through joint ownership. Since returns from rental housing were low, the formal private sector became oriented towards construction of housing for sale to owners. The Maharashtra Owner ship Flats Act, which was passed in 1963, required private developers to transfer own ership of housing to prospective buyers through a collective organization like a coop erative or a company. Another piece of legislation—the Maharashtra Apartment Ownership Act, 1970—allowed households to own housing on an individual basis (i.e. as condominiums), and required owners to form apartment associations. Although the two Acts did not specify cooperatives only, cooperatives became more popular than other types of organizations. I argue that there were two reasons for popularity of co operatives over other types of organizations. First, Tenure Cooperatives had a long history in Mumbai, so that ownership through cooperatives was broadly understood. Second, institutional structures supporting cooperatives were already well established, so that the transaction costs of formation and functioning of cooperatives were low. 7 A speculative question is, would Tenure Cooperatives have persisted if rent control policies were not there? If we remove such a constraint, still the demand for collective organizations like Tenure Cooperatives would exist in Mumbai due to its housing morphology and flat ownership regulations. Hansmann (1996) argues that rent control, which has been historically significant in N ew York, was not a major factor for the rise o f cooperatives in the city. Indeed, it is hard to argue that rent control would give rise to cooperatives in all contexts. In Mumbai, cooperatives became an alternative due to confluence o f specific institutional conditions. 216 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Further, cooperative policies in Mumbai were more liberal in providing greater autonomy to cooperatives. The state hardly intervened directly in the internal deci sion-making process of cooperatives. The institutional structures that emerged also ensured that the autonomy of cooperatives was not compromised. In this, the regula tory powers of the Cooperative Registrars were limited to overseeing the conformity of cooperatives with the Maharashtra Cooperative Societies Act. Unlike Mumbai, although rent control measures existed in Chennai, it did not form the core of housing policies. Hence, the rental housing market was not affected significantly.8 Further, in the context of plotted development, apartment ownership regulations had hardly been in vogue. Such regulations came up in 1995, when con struction of multi-family dwelling units had become more prominent. Thus, housing policies did not directly affect housing cooperatives. Further, the state did promote housing cooperatives in the initial years following Independence. In this, housing co operatives were promoted not only as Finance Cooperatives, but also as Building Co operatives (for land development and housing construction). Schemes for public housing used cooperatives as an alternative implementation vehicle. Yet, other forms of housing cooperatives hardly survived in the ensuing decades. Cooperative policies since the mid 1970s became more restrictive, as the state government controlled the formation and functioning of cooperatives. The Tamil Nadu Special Officer’s Act, 1976 effectively reduced the autonomy of cooperatives. Cooperative board elections were hardly held for nearly 20 years until 1996, when the Act was revoked. However, the Act was re-invoked again in 2001, so that government control over cooperatives continued. U nder the A ct, governm ent nom inees (i.e. Spe 8 Similar to Mumbai, a speculative question is, if rent control policies were significant, would Tenant Cooperatives have arisen in Chennai? In the context o f plotted development, it is doubtful if Tenant Cooperatives would have emerged; however, it is hard to speculate the alternative forms o f tenure that would have arisen. 217 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cial Officers) managed the daily affairs of cooperatives and had the power to override the decisions of cooperative management boards. In this, cooperatives became an arm of the state, and necessarily developed in the directions where the state laid an empha sis. With the state’s emphasis on use of cooperatives for rural housing, cooperatives came to be used as vehicles for housing finance in rural areas. The state government set up Taluk Cooperative Housing Society in each district for augmenting rural hous ing. Cooperatives’ role in urban housing became marginalized. Policies for broadening the scope of cooperatives have indeed been beneficial for rural housing. However, cooperatives lost their autonomy in their process. Loss of autonomy constrained the range of activities that cooperatives could otherwise have performed. Cooperatives became dependent on the state for their activities. In Chen nai, the role of Building Cooperatives reduced as state provided little support for allo cating land. In this context, cooperatives’ activities became confined primarily to pro viding finance for housing construction and repairs. 6.3. Role of Cooperative Institutional Structure The development of cooperative institutional structures differs in Mumbai and Chennai. In the case of Mumbai, the cooperative institutional structures are overall robust and well developed. The structures provide support in the three crucial areas where housing cooperatives are weak. They provide administrative support; they pro vide financial support; and there is a broad institutional support for access to land. In the case of Chennai, the institutional structure for financial support is well developed; other support systems hardly exist. I argue that the broader development of coopera tive institutional structures in Mumbai assisted in expanding the diverse roles of coop eratives; in the absence of such institutional structures the range of activities of hous ing cooperatives in Chennai narrowed. 218 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In the context of relatively liberal cooperative policies in Maharashtra, the in stitutional structures that came up ensured greater autonomy of cooperatives. The Reg istrar’s office, for example, plays an advisory role rather than a regulatory one. Under Maharashtra Cooperative Societies Act, the final decision of each cooperative lies with the “general body of members,” so that Registrars cannot interfere in the decision making process. Disputes between members are also settled in separate Cooperative Courts, which are separate from the Registrar’s office. Thus, structurally, coopera tives have relatively higher degree of autonomy from the Registrar’s Office. Relative independence from Registrar’s office also implies less state government interference. In its advisory role, the Registrar’s office provides administrative and legal support to cooperatives and their members. Mumbai District Cooperative Housing Federation, which is a secondary cooperative that evolved with the cooperative structure, provides other procedural support. Provision of such administrative, legal, and procedural sup port reduced the transaction costs of formation and functioning of cooperatives. Administrative, legal, and procedural support systems have hardly evolved in Chennai. Rather, in the context of Tamil Nadu Special Officer’s Act, the Registrar’s Office has more of a regulatory role as a controller. In addition, the Registrar’s Office also has the legal role of settling disputes between cooperatives and members. With the state controlling most of the cooperative’s functioning, the need for separate ad ministrative and legal support systems hardly arose. In the process, cooperatives have become more dependent on the state, wherein the Registrar’s office could effectively intervene in the daily functioning of housing cooperatives. The two-tiered institutional structure for financial support exists in both cities, since this structure is more broadly institutionalized in almost all states in India. However, the Apex cooperative in Tamil Nadu (the Tamil Nadu Cooperative Housing 219 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Federation, TNCHF) is more active than Apex cooperatives in other states (including Maharashtra) in terms of volume of lending, loan amounts and housing units financed. This is due primarily to the state government policies on rural housing, wherein TNCHF is the principal vehicle for financing Taluk Cooperative Housing Societies. Apex Cooperatives’ proportion of lending reduced in both Mumbai and Chennai dur ing the post Independence period due to the rural emphasis. However, due to emer gence of private Housing Finance Institutions, Tenure Cooperatives in the private sec tor in Mumbai depend little on funding from the Apex cooperative. In Chennai, Fi nance Cooperatives primarily depend on TNCHF for their funding. The nature of the institutional structure reinforced the development of housing cooperatives as financing mechanisms. In terms of access to land, although there is no formal structure through which land is mobilized for housing cooperatives, institutional support exists in Mumbai in the form of preferential treatment. Public housing agencies and metropolitan devel opment agencies allocate plots through housing cooperatives; housing schemes are also carried out through cooperatives. Large-scale programs like World Bank assisted sites and services schemes in Mumbai, for example, were carried out through provi sion of cooperative tenure. Although support for access to land existed in Chennai in the initial years following Independence, such support waned since the 1970s when the state government began to control the development of cooperatives. In this con text, the activities of Building Cooperatives reduced since Independence. Coopera tives did not even enter into consideration for large scale programs like World Bank assisted sites and services programs, w here tenure w as granted on an individual ow n ership basis. 2 2 0 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Both cooperatives’ internal organizational characteristics and the external insti tutional framework have influenced the cooperative institutional structures in the two cities. The structure and function of Registrar’s Office in Mumbai, for example, re sults from the broader policies on autonomy of cooperatives; on the other hand, the structure also evolved in response to the organizational weaknesses of cooperatives. In the context of state policies on utilizing cooperatives as financing mechanisms, TNCHF played a key role in financing primary cooperatives in Chennai. However, unlike Mumbai, there is little institutional support for providing access to land in Chennai. The state has thus a key role to play in setting up the institutional structure; however, the specific form of institutional structures depends on both the institutional framework and the organizational characteristics. 6.4. Policy Conclusions Table 6.1 summarizes the major characteristics with respect to internal features of cooperatives, the local housing market, the local institutional framework, and the cooperative institutional structure in Mumbai and Chennai. Clearly, the types of hous ing cooperatives that emerged in the two cities were distinctive, depending on the lo cal housing market conditions. Housing cooperatives emerged as Tenure Coopera tives in Mumbai in response to the housing morphology of multi-family dwelling units; in Chennai, where housing was owned on an individual basis, housing coopera tives were adapted as Finance Cooperatives. From a policy perspective, this high lights the need for housing policies to reconcile with the local housing market condi tions. 2 2 1 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 6.1. Summary of Comparison between Mumbai and Chennai Mumbai Chennai Endogenous level • Original Co-op Type Tenure Finance • Current Role Diverse Primarily Finance Exogenous level • Local Housing Market Multi-family housing Plotted development • Local Institutional Framework High Low a. Housing Policy i. Rent control ii. Flat Ownership Both policies not sig nificant b. Cooperative Policies Liberal Controlling Institutional Structure • Administrative Support Strong Weak • Support for access to finance Moderate Strong • Support for access to land Strong Weak Post-Independence Growth Phenomenal Modest Housing cooperatives in Mumbai expanded their activities in the post- Independence period. While rent-control measures created incentives for ownership, flat ownership regulations raised the need for forming collectives for ownership and management. Coupled with housing policies, cooperative policies in Mumbai were also relatively liberal in providing more autonomy to cooperatives. In such an institu tional environment, cooperatives’ activities expanded and were adapted for diverse roles in Mumbai. In Chennai, rent control was not significant, and the need for apartment owner ship regulations hardly arose in the context of its plotted development. Even in the context of apartment ownership regulations since 1995, apartments are mostly owned in the form of condominiums, rather than cooperatives. Cooperative policies were also restrictive in controlling the activities of cooperatives. Consequently, coopera tives developed in the directions where the state laid an emphasis, viz primarily as Fi nance Cooperatives. 2 2 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. From a policy perspective, the above raises the role of the state. That rent con trol influenced growth of housing cooperatives in Mumbai does not mean that it is necessarily a pre-condition. Indeed, Chapter 4 highlighted how rent control policies distorted the housing market in Mumbai, and there is much debate in housing litera ture whether rent control is beneficial. The larger point is the role of the state with respect to cooperatives. In Mumbai, cooperatives enjoy relatively greater autonomy than Chennai. While cooperatives diversified their role in Mumbai, housing coopera tives remained mainly as Finance Cooperatives in Chennai. Activities of Building Cooperatives reduced since Independence in Chennai. Yet, state involvement has been beneficial in widening the geographical focus of housing cooperatives over to the whole state. This raises the dialectic of state control vs cooperative autonomy. I argue that while state control may indeed be beneficial in some respects, cooperatives do require a degree of autonomy. The state could ensure that some basic legal principles are followed by cooperatives in their functioning; however, it should also ensure the independence of cooperative management boards in decision-making. Interference in the daily activities of cooperatives would restrain the range of activities they could otherwise perform. The state could also play a catalytic role in setting up the institutional struc tures supporting cooperatives. In Mumbai, such structures are robust and rather well developed. Institutional structures for administrative, legal and procedural support are strong. Similar structures hardly exist in Chennai. Institutional structure for financial support is stronger in Chennai, mainly due to the Apex Cooperative (TNCHF). Even though such institutional structure in Mumbai is comparatively moderate, the existence of the support is crucial. In the face of growth of private Housing Finance Institutions (HFIs), the role of the Apex Cooperative in Mumbai may be more sharply defined for 223 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. benefiting low-income households. Lastly, the institutional support for access to land is better developed in Mumbai than in Chennai. I argue that the institutional structures played a key role in the phenomenal growth and expansion of housing cooperative’s activities in Mumbai. These structures helped overcome the internal weaknesses of cooperatives. Once set-up, these struc tures assisted in reducing the transaction costs of formation and functioning of housing cooperatives. Hence, these structures acted as reinforcing mechanisms in the growth and expansion of the activities of housing cooperatives. The institutional structures in Mumbai and Chennai were clearly shaped by both the internal organizational features of cooperatives in the two cities and the external institutional framework. In Mumbai, for example, the flat ownership regulations did not explicitly stipulate the formation of cooperatives only; rather, it stipulated the need for collective organizations (companies or cooperatives). Apartment ownership regulations instituted in 1970 were oriented towards individual ownership. Yet, housing cooperatives proliferated in comparison to other types of organizations. As housing cooperatives had a longer history in Mumbai, procedures for forming cooperatives and their functioning were relatively well known. More importantly, the support structures were instrumental in institu tionalizing such procedures. This reduced the transaction costs of formation and func tioning of housing cooperatives. Further support in terms of access to finance and ac cess to land helped expand the activities of housing cooperatives. As highlighted in Chapter 4, cooperatives play diverse roles. The private sector utilizes cooperatives for collective ownership and management. Public housing agencies use cooperatives as vehicles for providing housing, and for transfer of ownership in rent controlled build ings to tenants. Public agencies and NGOs use cooperatives as vehicles for slum up grading and slum redevelopment. Lastly, NGOs utilize informal cooperatives as vehi 2 2 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cles for collective action (e.g. campaigning for housing rights, microfmance, self-help groups). In contrast to Mumbai, the institutional structure for financial support is strong in Chennai; in this context, activities of housing cooperatives as finance mechanisms persisted. 2 2 5 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 7. Potential of Housing Cooperatives for Low-Income Households In the pre-Independence period, housing cooperatives were hardly considered for low-income households. However, in the post-Independence period, national housing policies in India consistently promoted housing cooperatives as means for housing low-income households. In this chapter, I explore the potential of housing cooperatives to serve the housing needs of low-income households. At the endoge nous level, housing cooperatives are not necessarily oriented towards any particular income group. Yet, the collective feature of cooperatives’ governance could aid low- income households in terms of pooling of resources, achieving community action (col lective self-help), increasing creditworthiness, and limiting speculation. However, these features may not be achieved due to collective action problems endemic to the collective. At the institutional level, the external institutional framework may or may not support the formation and functioning of housing cooperative—the framework may lower or raise the transaction costs of cooperatives’ operations. The nature of the local housing market becomes significant at this level in influencing the demand for cooperatives. Lowering transaction costs would enable cooperatives to respond to the market demand and allow them to reach further down the income categories. At the national level, housing policies in India consistently promoted housing cooperatives as means for housing low-income households. The elaborate national two-tiered institutional structure for financial support assisted cooperatives in broadly reaching the low-income households. The two states of Maharashtra and Tamil Nadu also maintain a leading edge in serving the needs of low-income households. How ever, formal housing cooperatives hardly serve low-income households in their capital cities, Mumbai and Chennai respectively. While Mumbai’s local housing market has effectively priced formal housing market outside the reach of low-income households, 226 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. state controls constrained cooperatives’ activities in Chennai. Yet, Mumbai provides better prospects for housing low-income households through cooperatives due to low transaction costs of formation and functioning. In this, the institutional structure has played a critical role in reducing the transaction costs. Due to difficulties in defining “low-income” based on a particular income slab or percentile, I have considered the meaning of low-income on a qualitative ordinal scale. The Government of India defines four income groups, based on income slabs: economically weaker section (EWS), low-income group (LIG), middle-income group (MIG), and high-income (HIG). The income slabs are periodically adjusted for infla tion and consumer prices. My meaning of low-income group broadly corresponds to the LIG and EWS, i.e. the poorer sections of the society.1 The chapter is organized as follows. First, I make an endogenous level analy sis of housing cooperatives for the advantages and disadvantages that they hold for low-income households. Second, I make an exogenous level analysis of housing co operatives in the Indian context. Third, I compare the cases of Mumbai and Chennai. Fourth, I conclude with the role of cooperative institutional structure in aiding coop eratives to reach down the income groups. 7.1 Endogenous Level Analysis In Chapter 2, I highlighted that cooperatives entail open membership, so that they are not particularly oriented towards any particular income group. Cooperatives could be potentially formed across the range of income groups, from high to low. In such a context, what are the specific advantages and disadvantages for low-income 1 In 1996-97, EWS corresponded to households with monthly incom e upto 2,100 rupees; LIG between 2,101 and 4,500 rupees; MIG between 4,501 and 7,500 rupees; and HIG above 7,501 rupees. Som e definitions o f low-incom e also include the lower m iddle-income groups. 227 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. households to join or form cooperatives? To understand this, we must examine the incentives that cooperatives hold for low-income households. Of course, some of the incentives and disincentives that the cooperative holds for members in general (Chap ter 2) are also be applicable to low-income households. As a collective organization, the cooperative offers advantages that arise from being a collective. First, cooperatives entail collective pooling of resources, which offers economy of scale. The collective pooling lowers the individual housing costs that each household would otherwise incur. Land and infrastructure costs, for exam ple, could be shared in the form of cluster housing, group housing, or multi-family housing. Yet, cost sharing may or may not benefit low-income households depending on local real estate market. For example, under high housing demand conditions, land or housing costs may be so high that cost sharing may yet not make housing affordable for low-income households. Second, credit worthiness of low-income households rises due to collective pooling of resources. Low-income households typically have little access to formal housing finance due to lack of collateral and meager income stream. They have to de pend on informal credit or personal savings. Pooling their limited resources in the group increases their collective asset value. Group lending and micro-finance pro grams usually entail formation of a cohesive network of households for lending. At least two reasons are offered for why cooperatives may function better than conven tional banks for lending to low-income households. The first reason is that durable cooperatives avoid asymmetric information between the lender and the borrower due to local knowledge about each other’s assets, capabilities and character traits; thus they lower transaction costs of screening (Wenner, 1995). The second reason is that members have incentive to use group based social sanctions to enforce repayment 228 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. since all members will be deprived of future borrowing in case of default; such social sanctions are typically not available in a conventional bank (Besley and Coate, 1995). Peer monitoring among borrowers can also be more effective since they have better information about each other’s borrowing capacity (Varian 1990; Stiglitz 1990). Third, cooperatives foster community action (Clapham and Kintrea, 1995), community empowerment (Saegert and Winkel, 1996), and act as organizational vehi cles to foster collective action and self-help (UNCHS, 1989). As indicated in Chapter 4, NGOs in Mumbai have organized slum and pavement dwellers into informal coop eratives to mobilize their potential for collective action. Organizing cooperatives fo cused on particular gender based housing needs also encourages women’s participa tion and motivates them to undertake community leadership and action (Moser, 1992). Marcuse (1992) argues that self-help strategies would be more effective if they were based on collective self-help, collective democratic decision making and non exclusive “expansionist” structure. Fourth, speculation and gentrification can be potentially contained in coopera tively owned housing. Collective ownership of land, for example, limits ability of in dividual households to sell land in the open market. This can be especially beneficial in slums where granting of land tenure to households on individual basis could result in “upward filtering” (Johnson, 1987).2 Turnbull (1982) proposes “Cooperative Land Bank” as a means of limiting speculation through collective ownership. In this ar rangement, the cooperative owns the land, superstructures, and the infrastructure. In dividual households have perpetual leases to their unit, and each household owns 2 “Upward filtering” refers to the process by which the housing stock passes up the income group. This could oc cur when slum-dwellers are given ownership rights on individual plots. After legalization, the land could be “raided” by upper income households since it could be legally transacted. The pressure for upward filtering exists especially in slums in centrally located areas, where land values are high. 229 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. shares proportional to the area occupied by it. When a household sells the housing unit, the cooperative appropriates a part of the value of land, infrastructure and com mon areas; the household obtains rest of the value, including the value of housing unit. Thus, the whole profit does not accrue to the household when it sells the house. Community Land Trusts and Limited Equity Cooperatives in the United States simi larly capture a part of the equity when a household sells the housing unit. Speculation is potentially limited through such sharing of value. While the cooperative entails the above advantages for low-income house holds, it is also beset by concurrent collective action and management problems identi fied in Chapter 2. First, housing cooperatives, especially for low-income households, face entry problems and are seldom voluntarily formed. Usually an external agency (public housing agency or a NGO) is required to promote such cooperative. Second, self-management can be especially problematic in cooperatives of low-income households due to the lack of managerial, technical, legal, or financial ex pertise by members (Vakil, 1996). Third, low-income households face opportunity costs between participating in the collective and other productive income earning ac tivities. Households will participate if benefits from the collective outweigh costs of participation. Furthermore, cooperatives of low-income households also face the problems of access to finance and land. Although credit-worthiness of low-income households may rise by pooling their resources in the group, these cooperatives do not have broad access to finance. Lending to such cooperatives is not yet widely institutionalized, and usually requires a N G O as an interm ediary betw een the donor and the cooperative. Moreover, land and building costs in large cities may yet not be affordable for low- income households, despite the economy of scale obtained through the group. 230 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The above advantages and disadvantages are reflected to varying degrees in different types of housing cooperatives. Below, I consider the strengths of the three types of housing cooperatives for low-income households: Building Cooperatives, Fi nance Cooperatives, and Tenure Cooperatives. Building Cooperatives offer the advantage of collective pooling of resources, but formal cooperatives of this type require substantial up-front capital investment. Hence, formal cooperatives of this type may not serve low-income households, who may not have such savings to start with. Middle-income households are more likely to benefit from the scale economy provided by these societies. However, Building Co operatives could yield themselves to collective self-help and community participation. Low-income households could benefit from pooling their labor in the form of sweat equity in self-help groups for housing construction. However, access to finance and land could be major constraints for self-help groups. Also, such cooperatives are unlikely to contain speculation if they are only involved in plot development or house construction—members could sell their houses to reap capital gains. Often, self-help groups continue to collectively own and manage housing after construction, and they put constraints on the ability of members to sell their housing unit. Finance Cooperatives offer advantages in terms of collective pooling of finan cial resources for low-income households. However, if members need to borrow from formal Finance Cooperatives in India, they need to provide title deeds and other secu rities as collateral. Low-income households may not have access to such collateral, so that formal Finance Cooperatives may not serve their financial needs. Formal Finance C ooperatives usually have a loan ceilin g to allow m ore financing to reach low er- middle and low-income households. However, costs of urban plots could significantly exceed the ceiling limits, thus excluding low-income households from borrowing. 231 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Government subsidies and grants are also provided to finance cooperatives to enable broader financial lending to the Economically Weaker Section (EWS). Finance coop eratives offer better prospects for low-income households as microfinance organiza tions. Instead of using collateral, repayment is ensured through peer sanctions: unless a borrowing member has repaid, others in the group will not get access to credit. However, loans available through microfinancing are usually small, so that such amounts could be used mainly for incremental construction and repairs. Some NGOs also use the microfinance vehicle to pool enough credit so that the groups can ap proach other financial sources for supplemental funding for construction. Tenure Cooperatives also entail collective pooling of resources (land, finance, collective management). In this type of cooperative, land, housing, or both is collec tively owned and managed. The main constraint of Tenure Cooperatives may be ac cessibility to land. If this constraint is overcome, Tenure Cooperatives offers several advantages for low-income households. First, they offer scope for limiting speculation through various internal organizational procedures. Second, Tenure Cooperatives of fer a high degree of community organization and collective action. Third, credit worthiness is enhanced due to collective tenure. Fourth, other collective action fea tures like community action, provision of common infrastructure and amenities, group lending, etc. can also be more effectively carried out. Thus, Tenure Cooperatives offer the most prospects as policy tools for low-income households. A hybrid of the three types of cooperatives could also hold prospects for low- income households. For example, cooperatives could be formed for collective self- help, group lending, and collective ownership of housing. Thus, all three types of co operatives (or hybrids of them) offer different prospects for low-income households. Yet, such advantages exist when cooperatives overcome inherent constraints. While 232 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. internal management is an issue, access to finance or land are major constraints. Co operatives would be able realize their potential for housing low-income households to the extent they can overcome such constraints. Specific cooperative schemes for low- income housing could succeed when they receive subsidies that help overcome some of these constraints (e.g. land or finance at below market rates). Replicability of such schemes is, however, limited if similar concessions are not available at a wider scale. The external institutional framework plays a significant role in this respect. 7.2 Exogenous Level Analysis: The National Institutional Context During the pre-Independence period, housing cooperatives were hardly consid ered as vehicles of housing for low-income households. Tenure Cooperatives in Mumbai were primarily oriented towards the middle class. Talmaki (1931, p. 123), who was a pioneer of cooperative housing in Mumbai, explicitly admitted, “No coop erative housing societies are started among the working classes, either in Bombay or in other parts of the country” (p. 123). According to him, housing cooperatives re quired a large amount of “guarantee capital” from members before they could find outside investors willing to finance their construction. Hence, Talmaki (1920, p. 192) argued, This condition of large initial investment by members has the effect of excluding from the benefit of a housing society those classes of the people who live from hand to mouth and who cannot therefore effect a saving out of their income. Such a society can only cater for the middle classes and upper grades of the lower classes, such as clerks, shop- assistants, artizans and the like. Housing cooperatives in Mumbai hardly served the immigrant labor. Otto Rothfeld, an influential Registrar in Bombay province, observed that forming cooperatives of laborers, although desirable, was inherently problematic since they were mostly single, uneducated men who migrated in and out of the city: 233 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The most desirable object for Co-operative Housing would certainly be the relief of those classes by the provision of cooperative settlements with cheap but healthy tenements suited to working men and their fami lies. So far nothing has been done in this direction. The mill-hands themselves are uneducated and are to a large extent a floating popula tion, consisting in many cases of men alone without their families. To expect them at this stage to assimilate the principles of co-operation or apply them with any unity and consistency seems almost impossible.3 A few feeble efforts were yet made to encourage cooperatives for low-income housing. Hough (1932, pp. 194-195) took issue with the Bombay Registrar that co operative housing was not a solution for housing the poor. She argued: Provided long-term credit is available at low interest, it is only neces sary that the size of the initial payment be reduced to a minimum and that the term of repayment be so adjusted that the difference between their present rental and the monthly payment for the co-operative chawl shall be sufficient to cover the installments and interest. The Whitley Commission, which surveyed labor conditions in 1931, reported attempts by cooperatives to provide additional housing to workers, and recommended loans and subsidies to workers’ cooperatives for housing construction. The Saraiya Committee formed in 1945 was among the initial efforts to recognize use of housing cooperatives for low-income households. It recommended favorable treatment to housing coopera tives for allocation of land, and accessibility to finance: Whenever the Government, municipalities, improvement trusts and similar bodies have at their disposal lands to sell or lease for long terms, for construction of houses, they should give them, preferably to co-operative housing societies at favourable rates. The finance required by the housing societies should be supplied on easy terms by the State or by insurance companies (Government of In dia, 1951, p. 121) 3 Quoted in Mukherji (1923, pp. 308-309). 2 3 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The Committee also recommended setting up provincial cooperatives for providing long term loans to primary cooperatives. Unlike the pre-Independence period, housing policies during the post- Independence period explicitly promoted housing cooperatives as means of housing the low-income households. The Working Group on Housing Co-operatives, 1964, whose recommendations have had a lasting influence on housing cooperatives in In dia, emphasized that “co-operative activity is the best means of providing decent houses at reasonable costs to persons, particularly of low and middle-income groups who cannot afford to satisfy their housing needs by their individual efforts” (Govern ment of India, 1964, p.5). The Working Group formalized Saraiya Committee’s rec ommendation of channelizing finance into the national two-tiered institutional struc ture of Apex and primary cooperatives. Five Year plans also consistently promoted cooperatives as means of housing industrial workers and low-income groups. Cooperative housing programs came to be viewed as “an ideal form of self-help housing” which needed encouragement (Gov ernment of India, 1981). The idea of cooperative as an organizational vehicle for hous ing low-income households gained political currency and was consistently reinforced over time by different committees on housing. The Task Forces on Housing and Urban Development (1983) especially pro moted housing cooperatives for such use. It maintained that cooperatives are “ideally suited for cutting costs, better utilization of land and proper arrangements for mainte nance” (p. 5). It argued that cooperatives gained significance in the Indian scenario where private housing developm ent w as constrained and public housing agencies could not meet the urban housing demand. It concluded: The Task Force strongly recommends that low-income people should be encouraged to form cooperatives in order to meet their shelter re- 235 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. quirements. The principles underlying the cooperative movement and the experience of democratic and self-reliant functioning have much to commend themselves. ... Such cooperatives are playing an increasingly significant role in metropolitan areas where multi-storeyed housing has become necessary due to scarcity of land. (Task forces on Housing and Urban Development, 1983, pp. xxi) The National Commission on Urbanisation (NCU) formed in 1988 also rec ommended an expanded role of cooperatives for low-income housing. Its policy rec ommendations included: allocating land to cooperative societies formed of low and middle-income groups; establishing mutual benefit organizations for the poor; provid ing access to finance and building materials to augment pooling of resources in the collective; and providing soft loans to voluntary agencies that organize poor house holds into such cooperatives (Government of India, 1988, p.221). The National Housing Policy, which went through several revisions between 1988 and 1998 also recognized the role of housing cooperatives for low-income households. According to the policies, “cooperatives and other community-based or ganisations will be encouraged to take up various shelter-related activities, especially for slum-dwellers and the rural poor, assisted through preferential allotment of land, access to finance and fiscal support” (Government of India, 1998b). In the context of economic liberalization and decentralization in the 1990s, national urban slum policies (e.g. National Slum Development Program) and urban employment schemes (e.g. Swama Jayanti Shahari Rozgar Yojana) focus on local development through commu nity organizations such as neighborhood groups, neighborhood committees, Commu nity Development Societies, including cooperatives. The most recent thrust for usage of cooperatives for low-income households came with the national program for construction of 2 million housing units per year for the poor (Economically Weaker Section and Low-income Groups). Initiated in 236 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1999, the scheme consists of construction of 1.3 million housing units in rural areas, and 0.7 million units in urban areas. Of these, 100,000 units have been targeted for construction by housing cooperatives annually. The NCHF has been campaigning for States to allocate 30 percent of government land earmarked for housing to coopera tives and to increase lending to Apex Cooperatives so that cooperatives can achieve their goal under the 2 million housing program. Cooperatives constructed 175,000 units and 58,184 units against their target in 1998-99 and 1999-2000 respectively.4 National housing policies and laws thus consistently supported the develop ment of housing cooperatives for low-income households. However, there is scarce reliable time series data on income groups served by housing cooperatives. The Fourth Five Year Plan claimed that “provision of housing through private or co operative effort is usually directed to the affluent or middle classes,” although it main tained that cooperatives’ role can be expanded for serving the low- and middle-income groups (Government of India, 1970, p. 402). However, according to more recent es timates of the India National Report (Government of India, 1996, p. 33), “cooperatives cater mainly to the economically weaker sections and low-income group households, who account for three-fifth of the housing stock of this sector. The middle-income households account for two-thirds of the cooperative housing stock.” Apex cooperatives’ lending policies usually include ceilings on loan amounts and ceilings on the floor area of the house according to the income category. Interest rates are also higher for larger loan amounts. Such ceilings are intended as egalitarian lending policies and to discourage “luxury” housing. According to NCHF (1991), nearly 52 percent of the loan amounts sanctioned by state level Apex Cooperatives in 4 Ministry o f Urban Developm ent & Poverty Alleviatio, Annual R eport (2002-03). The bulk o f housing construc tion (54 percent) in 1998-99 was in Tamil Nadu through the rural housing cooperatives (about 95,554 units). 237 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1989 were to low-income households, 36 percent were to middle-income group households, and 12 percent were to high-income group households (Table 7.1). Loans to low-income households by Apex Cooperatives in Maharashtra and Tamil Nadu were 57 and 63 percent respectively. With respect to housing units constructed with financial assistance from Apex Cooperatives, about 72 percent of the housing units constructed were for the low-income; 19 percent were for the middle income, and 9 percent were for the high-income households (Table 7.2). The percentage of houses constructed for low-income households in Maharashtra and Tamil Nadu was 67 and 85 percent respectively. Thus, at the national level, routing finance through the two tiered structure of Apex and primary cooperatives, has indeed assisted housing coop eratives to reach down to low-income households. Maharashtra and Tamil Nadu take the lead in terms of providing such finance to low-income households. Table 7.1. Income-group wise distribution of loans sanctioned by Apex Coopera tives, 1988-89. State EWS LIG MIG HIG Total 1. Andhra Pradesh 30% 40% 30% 100% 2. Delhi 70% 30% 100% 3. Goa 3% 72% 25% 100% 4. Jammu and Kashmir 4% 34% 43% 19% 100% 5. Karnataka 4% 29% 52% 15% 100% 6. Madhya Pradesh 9% 24% 57% 10% 100% 7. Maharashtra 11% 46"., 38% 5% 100% 8. Manipur 4% 36".) 42% 18% 100% 9. Meghalya 13% 13% 18% 56% 100% 10. Pondicherry 99% 1% 100% 11. Punjab 2% S% 62% 28% 100% 12. Tamil Nadu 30% 33".. 33% 4% 100% 13. Uttar Pradesh 1% 42% 57% 160% National Average 21% 31% 36 12% 100% Source: NCHF (1991, p. 21). 2 3 8 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 7.2. Income-group wise distribution of houses assisted by Apex Coopera tives, 1988-89. State EWS LIG MIG HIG Total 1. Andhra Pradesh 30% 37% 33% 100% 2. Delhi 70% 30% 100% 3. Goa 6% 75% 19% 100% 4. Jammu and Kashmir 8% 38% 36% 18% 100% 5. Karnataka 10% 53% 32% 5% 100% 6. Madhya Pradesh 10% 35% 49% 6% 100% 7. Maharashtra 15% 52% 29% 4% 100% 8. Manipur 8% 43% 36% 13% l66% 9. Meghalya 40% 17% 14% 29% 100% 10. Pondicherry 100% 100% 11. Punjab 2% 8% 62 28% 100% 12. Tamil Nadu 74% 11% 11% 4% 100% 13. Uttar Pradesh 1% 42% 57% 100% National Average 49% 23% 19% 9% 100% Source: NCHF (1991, p. 23-24). 7.3. Exogenous Level Analysis: Cases of Mumbai and Chennai Although Maharashtra and Tamil Nadu are the two leading states in terms of reaching down to low-income households, formal housing cooperatives in the two cit ies of Mumbai and Chennai have hardly reached down to low-income households. In such a condition, what prospects do housing cooperatives have for serving low-income households in large cities? Comparison at the city level indicates that the nature of the local housing market is significant for the demand for housing cooperatives, and fewer state controls help expand the diversity of activities of housing cooperatives (Chapter 6). Diversification of housing cooperatives’ activities aided in the usage of coopera tives for low-income households in various ways by the public and Third sectors in Mumbia. Hence, I argue that Mumbai offers better prospects for expanding coopera tives’ potential to reach further down the income groups. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 7.3.1 Mumbai During the post-Independence period, real estate and housing prices have been historically very expensive in Mumbai when compared to other Indian cities. Accord ing to Society for Development Studies (1996, p. 17), house price to income ratio in Mumbai was 13.38, while the comparative ratio for Chennai was 7.01. Low-income households were effectively priced out of the formal housing market in this context. The formal private sector caters primarily to high-income groups. Housing coopera tives in the private sector are mainly used as means for transferring ownership and for collective management; they hardly benefit low-income households. These house holds are largely served by informal housing in slums, which accommodates over 50 percent of Mumbai’s population. Hence, cooperatives in Mumbai need to necessarily address informal housing if they are to serve low-income households. Housing cooperatives play diverse roles in Mumbai. They have been used for the benefit of low-income households in four ways in Mumbai, which offers prospects for broadening cooperatives’ roles for reaching down the income groups. First, public housing is provided through housing cooperatives. Large-scale schemes like Land Infrastructure Servicing Program and Slum Upgradation Programs of the Bombay Ur ban Development Project have been implemented through cooperatives. Second, co operatives have been used as vehicles for transferring ownership in inner city areas where landlords have had little incentive to maintain due to rent control. Third, coop eratives form the core of slum redevelopment policies. Fourth, non-government or ganizations use cooperatives in the informal sector to organize slum-dwellers, pave ment dwellers, and indigent women for fighting for their housing rights, for self-help, and for organizing them for microfinance. Such diversity of roles has been mainly facilitated by the relatively greater autonomy of housing cooperatives, as there are 240 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. fewer state controls. The cooperative institutional structure is also rather well devel oped, so that cooperatives have lower transaction costs of formation and functioning. Of course, the above roles of cooperatives could change with changes in the broader institutional framework. For example, rent control reforms—which are in deed required for increasing the efficiency of housing market in Mumbai—may pre empt the need for transferring control to tenant cooperatives. Yet, I argue that the di verse roles played by housing cooperatives in Mumbai offer prospects for coopera tive’s participation for housing low-income households. The first prospect of utilizing cooperatives in Mumbai is that it offers a better mode of providing tenure to slum-dwellers than other organizational forms. Several studies reinforce the need for security of tenure for housing improvements (e.g. Strassmann, 1988). There is also a broad change in public agencies’ attitudes from evicting slum-dwellers to providing tenure (Chapter 3). Cooperatives provide a better alternative since it provides an organizational mechanism to control speculation. In Mumbai, this is achieved in three ways: first, members need to seek permission from the cooperative to sell; second, cooperatives reserve the first right to refuse to buy the house if a household wants to sell; third, cooperatives screen new members. Govern ment regulations also prescribe that the member needs to occupy the house for a stipu lated period of time, and cannot sell the house. Such regulations could be imple mented through peer monitoring in the cooperatives.5 Alternative organizational forms like associations and companies do not embody similar kind of legal control over members since housing is individually owned. 5 To m y knowledge, an empirical study on cooperative’s ability to prevent speculation in Mumbai has not been conducted. International studies, however, confirm such notion (see M iceli, Sazama, and Sirmans, 1994; Sazama, 1996, 2000). Empirical studies on allocation o f land on individual basis, however, confirm “upward filtering” due to speculation (see Johnson, 1987). 241 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The second prospect of utilizing cooperatives in Mumbai is the ease of forma tion and functioning of housing cooperatives. Slum policies in Mumbai require that at least 70 percent of the slum-dwellers consent to form a cooperative for allocating ten ure. Organizing such a cooperative is of course not costless (O’Hare, Abbott, and Barke, 1998). Yet, the idea of cooperative is rather well known in Mumbai and there is less resistance for forming such cooperatives among slum-dwellers (Panwalkar, 1996). Since the support institutional structures are rather robust, the costs of forma tion and functioning of housing cooperatives are low as compared to other organiza tional forms. The third prospect of utilizing cooperatives is the physical design and usage of land. Under the Land Infrastructure Servicing Program, cooperative housing were provided in the form of cluster housing, where housing units were clustered around a courtyard. Cluster housing provided more efficiency in terms of service provision (water, sewage lines) than individually plotted housing (World Bank, 1985). Further more, cluster housing, by the nature of its design, does not allow much expansion of floor space. Such constraint may be beneficial to limit additional speculative con struction; but it also hinders the household’s ability for incremental construction due to genuine needs of household’s growth. The fourth prospect is the usage of cooperatives for organizing collective ac tion, for self-help, and for microfinance. Several NGOs in Mumbai (Society for Pro motion of Area Resources Center; Mahila Milan; Youth Urban Voluntary Action) have been involved in organizing slum-dwellers and pavement dwellers into coopera tives to empower them to fight for their housing rights. Although such cooperatives usually begin as informal cooperatives, they aim towards organizing themselves as formal cooperatives so that they can qualify for collective tenure under slum policies. 242 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. A few NGOs have also utilized cooperatives as vehicles for self-help housing con struction. Lastly informal cooperatives for microfinance (especially among women) mobilize finance of low-income households for emergency purposes, for entrepreneu rial use, and for house savings. Low costs of forming cooperatives offer prospects for such community organizations to emerge. My main point in illustrating the above four prospects is that housing coopera tives are being already utilized for slums and pavement dwellers in different ways. Moreover, the existence of rather robust support institutional structures also lowers their costs of formation and functioning. Although alternative organizational forms like associations also embody collective action, they do not hold the members as cohe sively as the cooperative. 7.3.2 Chennai Unlike Mumbai, housing cooperatives in Chennai are mainly Finance Coop eratives. While activities of Finance Cooperatives grew modestly in Chennai, the ac tivities of Building Cooperatives reduced since Independence. Thus, unlike Mumbai, cooperatives do not have play diverse roles in Chennai, due to both the local housing market and cooperative policies. Preference for individual plotted development pre cluded growth of Tenure Cooperatives. State controls also constrained the diverse roles that cooperatives could have otherwise played. Apart from support institutional structure for finance, there is little support for cooperatives so that the transaction costs of formation and functioning of housing cooperatives are high. The overall ac tivities of housing cooperatives are so low that Chennai offers fewer prospects for ex panding cooperatives’ potential for reaching down the income groups. 2 4 3 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. The Apex Cooperative (Tamil Nadu Cooperative Housing Federation, TNCHF) is the main source of finance for primary cooperatives in Chennai. Figure 7.1 shows the amount of funding from TNCHF and the number of housing units con structed in Chennai. As the chart shows, the funding for primary cooperatives hardly increased between 1990-91 and 1994-95. Primary cooperatives constructed a little over 1000 units at their peak in 1992-93, after which it reduced. Although the activi ties of cooperatives have increased modestly since Independence, their activities in housing construction are hardly significant in Chennai’s housing market. Primary co operatives also mainly serve upto middle-income households. TNCHF receives fund ing for EWS and LIG housing mainly from HUDCO, which is directed towards rural housing programs. Very little of such funding is directed towards urban areas, so that cooperatives mainly serve middle-income households. In the face of overall reduction of TNCHF funding to primary cooperatives in Chennai, cooperatives have not been able to reach down the income groups The principal prospect of Finance Cooperatives in Chennai for low-income households is that such cooperatives form the second largest source of formal financ ing for housing construction (Chapter 5). Yet, instances of microfinancing mecha nisms hardly exist. The main constraint on their activities is the state control. While the increase in the activities of housing cooperatives for rural housing is commend able, it did not have to be accompanied by a parallel reduction in emphasis on urban cooperatives. Since the cooperative institutional structure for financial support is well developed, it could be used for channelizing finance to primary cooperatives in Chen nai too. In this, the financial strength o f TNC H F could be leveraged for both urban and rural cooperatives. 2 4 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 7.1. TNCHF funding and number of Housing Units constructed with such funding in Chennai, 1982-83 to 1995- 96 T3 <D < D d o s o s < 1200 90 1000 75 800 60 600 45 400 200 1 5 0 82-83 83-84 84-85 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 95-96 Year ■ ■ — Amount - ♦ - Units Source: Compiled from TNCHF Annual Reports. to O l Housing Units 7.4 Conclusion: The Role of Institutional Structure Analysis at the national level shows that housing cooperatives have been able to serve low-income households, however, analysis of housing cooperatives in Mum bai and Chennai depict that formal cooperatives have hardly served low-income households. I argue that at the national level, the broad two-tiered national institu tional structure for finance to cooperatives aided in the ability of cooperatives to reach down to low-income households. The wide institutional structure channelized finance to primary cooperatives. Lending policies of the Apex Cooperatives also ensured that bulk of such financing is directed towards middle and low-income households. How ever, other institutional structures vary among states. Administrative, legal and other procedural support, for example varies between states; institutional support for provi sion of land also varies. In the case of Mumbai and Chennai, the institutional structures have again been influential in the role of housing cooperatives. In Mumbai, these structures are fairly well developed, so that the transaction costs of formation and functioning of housing cooperatives are low. Under these circumstances, cooperatives diversified their role, which holds several prospects for low-income households. Cooperatives hold better prospects than other organizational modes for fulfilling the housing needs of low-income households and for addressing the issues of informal housing. In Tamil Nadu, in the context of state emphasis on cooperatives as finance mechanisms (especially for the rural areas), the support structure for finance is well developed. Consequently, although cooperatives are mostly involved in providing housing finance in Chennai, funding from TNCHF leveled off in the 1990s for urban areas. While broadening the scope of cooperatives is indeed beneficial for rural hous ing, it has deprived urban areas. Cooperatives lost their autonomy due to state control, 246 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and have hardly diversified their activities. Similar to Mumbai, diversification could have been beneficial for low-income households in Chennai. 2 4 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8. Conclusion As I highlighted in Chapter 1, there are two central questions in my disserta tion: (a) what are the institutional conditions under which housing cooperatives grow? and (b) what is the scope of cooperatives in fulfilling the housing objectives of low- income households? These two questions have received little attention in housing re search. The first question is significant to understand the policy framework for en hancing the role of cooperatives in the housing market. The second question is par ticularly significant for developing countries like India where government policies have often promoted cooperatives as means of housing low-income households. An swering the questions is particularly relevant in the present context of domestic and international emphases on improving institutional and organizational mechanisms in the housing process. Analysis of evolution of housing cooperatives in India in general, and in Mumbai and Chennai in particular, holds broader policy answers for the above two questions. India has had a long history of evolution of housing cooperatives since the early twentieth century. While cooperatives were active in only two provinces (Bom bay and Madras) before Independence, they proliferated nationwide after Independ ence. Other states in India followed the model of either of the two provinces in pro moting cooperatives. Other countries have also promoted cooperatives with varying degrees of success. Hence the research does not only hold broader policy lessons for other Indian states, but also provides policy insights for development of cooperatives in other countries. In this chapter, I summarize the policy conclusions that answer the above two questions, and examine the relevance of these conclusions with respect to other coun- 248 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tries’ experiences. I examine the experiences of three countries in particular with re spect to housing cooperatives: Sweden, Turkey, and United States. Put together, these three countries represent a rich variety of institutional conditions in which housing co operatives have emerged. While cooperatives’ activities are quite pronounced in Sweden and Turkey, their activities are very limited in the United States. The chapter is organized as follows. First, I outline my analytical framework. Second, I summarize the broader policy conclusions with respect to the institutional conditions under which housing cooperatives grow. Third, I identify the institutional potential of housing cooperatives for low-income households. Fourth, I highlight the overall significance of cooperative institutional structure in the development of hous ing cooperatives and in enhancing their potential for serving the housing needs of low- income households. Fifth, I examine how the policy lessons relate with experiences of Sweden, Turkey, and United States with respect to evolution of housing cooperatives. Sixth, I conclude with the broad policy lessons. 8.1. Analytical Framework To answer the two questions posed above, I analyze the evolution of housing cooperatives in India at both endogenous (organizational) and exogenous (institu tional) levels (Chapter 2). Analysis at the endogenous level is useful to identify inter nal strengths and weaknesses of the cooperatives in controlling agency problems. At this level, cooperatives’ governance structure, their membership characteristics, and the nature of housing objectives gain significance. Cooperatives’ governance structure is characterized by collective ownership and democratic management, which endows it with both collective action prospects and problems. While cooperatives face inter nal management problems, they could also face problems in access to finance and land. In terms of membership, cooperatives are not explicitly oriented towards any 249 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. particular group of individuals (e.g. income group, class, etc.); members could join cooperatives if the benefits they derive from the collective (centripetal forces) out weigh the benefits from individual operation (centrifugal forces) (Bonus, 1986). In terms of housing objectives, cooperatives could be formed in response to as set specificity in production and consumption processes. Three types of housing co operatives could be identified in India based on their objectives: Building Coopera tives (which have been engaged in building construction and land development); Fi nance Cooperatives (which have been engaged in providing housing finance); and Tenure Cooperatives (which own housing collectively, and have been usually engaged in housing maintenance and management). While Building Cooperatives are least sta ble (due to finite longevity of production process), Tenure Cooperatives are most sta ble (due to perpetuity of ownership). Finance Cooperatives are stable as long as their financial base is in tact; however, they could have a high turnover rate of members since these members could exit if their repayment commitments are over. Analysis at the exogenous level is useful in identifying the elements of the in stitutional environment that constrain or enable the growth of housing cooperatives. At this level, institutions and the local housing market characteristics gain significance for the extent to which housing cooperatives emerge. Institutions form the incentive framework in which organizations like housing cooperatives evolve, and fundamen tally affect the transaction costs of housing cooperatives. In this, the political eco nomic context gains significance, since it shapes the institutional environment. The local housing market characteristics determine the demand for collective organizations like housing cooperatives. For exam ple, housing m orphology is important for the emergence of housing cooperatives (Tenure Cooperatives may be required for collec- 250 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tive management in multi-family housing; such demand may not exist for single family housing). I argue that both levels of analyses are required, and that ignoring one level of analysis may lead to partial lessons. As I propose in the next section, focusing exclu sively on either level ignores the tension that exists between endogenous organiza tional principles and the exogenous institutional framework. In synthesizing the two levels of analysis, I highlight the significance of cooperative institutional structure, which cuts across both levels of analyses. Institutional structure refers to the manifest relationships between organizations, both between cooperatives and between coopera tives and other organizations. I propose that such supportive institutional structure provides a reinforcing mechanism for cooperatives to persist and to enhance their ac tivities. Moreover, support provided by the institutional structures reduces the transac tion costs of formation and functioning of housing cooperatives. 8.2. Institutional Conditions for Growth of Housing Cooperatives Cooperatives were not indigenous organizations for India. They were neces sarily products of the state during British rule, and they emerged with the passage of Cooperative Societies Act, 1904. Housing cooperatives started to function after the Act was amended in 1912 to include non-credit cooperatives, including housing. Until Independence, housing cooperatives were active mainly in Bombay and Madras prov inces. While housing cooperatives emerged principally as Tenure Cooperatives in Bombay province, they emerged mainly as Finance Cooperatives in Madras province. In the post-Independence period, housing cooperatives grew phenomenally nation wide, and other states followed either of the two models. The two states of Maharash tra and Tamil Nadu that were carved out of Bombay and Madras provinces respec tively, maintained a leading edge in terms of cooperatives’ activities in India. How- 251 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ever, there is a differential evolution in the role of housing cooperatives in Mumbai and Chennai, the respective capitals of the two states. While housing cooperatives have grown enormously and have taken on diverse roles in Mumbai (Chapter 4), co operatives have remained principally as Finance Cooperatives in Chennai. Although Building Cooperatives were started in Chennai after Independence, their activities re duced over time (Chapter 5). We need to examine the above evolution of housing cooperatives in India at both the endogenous organizational level and the exogenous institutional level to draw broader lessons on institutional conditions in which housing cooperatives grow. Agency theorists focus on the endogenous organizational level characteristics for their explanation of growth or decline of cooperatives. On the other hand, Marxist critiques like Elster (1989) ignore the endogenous organizational problems of cooperatives as something that could be overcome. Rather, they focus on the exogenous institutional context in their examination of the growth of housing cooperatives. In the Indian con text, there has been a close relationship between cooperatives and the state histori cally. Hence, neither level of analysis can be ignored, and leaning either way too much could be problematic. Moreover, focus on one level of analysis ignores the crosscutting issues that arise at both endogenous and exogenous levels, which are important for drawing more informed policy lessons. For example, focusing exclusively on either level ignores the tension that exists between endogenous organizational principles in terms of coopera tive autonomy and the exogenous institutional framework in terms of state controls. This tension is clearly evident in the Indian case. In Mumbai, cooperatives have his torically enjoyed great degree of autonomy. Cooperatives grew phenomenally in the post-Independence period, and have diversified their roles in the housing market. In 252 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chennai, cooperatives have had little autonomy. State interference in the daily activi ties of cooperatives through government nominees appointed under Tamil Nadu (Spe cial Officers) Act, 1976 restrained the activities that cooperatives could have other wise performed in Chennai. Although state involvement has been beneficial in widen ing the geographical focus of housing cooperatives more broadly to rural areas, such benefits have come at a cost. Cooperatives became more state dependent and have hardly diversified since Independence. While activities of Finance Cooperatives grew, activities of Building Cooperatives reduced since Independence. Autonomy is significant for cooperatives, since they are self-managed, member controlled organizations. Yet, these organizations also face internal collective action problems and difficulties in access to finance and land, so that they require external assistance. While the state could provide such external assistance, it comes with the intrinsic danger of the state controlling the functioning of housing cooperatives. Such state controls could lead the development of housing cooperatives in the directions where the state puts an emphasis. Consequently, autonomy of housing cooperatives could be compromised and stymie the diverse activities that cooperatives could other wise perform. From an endogenous organizational perspective, autonomy is significant for cooperatives. First, as highlighted above, autonomy is instrumentally important for cooperatives to diversify their activities. Second, autonomy is also intrinsically impor tant for cooperatives’ governance structure. Compromising autonomy implies that cooperative management boards cannot control the internal decision making proc esses, w hich is one o f the fundam ental elem ents o f cooperative’s organizational char acteristics. From a policy perspective, this leads to the first policy conclusion, that is, housing cooperatives require autonomy for their development. 253 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. From an exogenous institutional perspective, the question remains as to what the state’s role should be while maintaining cooperative’s autonomy. Cooperatives could fall into the trap of state dependence and control if they seek direct state support. Shah (1996) argues that cooperatives “often get co-opted by the state and the national political elite into the larger policy framework which invariably draws them into po litical and bureaucratic power play” (p. 39). Although state controls may be well in- tentioned and have some beneficial effects, they constrain the autonomous growth and expansion of cooperatives’ activities in the long run. While state involvement is in deed required for cooperatives to comply with basic legal principles, overt interference in the decision making processes compromise cooperative autonomy. This leads to the second policy conclusion, that is, state controls could constrain the growth o f housing cooperatives in the long run. As the Indian case depicts, although the above two policy measures may be normatively desirable, such measures are obviously not carried out in a political and institutional vacuum. In the Indian context, the national government exhibited an “en dogenous preference” in Elster’s (1989) terms for cooperatives in the post- Independence political economic context of democratic socialism. State support, however, did not translate into state controls—the national government could not un dermine the autonomy of housing cooperatives as both housing and cooperatives were state government subjects. State governments have more significance in India for the extent to which the government controls the functioning of housing cooperatives. The nature of such con trols evolved in the context of specific historical and institutional factors, as illustrated in the evolution of housing cooperatives in the two cities of Mumbai and Chennai. As Chapter 4 shows, Mumbai historically had pro-active cooperative leaders, who strived 254 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to maintain independence of cooperatives. The Cooperative Registrar’s Office played an advisory role in this context to govern housing cooperatives. In contrast, coopera tives in Chennai became politicized, with the Registrar’s Office played a controlling role and overtly interfering in the daily activities of cooperatives. While cooperatives have retained their autonomy in Mumbai historically, cooperatives in Chennai lost such autonomy. This leads to the third policy conclusion, viz, cooperative policies are themselves contingent on local institutional history. Cooperative autonomy or overt state controls are not determined a priori', rather they are set in the specific political and institutional conditions. Independent institutional structures could provide a greater degree of coopera tive autonomy from state controls. As I argue later (Section 8.3), the development of a national institutional structure supporting cooperatives acted as a reinforcing mecha nism in the growth of housing cooperatives in India. Although these institutional structures were state sponsored in the political economic context of democratic social ism, they gained an independent life of their own, and continued to support growth of housing cooperatives despite India’s political economic transition towards economic liberalization. This leads to the fourth policy lesson that the state could play a cata lytic role in establishing independent cooperative institutional structures. Of course, independence of such institutional structures cannot be defined a priori', these struc tures can themselves become state controlled, as in the case of Chennai. Yet, such structures reduce direct dependence of cooperatives on the state. Lastly, the local housing market plays a critical role in the evolution of housing cooperatives. First, the local housing market may or may not be ready for adapting cooperatives. Second, specific types of housing cooperatives could emerge in re sponse to the particular demands of the local housing market, as illustrated in the case 255 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of Mumbai and Chennai. The rather detailed history of pre-Independence local hous ing market in Mumbai (Chapter 4) shows that multi-family housing units emerged in the city in the context of its sudden economic boom and population growth due to im migrant male laborers in the late nineteenth century. “Modem” urban development following the economic boom almost erased the city’s underlying rural character. In this context, Tenure Cooperatives emerged as “modem” organizational vehicles for providing decent housing for the middle class through collective ownership and man agement in multi-family housing. Unlike Mumbai, Chennai’s economic growth was sluggish during pre- Independence period and grew mainly as a “city of villages” where the internal mral stmcture of the city was maintained (Chapter 5). Mainly families migrated from nearby rural areas, and there were strong rural-urban connections socially. Conse quently housing followed the mral settlement pattern, and housing development was predominantly in the form of plotted development. While housing in Mumbai grew in response to accommodate single-male immigrant laborers, housing in Chennai emerged to accommodate families. Thus housing in Mumbai consisted of single room units in multi-family housing, and housing in Chennai consisted of single-family units. With population growth, Mumbai expanded on the vertical plane, and Chennai densi- fied on the horizontal plane. Little demand emerged for collective organizations like Tenure Cooperatives for housing management in Chennai. In this context, housing cooperatives were adapted as financing mechanisms for individual housing. The local housing market thus affected the specific types of housing cooperatives that emerged in the two cities. This leads to the fifth nolicv conclusion, that is, the nature o f local housing market is critical for the specific type o f development o f housing cooperatives, and housing policies should reconcile with such local market conditions. 256 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8.3. Potential of Housing Cooperatives for Low-income Households In the pre-Independence period, housing cooperatives in India were hardly considered for low-income households. However, in the post-Independence period, national housing policies in India consistently promoted housing cooperatives as means for housing low-income households. National statistics on housing coopera tives do indicate that cooperatives cater to low-income households, and the two states of Maharashtra and Tamil Nadu maintain a leading edge in this respect. However, low-income households have had limited access to formal housing constructed by co operatives in the two cities of Mumbai and Chennai. Yet, Mumbai offers more pros pects of housing cooperatives for reaching further down the income groups, as indi cated by the use of diverse roles of cooperatives for low-income households. Finance Cooperatives in Chennai serve up to lower middle-income households, who have sta ble employment opportunities or have collateral. Other than this, housing coopera tives have hardly been used as vehicles for low-income households. To examine the potential of housing cooperatives for low-income households, we need to examine the incentives offered by the internal organizational features of cooperatives for low-income households. However, as the Indian case depicts, such potential may or may not be achieved due to the institutional conditions in which housing cooperatives function. Thus, examination at both the endogenous organiza tional level and the exogenous institutional level again become significant for identify ing the conditions in which housing cooperatives could reach low-income households. At the endogenous level, housing cooperatives are not necessarily oriented to wards any particular income group (Chapter 7). Yet, the collective feature of coopera tives’ governance structure offers organizational potential for housing low-income households in several ways. First, cooperatives entail collective pooling of resources, 257 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. which offers economy of scale. Second, credit worthiness of low-income households rises due to collective pooling of resources. Group lending and micro-finance pro grams usually entail formation of a cohesive network of households for lending, wherein cooperatives offer an alternative for group formation. Third, cooperatives foster community action, community empowerment, and act as organizational vehicles to foster collective action and self-help. Fourth, speculation and gentrification can be potentially contained in cooperatively owned housing. However, the above features may not be achieved due to collective action problems endemic to the cooperative. First, housing cooperatives, especially for low- income households, face entry problems and are seldom voluntarily formed. Usually an external agency (public housing agency or a NGO) is required to promote such co operatives. Second, self-management can be especially problematic in cooperatives of low-income households due to the lack of managerial, technical, legal, or financial ex pertise by members. Third, low-income households face opportunity costs between participating in the collective and other productive income earning activities. House holds will participate if benefits from the collective outweigh costs of participation. Fourth, cooperatives also obtain the collective action problem of free-riding. In addition to the above internal organizational problems, cooperatives of low- income households also face the problems of access to finance and land. Although credit-worthiness of low-income households may indeed rise due to pooling of their resources in the group, these cooperatives do not have broad access to finance. Lend ing to such cooperatives is not yet widely institutionalized, and usually requires a NGO as an intermediary between the donor and the cooperative. Moreover, land and building costs in large cities may not be affordable for low-income households despite the economy of scale obtained through the group. Thus, all three issues of internal 258 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. management, access to finance, and access to land become critical in determining the potential of cooperatives to serve the housing needs of low-income households. At the endogenous level, cooperatives require a certain degree of autonomy and fewer institutional restrictions to realize their internal strengths. In Mumbai, the autonomy helped cooperatives play diverse roles to the benefit of low-income house holds. First, public housing schemes have been implemented through cooperatives. Second, cooperatives have been used as vehicles for transferring ownership in inner city areas where landlords have had little incentive to maintain due to rent control. Third, cooperatives form the core of slum upgradation and redevelopment policies. Fourth, non-government organizations use cooperatives in the informal sector to or ganize slum-dwellers, pavement dwellers, and indigent women for fighting for their housing rights, for self-help, and for organizing them for microfmance. Cooperatives have hardly been used for such wide range of activities in Chennai. This reinforces the first policy conclusion. While autonomy is desirable for cooperative’s develop ment, it is also helpful to serve low-income households. At the exogenous level, cooperatives also require institutional support to over come the critical areas where cooperatives face weaknesses. These critical areas in clude internal organizational management, access to finance, and access to land. Of ten, state support is provided to specific cooperative schemes for low-income housing in the form of subsidized finance or land. Such project specific subsidies, however, may not be replicable if similar concessions are not available at a wider scale. Hence, policies need to ensure broader institutional mechanisms through which internal or ganizational problems of cooperatives could be overcome. As the Indian case depicts, the establishment of national institutional structures for administrative support, finan cial support, and access to land helped in overcoming some of the internal weaknesses. 259 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This reinforces the fourth policy conclusion in terms of strengthening institutional structures, where the state could play a catalytic role. As I have highlighted in Chapter 7, one of the principal issues at the exogenous level is the transaction costs of formation and functioning of housing cooperatives. The costs of formation for housing cooperatives include entry and establishment costs. The costs of functioning include administrative and monitoring costs, costs of resolv ing disputes (between members or cooperatives or both), and so on. Higher transac tion costs—that is, higher costs of formation and functioning—could lower the range of activities of cooperatives in a given institutional context. Higher transaction costs could also deter low-income households from forming cooperatives. Lowering the transaction costs is thus significant for housing cooperatives to reach further down the income groups. This leads to the sixth policy conclusion, that is, cooperative policies need to lower transaction costs offormation and functioning o f housing cooperatives. Ambiguously defined procedures in formation, inefficient approval mechanisms, or prohibitions on functioning of cooperatives could raise the transaction costs of coop eratives. Streamlining such procedures and providing greater autonomy could lower the transaction costs of formation and functioning of cooperatives. The cases of Mumbai and Chennai support the above policy conclusions. In Chapter 7, I have argued that Mumbai offers better prospects for use of cooperatives for low-income households. The transaction costs of formation and functioning of housing cooperatives are very low as compared to other organizational forms. The concept of forming cooperatives is widely known in Mumbai, and there is little resis tance for forming such cooperatives among slum-dwellers (Panwalkar, 1996). Hence, the internal organizational advantages of cooperatives have indeed been capitalized. Cooperatives have been used by NGOs for organizing collective action, for self-help, 260 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and for microfmance; public agencies have used cooperatives to allocate collective tenure to slum-dwellers. Although enabling laws for condominium (apartment) own ership exist, cooperatives have been more popular due to such low costs. In Chennai, cooperatives have limited role due to restrictions imposed by the Tamil Nadu Special Officers’ Act, 1976. The transaction costs are so high that cooperatives are hardly considered in formulating low-income housing policies. The high costs have also de terred the growth of cooperatives in the private sector; multi-family housing is primar ily owned as condominiums. 8.4. Significance of Institutional Structure Development experts like Peter Bauer (1961) concluded in the early years of cooperatives’ development that they grew in India because of direct state support. Under the leadership of Nehru, the post-Independence context of democratic socialism was conducive for such direct state support. Bauer (1961, pp. 75-77) argued against providing such preferential state support to cooperatives. I argue that indeed the po litical economic environment of democratic socialism provided a favorable environ ment for emergence of housing cooperatives. For, the socialist ideology of collective was in alignment with that of cooperative’s internal principles. However, direct state support, by itself, does not explain the phenomenal growth of housing cooperatives in the longer run during the post-Independence period. First, cooperatives had very small role in implementing government schemes, so that housing cooperatives had limited state support. Second, although direct state support waned with the transfor mation of India’s political economy towards economic liberalization in the 1990s, the activities of housing cooperatives continued to grow. I argue that cooperative institutional structures have played a key role in the development of housing cooperatives in India. Institutional structure refers to the 261 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. manifest arrangement o f relationships between organizations fostered by the institu tional framework. In the context of cooperatives, the institutional structure refers to the relationships of cooperative organizations with each other and with other organiza tions. The institutional structure provides a reinforcing mechanism for cooperatives to persist and to enhance their activities. Vanek has similarly argued that a “supporting structure” is needed for the propagation of cooperatives.1 Horvat (1982, p. 459) also argues, “cooperatives can develop only if they are integrated economically into a sec tor.” As the Indian case depicts, support cooperative institutional structures enable the growth o f housing cooperatives. The institutional structure mediates both the endogenous organizational fea tures of cooperatives and the exogenous institutional framework. At the endogenous organizational level, the structure provides support for enhancing the internal collec tive strengths, while mitigating the internal weaknesses of cooperatives. At the ex ogenous level, the supportive structure is crucial for reducing transaction costs in volved in the formation and functioning of housing cooperatives. The specific form of cooperative institutional structure is also a function of both the endogenous features of cooperatives and the exogenous features of the institutional framework. In the Indian context, the institutional structures clearly bear the imprint of the state. The institutional structure could support the development of cooperatives in three ways, which act at both the internal organizational level and the external institu tional level: (a) by providing administrative support; (b) by providing financial sup port; and (c) by providing support for access to land. First, providing administrative, legal, and procedural support could enhance the internal strengths while overcoming the internal weaknesses of cooperatives. Administrative support could facilitate pri 1 Quoted in Horvat (1982, p. 459). 262 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mary cooperatives to overcome some of the collective action problems inherent in co operatives, especially in terms of monitoring. Legal support could reduce the conflicts that arise between members or between members and cooperatives. Lastly, procedural support could be in the form of establishing routine procedures for formation and functioning of cooperatives. Such support is crucial for reducing the transaction costs of formation and functioning of housing cooperatives. Second, the institutional structure could enhance access to finance, which is a key component of housing. As cooperatives are member organizations, they usually raise their finances through their members. However, given the inherent characteris tics of housing market in terms of dedicated assets and lumpy investments, finance raised through members needs to be leveraged with external funding for adequate in vestment in housing. Cooperatives could also face obstacles in obtaining such finance in the open market, i.e. their transaction costs for obtaining finance could be high. In the absence of external financial support, cooperatives could grow only to the extent they are able to mobilize finance internally for productive housing investments. Third, the institutional structure could enhance access to land, which is another key component of housing. As housing market is characterized by high degree of site specificity, cooperatives typically require large parcels of land for contiguous housing development. Cooperatives could face high transaction costs in access to such land. In the absence of access to land, formation of cooperatives may be limited. Both housing finance and land are two critical factors of housing, where exogenous institu tional support is required to enhance the activities of housing cooperatives. Such sup port is required to reduce the transaction costs of functioning of housing cooperatives. In the Indian context, the state played a central role in establishing the coopera tive institutional structures. The British established the Registrar’s Office under the 263 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1904 Act. The office has since gained immense importance over the years, and has been a deciding factor in the development of cooperatives in the states. The Regis trar’s Office in each state oversees the functioning of cooperatives, and is considered as the gods of birth, life, and death of cooperatives.2 Indeed, autonomy of coopera tives centrally depends on the role accorded to the Registrar’s office by the state gov ernment. In Mumbai, the Registrar’s office plays more of an advisory role and inter feres very little in the daily activities of cooperatives. Disputes are handled through separate Cooperative Courts. Additional structures for providing procedural support and legal, technical, and financial advice also emerged in Mumbai. Consequently, the transaction costs of formation and functioning of cooperatives are low in Mumbai. In Chennai, the Registrar’s office plays more of a controlling role, interfering in the daily activities of cooperatives. Few other structures exist to provide administrative, legal, and procedural support. Consequently, transaction costs of formation and functioning of cooperatives are high. Other institutional structures were also essentially established with state sup port. Cooperative institutional structures for finance, for example, were established in the political economic context of democratic socialism, which favored the setting up of such structures. The Working Group on Housing Cooperatives in 1964 formalized the two-tiered structure of Apex and Primary cooperatives in each state, whereby Apex cooperatives mobilize finance for onlending to primary cooperatives. Apex co operatives borrow from various sources (Life Insurance Corporation, Housing and Ur ban Development Corporation, Cooperative Banks, and so on), and on-lend such money to primary cooperatives. Primary cooperatives thus have a steady source of finance, which has reduced their transaction costs of obtaining finance. Presently, al 2 These gods are Brahma, Shiva, and Vishnu respectively. 264 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. most all states have an Apex cooperative, which has facilitated the nationwide growth of housing cooperatives. Indeed, the two tiered financial structure is the most impor tant for having facilitated a nationwide growth of housing cooperatives. Cooperative institutional structures for land are less well defined as compared to that of financial support. National policies consistently supported cooperatives in the preferential allocation of land. State support for land has been provided either di rectly (e.g. through preferential allocation of government land to cooperatives) or indi rectly (e.g. by providing exemption from land acquisition under Urban Land (Ceiling and Regulation) Act, 1976). However, allocation of land to cooperatives principally depends on the state governments, especially the local housing agencies. In Mumbai, public housing agencies consciously implemented mass housing programs through housing cooperatives; in Chennai the public housing agency hardly included coopera tives as implementation mechanisms. The structure for land allocation is perhaps more developed in New Delhi (see Appendix 1), where the public housing agency has had an explicit policy of allocating land to cooperatives. 8.5. Housing Cooperatives in Sweden, Turkey, and United States In this section, I compare the Indian experience with three other countries’ his torical experiences to examine if the lessons I have drawn above are more broadly ap plicable to the international context. I have chosen Sweden, Turkey, and United States to compare such experiences. These three countries provide a rich variety of institu tional conditions under which housing cooperatives have evolved. While democratic socialism has been the dominant state ideology of Sweden historically, United States is an advanced capitalist country. Although Turkey has not had a similar dominant state ideology, it had followed import substitution policies before liberalizing its econ omy in the 1980s. Moreover, Turkey has intermittently had military rule. While 265 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sweden and United States are “developed” countries, Turkey is a “developing” coun try. Lastly, while cooperatives’ activities are quite pronounced in Sweden and Turkey, their activities are quite limited in the United States. Comparing Indian experience with these three countries thus provides an international basis for testing the policy conclusions. Below, I first make a brief historical review of evolution of housing co operatives, and then make the comparative study for applicability of policy lessons. 8.5.1. Sweden In his historical review of the phenomenal growth of cooperative housing in Sweden, Bengtsson (1992) concludes, “the development is probably best described as mutually reinforcing processes of internal organizational consolidation and external institutional integration” (p. 102). At the organizational level, activities of housing cooperatives—principally in the form of Tenure Ownership Cooperatives (TOCs)— grew enormously since the 1920s. As Table 8.1 shows, the overall share of coopera tive housing stock rose from 4 percent in 1945 to 15 percent in 1990. The share of cooperatives in multi-family housing was, however, much higher: by 1945, housing cooperatives accounted for nearly 11 percent, which more than doubled in 1975 to 24 percent (Lundqvist, 1988, p. 39). Table 8.1. Tenure by share of housing stock in Sweden, 1945-1990 1945 1960 1970 1980 1990 Cooperative (Tenant Owner) 4% 9% 13% 14% 15% Private rental 52% 43% 30°., 21% 20% Public rental 6% 14% 23"., 24% 25% Home ownership 38% 34% 34% 41% 40% Total 100% 100% 100% 100% 100% Source: Turner (1997, p. 478). At the institutional level, formation of Tenant Ownership Cooperatives (TOCs) was facilitated with the Tenant-Owner Act in 1930, which defined the relationships 266 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. between members and cooperatives. TOCs also grew due to Swedish law on home- ownership—individual households are prohibited from individual ownership of an apartment in multi-family housing. Hence multi-family housing units are either coop eratively owned or are rented. Moreover, the Social Democratic Party has held power alone or in coalitions since 1932 in Sweden, except for brief periods during 1976-82 and 1991-94. Even the short-lived non-socialist governments favored converting rental housing to cooperatives, but such conversion was carried out only in a limited way in private rental housing. Thus the overall political economic context of democ ratic socialism provided a favorable environment for development of housing coopera tives. Cooperatives’ activities increased due to two reasons. First, they played a sub stantial role in the Million Program, which was a housing construction program launched between 1965 and 1974. Cooperatives accounted for nearly 16 percent of housing constructed under the program (Bengtsson, 1993, p. 15). Second, coopera tives’ role expanded with the emergence of Municipal Housing Companies (MHCs) after World War II. Between 1945 and 1990, there was a significant increase in public rental housing (which are primarily MHC housing) from 6 percent to 25 percent (Ta ble 8.1). Although municipal boards own MHC housing, much of the housing is man aged through TOCs affiliated with HSB or Riksbyggen, which are two of the largest cooperative institutional structures in Sweden (see below). The role of housing cooperatives reduced in the aftermath of Million Program due to reduction in construction of multi-family housing, and increase in single-family housing construction. H ow ever, cooperatives w ere also adapted into the single fam ily housing market in the 1980s—cooperatives produced about one-third of the single family housing in the late eighties (Bengtsson, 1993, p. 16). Demand for cooperative 267 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. housing also increased, as indicated by the increase in number of people queuing for HSB housing from 110,000 to 260,000 between 1980 and 1988. TOCs could tradi tionally receive subsidized loans upto 99 percent of housing construction costs, the rest 1 percent coming from members’ initial deposits. Cooperative housing in Sweden emerged essentially as social housing for the working class, wherein dwellings were provided with basic housing amenities (e.g. cooperative housing came with bathrooms, which were considered a luxury for work ers before World War II). HSB housing also included social amenities, such as laun- drettes, refuse chutes, children’s play areas, etc. Although social housing in Sweden serve low-income households, they do not serve the very poor. In the early years of cooperatives’ growth (1927 to 1939), between 42 to 46 percent of HSB members were workers and craftsmen (Bengtsson, 1993, p. 33). According to Turner (1995), the household structure in cooperative housing resembled that of rental housing in the post war period, although the average income was somewhat higher in cooperatives. Co operatives largely serve young households and elderly householders. According to Lundqvist (1988, p. 52), the gross income of cooperative households fell between renters and homeowners. Lundqvist, Elander, and Danermark (1990) argue that al though newly constructed cooperative housing is still egalitarian, older cooperative housing became increasingly commodified since the 1980s, due largely to speculation in the wake of government deregulations. A significant aspect of the mutual reinforcement between the internal organiza tional features and the external institutional environment is the cooperative institu tional structure, which emerged differently from India. The institutional structures emerged endogenously within the cooperative system, with little state assistance. Be ing independent, cooperatives have a greater degree of autonomy in their decision- 268 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. making processes. Three such independent structures can be identified in Sweden: the Swedish Central Organization of Tenant Owners Cooperatives (SBC) in 1921; Ten ants Savings and Building Society (HSB Society) in 1923; and Cooperative Building Organization of the Swedish Trade Union (Riksbyggen) in 1940. All three organiza tions have Tenant Ownership Cooperatives (TOCs) as members. HSB and Riks byggen together represent the major part of cooperative development in Sweden. By 1991, about 45 percent of cooperatively owned dwellings were affiliated with HSB and about 22 percent with Riksbyggen. In the same year, there were 3700 TOCs affili ated with HSB; 1800 TOCs affiliated with Riksbyggen', and 6000 independent TOCs (about half of which were affiliated with SBC) (Bengtsson, 1993, p. 31). HSB is organized at three levels: primary TOCs, regional HSB societies, and HSB National Federation. The National Federation provides technical, financial, le gal, and administrative support to the 59 regional HSB societies. HSB societies under take construction and allocate housing to members. HSB runs its own savings bank affiliated with the National Federation. Members deposit money in the bank to join the queue for house allocation. In Riksbyggen, housing is owned by trade unions. Its structure is similar to HSB, with 26 regional officies. However, it does not have a bank of its own—it has special agreements with other local savings bank for obtaining loans. Also, in Riksbyggen affiliated TOCs, building guilds and union owned building company undertake construction. In the SBC structure, support to primary coopera tives is relatively weaker than the other two and TOCs are more independent. SBC has five regional offices, and also has its own home savings system started in 1987. 8.5.2. Turkey State policies fundamentally influenced the evolution of housing cooperatives in Turkey. Kemal Ataturk, the country’s Founding father, promoted cooperatives in 269 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. different sectors since the country’s Independence in 1923. However, the activities of housing cooperatives were limited until World War II. In 1945, the number of hous ing cooperatives reached 60 with about 5,000 members. After World War II coopera tives’ activities began to expand with increasing access to finance and land. In 1962, there were 1,600 cooperatives with 150,000 members. Housing cooperatives received subsidized credit from various sources, including Housing Credit Bank and Social In surance Organization. In 1969, a new cooperative bill also created opportunities for cooperatives to obtain land cheaply (Cakin, 1991). Despite such concessions, coop erative housing construction did not increase dramatically during the 1970s due to the oil crisis. Cooperative’s activities grew phenomenally in the 1980s when mass housing bills were introduced. The Mass Housing Fund created in 1984 provided loans to co operatives on a preferential basis. The number of housing cooperatives, which had stagnated during the 1970s, surged to nearly 4,000 in 1985 with a membership of nearly 300,000. Of the total number of housing cooperatives formed between 1941 and 1991, about 79 percent were formed in the post 1980 period. The share of coop eratives in housing construction (based on occupancy permits), which was 5.2 percent in 1970 and 8.7 percent in 1980, rose sharply to 25.2 percent in 1990 (Bugra, 1998, p. 308). Housing cooperatives in Turkey have been mainly Building Cooperatives, in volved in housing construction, although many of them have also been transformed into Tenure Cooperatives in multi-family housing The Turkish government provided subsidized credit and land for mass housing to cooperatives so that families of limited means could gain access to housing. There are, however, few systematic studies on the extent to which housing cooperatives have reached down the income groups. According to Pamuk (1996, p. 105), cooperatives 270 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. have “remained primarily a housing finance solution for middle-income households for first time homeowner ship.” According to Turel (2000), however, about 85 percent of credits through Mass Housing Fund had been given to cooperative members at sub sidized interest rate, so that housing acquisition through cooperatives required much less equity per household as compared to other methods. Unlike India and Sweden, the institutional structures supporting housing coop eratives are not well developed, principally because cooperatives received direct state assistance in terms of subsidized credit and land. The procedures for establishing co operatives are quite cumbersome (Pamuk, 1991, p. 18), and there is little administra tive, legal, and procedural support mechanism. Yet, the state has been promoting for mation of such supportive structures in the form of secondary associations of coopera tives through various concessions. Two such major associations have currently emerged. The first is KENTKOOP (Union of Batikent Housing Construction Coop eratives), Batikent (near Ankara), which was founded in 1979 by trade unions, asso ciations of tradespeople, artisans, and other groups in cooperation with the Metropoli tan Municipality of Ankara. The second is Turkkonut (Central Union of Construction Cooperatives of Turkey), which provides technical assistance to cooperatives. 8.5.3. United States Although housing cooperatives emerged in the United States since the early decades of the twentieth century, their activities are sparse and their overall role in the housing market is marginal. At the endogenous organizational level, housing coopera tives have hardly been able to mobilize themselves as collective vehicles for housing. Although condominiums emerged later during the 1960s, they overtook cooperatives. While cooperatives have historically accounted for less than 1 percent of the country’s owner occupied units, condominium’s share rose from 1.3 percent in 1976 to 4.3 per- 271 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cent in 1995 (Table 8.2). Recent multi-family housing is also mostly owned as con dominiums, rather than as cooperatives. Goodman and Goodman (1997) estimate that condominium units maintain a 12 percent price premium over cooperative units. Table 8.2. Share of Cooperative and Condominium Housing among owner occu pied units in United States Condominium Cooperative 1976 1.3% 0.8% 1980 2.1% 0.6% 1983 2.5% 0.7% 1987 3.3% 0.9% 1991 3.9% 0.7% 1995 4.3% 0.6% Source: Based on U.S. Census Bureau American Housing Surveys At the exogenous level, the institutional framework of advanced capitalism hardly favored the growth of collective organizations like housing cooperatives. Al though leading housing activists like Catherine Bauer championed for inclusion of co operatives in federal and other state policies, they were hardly included in such sig nificant pieces of legislation as the 1949 Housing Act (Bailey, 1988), presumably due to private real estate interests. In her study of the historical evolution of housing co operatives in the United States, Leavitt (1995) argues that particular confluence of in stitutional factors after World War II facilitated growth of single-family homes, wherein housing cooperatives were marginalized. These factors include post-war eco nomic growth, federal subsidies that made single-family houses more affordable, and changes in social structure. The primary use of cooperatives in the post-war period has been in the form of Limited Equity Cooperatives (LECs) for converting public housing to ownership housing. Various programs of Housing and Urban Development (HUD) provided for such conversion since the 1950s. LECs have also been promoted as vehicles for af- 272 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fordable housing. According to Miceli, Sazama, and Sirmans (1994, p. 483), nearly 224,000 LEC housing units were promoted since the 1960s through various federal programs, state programs, and nonprofit organizations. In the United States, housing cooperatives are most pronounced in New York. The passage of Limited Dividend Housing Companies Law in 1926 paved the path for expansion of cooperatives’ activities. Trade unions took benefit of the condemnation rights and tax abatements in exchange for limited dividends to house workers. Al though activities of housing cooperatives reduced after the 1929 Depression, their ac tivities picked up after World War II, particularly for exclusive housing among elites (Hansmann, 1996). The 1971 New York State Law authorizing commercial and sav ings banks to make long-term loans to cooperative buyers also assisted growth of co operatives. In the mid 1970s lenders began to increase loan to value ratios on coop erative loans (HUD, 1980). When Housing and Urban Development Department un dertook conversion of public rental housing to condominiums and cooperatives under its various programs, more than 70% of conversions to cooperatives were in New York (HUD, 1980). Broad, nationwide institutional structures supporting housing cooperatives hardly exist in the United States. The National Association of Housing Cooperatives is a non-profit organization promoting the role of housing cooperatives. However, NAHC has had limited outreach. There is no organized structure of cooperatives in each state. Hence, institutional structures for administrative, legal, and procedural support are quite weak. Although housing financial instruments are generally well developed, such instruments are oriented towards individual housing m ortgages; these instruments have limited applicability for cooperatively owned housing (although a few HUD and Fannie Mae programs include cooperatives). Thus, there is little institu- 273 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tional structure for financial support. Failure of public housing conversions to LECs due to lack of such institutional support has been highlighted in several studies (Rohe and Stegman, 1995; Leavitt and Saegert, 1988; Rohe, 1995). 8.5.4. Comparison of Housing Cooperatives in Sweden, Turkey, U.S. and India Table 8.3 summarizes the various features of housing cooperatives in the four countries of Sweden, Turkey, United States, and India. As the table shows, housing cooperatives’ activities have been most pronounced in Turkey, followed by Sweden, India, and United States. The table should be interpreted with some caution with re spect to ability of cooperatives to serve low-income households. Although social housing in Sweden serve low-income households, they do not serve the very poor. Housing cooperatives reach upto middle-income groups in Turkey. If we include the role of LECs in affordable housing, cooperatives in United States could be interpreted to reach down to low-income households. However, as the role of cooperatives itself is low in the housing market, LECs may arguably not be representative of overall co operatives’ activities. Moreover, market based cooperatives hardly reach low-income households in the United States. India’s reach down to low-income households should also be interpreted with caution. As I argued in Chapter 7, although statistics at the national level show such outreach, housing cooperatives in Mumbai and Chennai hardly serve low-income households. How do my policy conclusions relate with the country experiences of Sweden, Turkey, and United States? My first policy conclusion was that cooperative policies should ensure autonomy of housing cooperatives for their growth and development. Housing cooperatives enjoy a great degree of autonomy in Sweden and United States, and less so in Turkey. In India, although cooperatives enjoy autonomy at the national level, such autonomy differs between states. 274 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 8.3. Comparison of Housing Cooperatives in Sweden, Turkey, United States and India Sweden Turkey United States India a. Cooperative Activity i. Share of housing market 15.0% 25.2% <1.0% 10.8% ii. Income groups served Upto lowe income group Upto lower-middle- income group Upto low-income group (inclu. LECs) Upto low-income group b. Endogenous level features i. Type of Cooperatives Tenure Co-ops Building Co-ops Tenure Co-ops Building; Finance; Tenure Co-ops ii. Autonomy of Cooperatives Autonomous Less autonomous Autonomous Autonomous, but varies among states c. Exogenous level features i. Political Economy Democratic socialism Economic liberaliza tion (1980s) Advanced Capitalism Democratic socialism ii. State controls Low Moderate Low Moderate, but varies among states iii. Institutional Framework Supportive Supportive Competitive Supportive, but var ies among states d. Institutional Structure i. National framework Strong Weak Weak Strong ii. Administrative Support Strong Weak Weak Strong iii. Financial support Strong Strong* Weak Strong iv. Support for access to land Strong* Strong* Weak Moderate* e. Transaction Costs Very Low Moderate High Low * Support is provided directly by the state, rather than through independent cooperative institutional structures. N> While the case of Sweden reinforces the first policy conclusion, United States and Turkey differ. Despite autonomy, cooperatives’ activities are limited in United States; also, despite lesser autonomy, cooperatives’ activities have expanded enor mously in Turkey. Although paradoxical, the cases of United States and Turkey do not necessarily invalidate the policy conclusion. For, as I have argued before, there is an inherent tension between the endogenous feature of cooperative autonomy and ex ogenous institutional framework. While cooperative autonomy is indeed desirable, cooperatives also have internal weaknesses. A supportive institutional framework is required to mitigate such weaknesses. In this sense, the competitive institutional framework of United States is hardly supportive for housing cooperatives. The institu tional framework in Turkey is quite supportive of housing cooperatives. My second policy conclusion was that state controls could constrain the growth of housing cooperatives in the long run. While there are few state controls on the functioning of cooperatives in Sweden and United States, cooperatives in Turkey and India face a moderate level of state controls. Again, United States and Turkey seem to be contrary to the policy conclusion. However, there is a difference between state control and no state involvement. In the United States, there is no state control over cooperatives; at the same time, there is little state involvement. Federal policies have hardly included cooperatives. Thus, while state controls have not stymied the growth of housing cooperatives, federal policies have not aided the growth of cooperatives either. In Elster’s terms, there is little “endogenous preference” for cooperatives in the capitalist context of United States. In Turkey, cooperatives face moderate level of controls due to direct state involvem ent in the provision o f housing finance and land; however, there is little state interference in the daily decision making processes of co operatives. 276 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. My third policy conclusion was that cooperative policies are contingent on lo cal institutional history. It is hard to negate such a policy conclusion, but my point was that cooperative autonomy or state control is not given a priori—they emerge within specific political and institutional contexts. Indeed, the historical evolution of cooperative policies in all four countries attests such a conclusion. More broadly, however, the conclusion is significant for transfer of policies from one context to an other. Despite the desirability of certain policies, they may or may not be adaptable to certain contexts, since such policies are hardly implemented in a vacuum. My fourth policy lesson was that the state could play a catalytic role in estab lishing independent cooperative institutional structures. The Indian case stands apart from other countries’ experiences in this regard. While the Indian government was directly involved in establishing cooperative institutional structures, such structures emerged endogenously within the cooperative system in Sweden. Cooperative institu tional structures hardly exist in Turkey and United States. As I mentioned in the third policy conclusion above, cooperative policies are hardly carried out in a vacuum. In Sweden, little state involvement was necessary to setup such structures due to a vi brant cooperative movement; therefore, the structures emerged endogenously. Yet, state involvement in Sweden is required for cooperatives’ access to land. In United States, some feeble support structures have emerged within the cooperative and non profit sectors. In Turkey, although nationwide structures do not exist yet, the state has been providing some concessions for such structures to evolve. My fifth policy conclusion was that the nature of local housing market is criti cal for the specific type of development of housing cooperatives, and housing policies should reconcile with such local market conditions. While this is evident in the cases of Mumbai and Chennai in India, this is also reflected in the local housing markets of 277 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. other countries. For example, although cooperatives have not been active in the United States countrywide, their activities have been quite pronounced in New York, where specific legislations enabled the development of cooperatives. Cooperatives arose in the context of New York’s predominantly multi-family housing in the form of Tenure cooperatives for collective ownership and management. In Sweden, similarly, Tenure Cooperatives proliferated in the context of the country’s multi-family housing units constructed through the MHCs and the Million Program. In Turkey, coopera tives were promoted primarily for housing construction. Lastly, I raised the significance of the institutional structure and the need to lower transaction costs of formation and functioning of housing cooperatives for en hancing their potential for serving low-income households. Cooperative policies have a significant role to play in this: ambiguous procedures and prohibitions could raise the transaction costs of cooperatives. A well developed institutional structure could reduce the transaction costs of cooperatives. I identified three areas in which the insti tutional structure could provide support: internal management, access to finance, and access to land. In Sweden, all three institutional structures have strongly developed. Support for internal management has developed endogenously within the cooperative sector; financial support is also endogenously developed, but is supplemented with govern ment assistance; support for access to land is provided through the municipal govern ments. The transaction costs of cooperatives are very low, and cooperatives have in deed become significant for social housing. Although these cooperatives do not serve the very poor, their reach down the income groups is noteworthy. In Turkey, support for internal management has not fully evolved; support for access to finance and land is directly provided by the government. Thus, the transac- 278 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tion costs of cooperatives are quite low. However, cooperative institutional structure in Turkey is not independent, and it is arguable if the extraordinary growth of housing cooperatives will continue if the government support were withdrawn. Cooperatives serve upto middle-income households, principally because of moderately higher trans action costs. In the United States, support structures have weakly evolved. There is little in stitutional support for internal management, access to finance, and access to land. Consequently, transaction costs of cooperatives are quite high. Market oriented coop eratives hardly reach low-income households; indeed, they are explicitly exclusive en claves for the elite. Although LECs have been promoted for affordable housing, their success depends on the managerial and financial support available through other agen cies (mainly nonprofit organizations). Thus, experiences of all three countries rein force the conclusion with respect to the significance of the institutional structure. 8.6. Conclusion My dissertation answers two questions that have been received little attention in housing research. The first question is, what are the institutional conditions under which housing cooperatives grow? and the second question is, what scope do coopera tives have for fulfilling the housing objectives of low-income households? I have made endogenous organizational level analysis and exogenous institutional level analysis of evolution of housing cooperatives in India. The first level of analysis is useful in identifying the internal strengths and weaknesses of cooperatives. The sec ond level of analysis is useful in identifying elements of the external institutional framework for enhancing the development of cooperatives. In this, I have argued that the institutional structure has a key role to play at both levels. At the endogenous level, the institutional structure could reinforce the internal strengths while mitigating 279 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the internal weaknesses. At the external level, the institutional structure is crucial to reduce the transaction costs involved in the formation and functioning of housing co operatives. I draw six policy lessons based on the institutional analysis of evolution of housing cooperatives in India in general, and the two cities of Mumbai and Chennai in particular. I have tested these policy conclusions against the experiences of housing cooperatives in Sweden, Turkey, and United States. The conclusions seem to broadly attest the other countries’ experiences, although some paradoxes need to be explained. First, I propose that cooperative policies should ensure autonomy of housing coopera tives for their growth and development. Second, I argue that state controls could con strain the growth of housing cooperatives in the long run. Third, I argue that coopera tive policies are contingent on local institutional history. Fourth, I suggest that the state could play a catalytic role in establishing independent cooperative institutional structures. Fifth, I contend that the nature of local housing market is critical for the specific type of development of housing cooperatives, and housing policies should reconcile with such local market conditions. Lastly, I raise the significance of the in stitutional structure and the need to lower transaction costs of formation and function ing of housing cooperatives for enhancing their potential for serving low-income households. 2 8 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bibliography Aiyar, S. 1997. “The Crash of 1997,” India Today, 4 August, pp.60-65. Alchian, Alchian and Harold Demsetz. 1972. “Production Information Costs, and Economic Organization,” American Economic Review, Vol 62, pp. 777-795. Alchian, Armen A. 1950. “Uncertainty, Evolution, and Economic Theory,” The Journal o f Political Economy, Vol. 58, No. 3, p. 211-221. Amitabh. 1997. Urban land markets and land price changes: a study in the Third World context. Aldershot, Hants, England; Brookfield, USA: Ashgate. Axelrod, Robert M. 1984. The evolution o f cooperation. New York: Basic Books. Bailey, Kristin. 1988. The Federal Government and Cooperative Housing Movement, 1917-1955. Ph.D. diss., Camegie-Mellon University. Baken, Robert-Jan and Jan van der Linden. 1992. Land Delivery for Low Income Groups in Third World Cities. Aldershot, Brookfield: Avebury. Baker, Christopher. 1978. “Debt and Depression in Madras, 1929-1936,” in The Im perial Impact: Studies in the Economic History o f Africa and India, ed. Clive Dewey and A. G. Hopkins. London: The University of London, Athlone Press. Baker, Christopher. 1981. “Colonial Rule and the Internal Economy in Twnetieth- century Madras,” Modern Asian Studies, Vol. 15, No. 3, pp. 575-602. Banerjee, Abhijit V., Timothy Besley, and Timothy Guinnane. 1994. “Thy Neighbor’s Keeper: The design of a Credit Cooperative with Theory and Test,” The Quarterly Journal o f Economics, Vol. 119, Issue 2, pp. 491-515. Baneijee-Guha, Swapna. 1996. “Urban Development Process in Bombay: Planning for Whom?” in Bombay: Metaphor for Modern India, ed. Sujata Patel and Al ice Thomer. Bombay: Oxford University Press. Barlow, James and Simon Duncan. 1988. “The Use and Abuse of Housing Tenure,” Housing Studies, Vol. 3, No. 4, pp. 219-231. Barzel, Yoram. 1997. E con om ic A n alysis o f P ro p erty R ights. Cambridge, M A: Cambridge University Press. Bauer, P. T. 1961. Indian Economic Policy and Development. New York: Frederick A. Praeger. 2 8 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bengston, Bo. 1993. Prosperous Ambivalence: Housing Cooperatives in Sweden. Gavle, Sweden: The National Swedish Institute for Building Resaerch. Bengtsson, Bo. 1992. “Not the Middle Way But Both Ways—Cooperative Housing in Sweden,” Scandinavian Housing and Planning Research, Supplement No. 2, Vol. 9, No. 2, pp. 87-104. Berger, Peter and Thomas Luckmann. 1966. The Social Construction o f Reality. New York: Anchor Press. Besley, T. and S. Coate. 1995. “Group Lending, Repayment Incentives, and Social Collateral,” Journal o f Development Economics, Vol. 46, no. 1, pp. 1-18. Bhattacharya, A. 1995. “An Integrated Approach to Urban Development—A CIDCO Experience in New Bombay,” Nagarlok, Vol. 27, No. 3, pp. 1-14. Birchall, Johnston. 1995. “The Hidden History of Cooperative Housing in Britain,” in The Hidden History o f Housing Cooperatives, ed. A. Heskin and J. Leavitt. Davis, CA: University of California. Birchall, Johnston. 1997. The international co-operative movement. Manchester, New York: Manchester University Press. Boleat, M. 1987. “Housing Finance Institutions,” in Shelter, Settlement, and Devel opment, ed. Lloyd Rodwin. A study sponsored by the United Nations. Bos ton: Allen and Unwin. Bombay Metropolitan Region Development Authority (BMRDA). 1995. Regional Plan for Bombay Metropolitan Region, 1996-2011. Bonus, Holger. 1986. “The Cooperative Association as a Business Enterprise: A Study in the Economics of Transactions,” Journal o f Institutional and Theo retical Economics, Vol. 142, pp. 310-339. Brett, E.A. 1993. “Voluntary Agencies as Development Organizations: Theorizing the Problem of Efficiency and Accountability,” Development and Change, Vol. 24, No. 2, pp. 269-303. Buckley, Robert. 1996. Housing finance in developing countries. New York: St. Martin’s Press. Bugra, Ayse. 1998. “The Immoral Econmy of Housing in Turkey,” International Journal o f Urban and Regional Research, Vol. 22, No. 2, pp. 303-317. 2 8 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bumett-Hurst, A. R. 1925. Labour and Housing in Bombay: A Study in the Eco nomic Condition o f the Wage Earning Classes o f Bombay. London: P. S. King and Son, Ltd. Cakin, Sahap. 1991. “Assessment of Co-operative Housing,” Habitat International, Vol. 15, No.1/2, pp. 39-50. Catanach, I. J. 1970. Rural credit in western India, 1875-1930; rural credit and the co-operative movement in the Bombay Presidency. Berkeley: University of California Press. Chandavarkar, Rajnarayan. 1994. The Origins o f Industrial Capitalism: Business strategies and the working classes in Bombay, 1900-1940. Cambridge (Eng land) and New York, NY, USA: Cambridge University Press. Chandavarkar, Rajnarayan. 1998. Imperial Power and popular politics: Class, Re sistance and the state in India, c. 1850-1950. Cambridge (England): Cam bridge University Press. Chouinard V. 1989. “Social production and housing alternatives: cooperative hous ing in post-war Canada,” in The Power o f geography: how territory shapes social life, ed. Jennifer Wolch and Michael Dear. Boston: Unwin Hyman. Chouinard, V. 1990a. “The uneven development of capitalist states: 1. Theoretical proposals and an analysis of postwar changes in Canada’s assisted housing programmes,” Environment and Planning A, Vol. 22, pp. 1291-1308. Chouinard, V. 1990b. “The uneven development of capitalist states: 2. The Struggle for cooperative housing,” Environment and Planning A, Vol. 22, pp. 1441- 1454. City and Industrial Development Corporation of Maharashtra Ltd. (CIDCO). 1995. Socio-economic Survey o f Households in Various Nodes and Villages in New Bombay. Clapham David and Keith Kintrea. 1995. “Housing Cooperatives and Community,” in The Hidden History o f Housing Cooperatives, ed. A. Heskin and J. Leavitt. Davis, CA: University of California. Clapham, David and Keith Kintrea. 2000. “Community-based Housing Organisations and the Local Governance Debate,” Housing Studies, Vol. 15, No. 4, pp. 533- 559. Coase, Ronald. 1937. “The Nature of the Firm,” Economica, Vol. 5, pp. 386-405. 2 8 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Coase, Ronald. 1960. “The Problem of Social Cost,” Journal o f Law and Economics. Vol. 3, pp. 1-44. Cooper and Rodman. 1994. “Accessibility and Quality of Life in Housing Coopera tives,” Environment and Behavior, Vol. 26, No. 1, pp. 49-70. Dale, Peter. 1997. “Land Tenure Issues in Economic Development,” Urban Studies, Vol. 34, No. 10,1621-1634. Dattatri, G. and P. B. Anand. 1991. “Development of Madras: A Historical Review,” Chennai-2011: Policy Imperatives, an Agenda fo r Action (Seminar Theme Pa per), based on a Research Programme conducted by the Times Research Foun dation. Madras: Madras Metropolitan Development Authority. Delhi Development Authority (DDA). 1961. Delhi Master Plan, 1962-1982. New Delhi. Delhi Development Authority (DDA). 1996. Annual Administration Report, 1995-96. New Delhi. Desai, Vandana. 1988. “Dharavi, the Largest Slum in Asia: Development of Low- income Urban Housing in India,” Habitat International, Vol. 12, No. 2, pp. 67- 74. Desai, Vandana. 1996. “Access to power and participation,” Third World Planning Review, Vol. 18, No. 2, pp. 217-242. Deshpande, Sudha. 1996. “Impact of Urbanisation on Women and Family in Bom bay,” in Urban Explosion o f Mumbai, ed. M. D. David. Bombay: Himalaya Publishing House. Deverreux, John and Raymond P. H. Fishe. 1993. “An Economic Analysis of Group Lending Programs in Developing Countries,” The Developing Economies, Vol. XXXI, no. 1, pp. 102-121. DiMaggio, Paul and Walter Powell. 1983. “The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields,” American So ciological Review, Vol. 48, No. 2, pp. 147-160. Doebele, W. A. 1987. “Land Policy,” in Shelter, Settlement, and Development, ed. Lloyd Rodwin. Boston: Allen and Unwin. Doebele, William A. 1983. “Concepts of Urban Land Tenure,” in Urban Land Pol icy: Issues and Opportunities, ed. Harold B. Dunkerley. New York: Oxford University Press. Published for the World Bank. 2 8 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dua, Ajay. 1991. Management o f Bombay's Housing Renewal Programme. Re search Report 27, Indian Human Settlements Programme. Economist Group. 1991. Consumer Preference for Housing in Madras Metropolitan Area. Prepared for the Times Research Foundation, Sponsored by the Project Management Group (Tamil Nadu Urban Development Project), Government of Tamil Nadu. Chennai: MMDA. Elster, Jon. 1989. “From Here to There; or, If Cooperative Ownership is So Desir able, Why Are There So Few Cooperatives?” Social Philosophy and Policy, Vol. 6, No. 2, pp. 93-111. Etzioni, Amitai. 1964. Modern organizations. Englewood Cliffs, N.J., Prentice-Hall Evans, Alan E. 1995. “The Property Market: Ninety Per Cent Efficient?” Urban Studies, Vol. 32, No. 1, pp. 5-29. Evans, Peter. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton, N.J.: Princeton University Press. Evenson, Norma. 1996. “An Architectural Hybrid,” in Bombay: Mosaic o f Modern Culture, ed. Sujata Patel and Alice Thomer. Bombay: Oxford University Press. Fama, E. F. and M. C. Jensen. 1983. “Separation of ownership and control,” Journal o f Law and Economics, Vol. 26, pp. 301-325. Gamble, Andrew and Gavin Kelly. 1996. “The New Politics of Ownership,” New Left Review, No. 220, pp. 62-97. Goodman, Allen C. and John L. Goodman. 1997. “The Co-op Discount,” Journal o f Real Estate Finance and Economics, Vol. 14, No. 1/2, pp. 223-233. Gordon, A.D. 1978. “Businessmen and Politics in a Developing Colonial Economy: Bombay City, 1989-1933,” in The Imperial Edge: Studies in the Economic History o f Africa and India, ed. Clive Dewey and A. G. Hopkins. London: University of London, The Athelone Press. Government of Bombay. 1946. The Greater Bombay Scheme: Report o f the Housing Panel. Bombay. Government of Bombay. 1959. Reports o f the Panels Appointed by the Study Group for Greater Bombay. Bombay. Government of Delhi. 1961. Annual Administration Report, 1960-61. New Delhi. 2 8 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Government of Delhi. 1976. Annual Administration Report, 1975-76. New Delhi. Government of Delhi. 1991. Annual Administration Report, 1990-91. New Delhi. Government of Delhi. 1996. Annual Administration Report, 1995-96. New Delhi. Government of India. 1950. First Five Year Plan. New Delhi: Planning Commis sion. Government of India. 1951. Report o f the Co-operative Planning Committee Ap pointed by the Government o f India on the Recommendation o f the Fourteenth Registrar's Conference. New Delhi. Government of India. 1956. Second Five Year Plan (1956-1961). New Delhi: Plan ning Commission. Government of India. 1964. Report o f the Working Group on Housing Co-operatives. New Delhi. Government of India. 1970. Fourth Five Year Plan(l969-1974). New Delhi: Plan ning Commission. Government of India. 1974. Fifth Five Year Plan (1974-1979). New Delhi: Plan ning Commission. Government of India. 1981. Sixth Five Year Plan(1980-1985). New Delhi: Planning Commission. Government of India. 1985. Seventh Five Year Plan (1985-1990). New Delhi: Plan ning Commission. Government of India. 1988. Report o f the National Commission on Urbanisation, Vol. II. New Delhi. Government of India. 1991. Report o f the Committee on Model Cooperatives Act. New Delhi: Planning Commission. Government of India. 1992. Eighth Five Year Plan (1992-1997). New Delhi: Plan ning Commission. Government of India. 1996. India National Report, Habitat II. New Delhi. Government of India. 1997. Ninth Five Year Plan (1997-2002). New Delhi: Plan ning Commission. 2 8 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Government of India. 1998a. Jawaharlal Nehru: Letters to Chief Ministers, 1947- 1964, Vol. 4 (1954-57). A Project of Jawaharlal Nehru Memorial Fund, Teen Murti House, New Delhi. Government of India. 1998b. National Housing and Habitat Policy. New Delhi. Government of India. 2002. Tenth Five Year Plan, 2002-2007. New Delhi: Planning Commission. Government of Madras. 1928. Report o f the Committee on Co-operation in Madras 1927-28. Madras. Government of Madras. 1948. Report o f the Madras Provincial Housing Committee. Madras. Government of Madras. 1951. Report on the Working o f Co-operative Societies in the State o f Madras. Madras. Government of Maharashtra. 1962. Report o f the Ownership Flats Inquiry Commit tee. Bombay. Government of Maharashtra. 1977. Report o f the Rent Acts Enquiry Committee. Bombay: Government Central Press. Government of Maharashtra. 1981. Report o f the High Power Steering Group for Slums and Dilapidated Houses Appointed by the Government o f Maharashtra. Bombay. Government of Maharashtra. 2001. Co-operative Movement At a Glance in Ma harashtra. Mumbai. Greif, Avner. 1998. “Historical and Comparative Institutional Analysis,” American Economic Review, Papers and Proceedings of the Hundred and Tenth Annual Meeting of the American Economic Association, Vol. 88, No. 2, pp. 80-84. Greif, Avner. 2001. Historical Institutional Analysis. Cambridge, Mass.: Cambridge University Press. Grief, Avner. 1994. “Cultural Beliefs and the Organization of Society: A Historical and Theoretical Reflection on Collectivist and Individualist Societies,” Journal o f Political Economy, Vol. 102, No. 5, pp. 912-950. Guinnane, Timothy. 2000. Cooperatives as Information Machines: German Rural Credit Cooperatives, 1883-1914 (mimeo). 2 8 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Guinnane, Timothy. 2001. A ‘Friend and Advisor': Management, Auditors, and Con fidence in Germany's Credit Cooperatives, 1889-1914. Discussion Paper No. 824, Economic Growth Center (Yale University). Gupta, Devendra B., Sanat Kaul and Rita Pandey. 1993. Housing and India's Urban Poor. New Delhi: Har-Anand Publications. Hall, Peter and Rosemary Taylor. 1996. “Political Science and the Three New Insti- tutionalisms,” Political Studies, Vol. 44, pp. 936-957. Hansmann, Henry. 1991. “Condominium and Cooperative Housing: Transactional Efficiency, Tax Subsidies, and Tenure Choice,” The Journal o f Legal Studies, Vol. XX, pp. 25-55. Hansmann, Henry. 1996. The Ownership o f Enterprise. Cambridge, MA: The Belknap Press of Harvard University Press. Hechter, A. 1990. “The Attainment of Solidarity in Intentional Communities,” Ra tionality and Society, Vol. 2, No. 2, pp. 142-155. Hirchman, Albert O. 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge, MA and London: Harvard University Press. Horvat, Branko. 1982. The Political Economy o f Socialism: A Marxist Social The ory. Armonk, N.Y.: M. E. Sharpe, Inc. Hough, Eleanor. 1932. The co-operative movement in India. London: P.S. King & son, Ltd. [Editions by Oxford University Press: 2n d edition enlarged and re vised for India and Pakistan in 1950; 3rd edition in 1953; 4th edition in 1959; 5th edition in 1966], Housing and Urban Development Department of United States (HUD). 1980. The Conversion o f Rental Housing to Condominiums and Cooperatives: A Na tional Study o f Scope, Causes and Impacts. Washington D.C.: Division of Policy Studies (HUD). Immergut, Ellen M. 1992. “The Roles of the Game: The Logic of Health Policy making in France, Switzerland, and Sweden,” in Structuring Politics: Histori cal Institutionalism in Comparative Analysis, ed. Sven Steinmo, Kathleen Thelen, Frank Longstreth. New York: Cambridge University Press. International Cooperative Alliance (ICA). 1996. ICA News, No. 5/6. Geneva. 2 8 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jain, L. C., and Karen Coelho. 1996. In the Wake o f Freedom: India's Tryst with Co operatives. New Delhi: Concept Publishing. Johnson, Thomas E. 1987: “Upward Filtering of Housing Stock,” Habitat Interna tional, Vol. 11, No. l,pp. 173-190. Keogh, Geoffrey and Eamonn D’Arcy. 1998. “Territorial Competition and Property Market Process: An Exploratory Analysis,” Urban Studies, Vol. 35, No. 8, pp. 1215-1230. Keogh, Geoffrey and Eamonn D’Arcy. 1999a. “Property Market Efficiency: An Insti tutional Economics Perspective,” Urban Studies, Vol. 36, No. 13, pp. 2401- 2414. Keogh, Geoffrey and Eamonn D’Arcy. 1999b. “The Property Market and Urban Competitiveness: A Review,” Urban Studies, Vol. 36, Nos. 5-6, 917-928. Khurana, M. L. 1995. Organisation and Management o f Housing Cooperatives. New Delhi: National Cooperative Housing Federation of India (NCHF). Klein, Ira. 1986. “Urban Development and Death: Bombay City, 1870-1914,” Mod ern Asian Studies, Vol. 20, No. 4, pp. 725-754. Kosambi, Meera. 1986. Bombay in Transition: The Growth and Social Ecology o f a Colonial City, 1880-1980. Stockholm: Almqvist and Wiksell International. Leavitt, J. 1995. “The Interrelated History of Cooperatives and Public Housing from the Thirtees to the Fifties,” in The Hidden History o f Housing Cooperatives, ed. A. Heskin and J. Leavitt. Davis, CA: University of California. Leavitt, J. and S. Saegert. 1988. “The Community Household: Responding to Hous ing Abondonment in New York City,” Journal o f American Planning Associa tion, Vol. 54, Number 4, pp. 489-500. Lewandowski, Susan J. 1975. “Urban Growth and Municipal Development in the Colonial City of Madras, 1860-1900,” Journal o f Asian Studies, Vol. 34, No. 2, pp. 341-360. Lewin, A.C. 1981. Housing Co-operatives in Developing Countries: A Manual for Self-help in Low-Cost Housing Schemes. Intermediate Technology Publica tions and John Wiley & Sons. Loganathan, S. 1995. An Exploratory Study o f the Functioning o f the Cooperative Group Housing Societies in Delhi. M. Phil. Diss., Indian Institute of Public Affairs, New Delhi. 2 8 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lundqvist , J. L., Ingemar Elander, and B. Danermark. 1990. “Housing Policy in Sweden,” International Journal o f Urban and Regional Research, Vol. 14, No. 3, pp. 445-467, Lundqvist, L. J. 1988. “Cooperative Housing,” in Housing Policy and Tenures in Sweden: The Quest for Neutrality, ed. L. J. Lundqvist. Aldershut (England): Avebury. Madras Metropolitan Development Authority (MMDA). 1980. Structure Plan for Madras Metropolitan Area, Volume I and II. Chennai: MMDA. Madras Metropolitan Development Authority (MMDA). 1991. Madras 2011— Policy Imperatives, An Agenda for Action. Theme Paper and Volumes 1 to 5. Based on Research Program Conducted by the Times Research Foundation. Spon sored by the Project Management Group (Tamil Nadu Urban Development Project), Government of Tamil Nadu. Chennai: MMDA Madras Metropolitan Development Authority (MMDA). 1995. Master Plan for Ma dras Metropolitan Area-2011. Chennai: MMDA. Maharashtra Housing and Area Development Authority (MHADA). 1993. Activities and Achievements. Bombay. Maitra, Sipra. 1991. “Housing in Delhi: DDA’s Contoversial Role,” Economic and Political Weekly, Vol. 26, No. 7, pp. 344-346. Marcuse, Peter. 1992. “Why conventional self help projects won’t work,” in Beyond Self-Help Housing, ed. Kosta Mathey. London and New York: Mansell. Markovits, Claude. 1996. “Bombay as a Business Centre in the Colonial Period: A comparison with Calcutta,” in Bombay: Metaphor for Modern India, ed. Sujata Patel and Alice Thomer. Bombay: Oxford University Press. Mathur, B. S. 1977. Co-operation in India. Agra: Sahitya Bhawan (2nd Edition). Mathur, B. S. 1971. Co-operation in India; a critical analysis o f the co-operative movement in India’s planned economy. Agra: Sahitya Bhawan. Miceli, Thomas J., and Gerald W. Sazama. 1998. “Managing Externalities in Multi- Unit Hosuing: Limited Equity Cooperatives as Alternatives to Public Hous ing,” Journal o f Policy Modeling, Vol. 20, No. 5, pp. 649-668. Miceli, Thomas, Gerald W. Sazama, and C.F. Sirmans. 1994. “The Role of Limited Equity Cooperatives in Providing Affordable Housing,” Housing Policy De bate, Vol. 5, Issue 4, pp. 469-490. 2 9 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ministry of Urban Affairs. 1996. A Compendium o f Indian Slums. New Delhi: Gov ernment of India. Misra, B. 1986. “Public Intervention and Urban Land Management: The Experience of Three Metro-Cities of India,” Habitat International, Vol. 10, No. 1/2, pp. 59-77. Moser, Caroline. 1992. “Women and Self-Help Housing Projects: A Conceptual Framework for Analysis and Policy Making,” in Beyond Self-Help Housing, ed. Kosta Mathey. London and New York: Mansell. Mukherji, Panchanandas. 1923. The Co-operative Movement in India. Calcutta and Simla: Thacker, Spink and Co. Mukhija, Vinit. 2001a. “Enabling Slum Redevelopment in Mumbai: Policy Paradox in Practice,” Housing Studies, Vol. 16, No. 6, pp. 791-806. Mukhija, Vinit. 2001b. “Upgrading Housing Settlements in Developing Countries: The Impact of Existing Physical Conditions,” Cities, Vol. 18, No. 4, pp. 213— 222. Mukhija, Vinit. 2002. “An analytical framework for urban upgrading:property rights, property values and physical attributes,” Habitat International, Vol. 26, pp. 553 -570. Naik, D.D. 1985. 25 Years on the Co-op Housing Front: History o f the Maharashtra Co-operative Housing Finance Society Limited 1960-1985. Mumbai: The Ma harashtra Co-operative Housing Finance Society Limited. Naik, K. N. 1953. The Co-operative Movement in the Bombay State. Bombay: The Popular Book Depot. Nath, V. 1995. “Planning for Delhi and National Capital Region: Review of Plan Formulation and Implementation,” Economic and Political Weekly, Vol. XXX, No. 35, pp. 2191-2202. National Buildings Organisation (NBO). 1968. Rent Control and Housing Produc tion in Bombay. Technical Report 25. Study conducted by Tata Institute of Social Sciences on behalf of NBO and UN Regional Housing Centre Ecafe. National Buildings Organisation (NBO). 1996. Handbook o f Housing Statistics Part- I. New Delhi: Ministry of Urban Affairs and Employment, Government of India. 2 9 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. National Cooperative Housing Federation (NCHF). 1991. Study o f Distribution o f Loans among Various Income Groups Provided by Apex Cooperative Housing Finance Societies. New Delhi. National Cooperative Housing Federation (NCHF). 1996. NCHF: What it is? What it does? (pamphlet). New Delhi. National Cooperative Union of India (NCUI). 1997. Convention o f Parliamentari ans'1 Forum for Development o f Cooperative Movement. New Delhi: NCUI. National Housing Bank (NHB). 1996. Annual Report, 1995-96. New Delhi. National Institute of Urban Affairs (NIUA). 1993. Settlement Status in Sites and Ser vices Schemes at Madras. Research Study Series Number 53. New Delhi. Neild, Susan. 1979. “Colonial Urbanism: The Development of Madras City in the Eighteenth and Nineteenth Centuries,” Modern Asian Studies, Vol. 13, No. 2, pp. 217-246. Nicholson, F. 1895. Report Regarding the Possibility o f Introducing Land and Agri cultural Banks into the Madras Presidency, 2 volumes. Nijman, Jan. 2000. “Mumbai’s Real Estate Market in 1990s: De-regulation, Global Money and Casino Capitalism,” Economic and Political Weekly, Vol. 35, No. 7, pp. 575-582. North, Douglass. 1981. Structure and Change in Economic History. New York: W.W. Norton and Company. North, Douglass. 1990. Institutions, Institutional Change and Economic Perform ance. Cambridge, Mass.: Cambridge University Press. O’Hare, Greg, Dina Abbott, and Michael Barke. 1998. “A review of slum housing policies in Mumbai,” Cities, Vol. 15, No. 4, pp. 269-283. Olson, Mancur. 1965. The logic o f collective action; public goods and the theory o f groups. New York: Schocken Books. Orr, J. P. 1915. The Need for Co-operation between Neighbours in the Development o f Building Estates. Lectures delivered before the Bombay Cooperative Hous ing Association. Ostrom, E. 1986. “An Agenda for the Study of Institutions,” Public Choice, Vol. 48, p p .3-25. 2 9 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pamuk, Ayse. 1992. “Comparative analysis of housing delivery systems for low- income households : policy options for Turkey,” Institute o f Urban and Re gional Development Working Paper No. 589. Berkeley, Calif. : University of California Pamuk, Ayse. 1996. “Convergence Trends in Formal and Informal Housing Markets: The Case of Turkey,” Journal o f Planning Education and Research, Vol. 16, pp. 103-113. Panwalkar, Pratima. 1996. “Upgradation of Slums: A World Bank Programme,” in Bombay: Metaphor for Modern India, ed. Sujata Patel and Alice Thomer. Bombay: Oxford University Press. People’s Dialogue Report. 1996. From Ends to Means: Savings and credit as a means to community mobilisation, people's power and housing finance. ACHR/TAP Workshop held in Mumbai, 25 September to 1 October. Pugh, Cedric. 1994a. “Housing Policy development in developing countries: The World Bank and internationalization, 1972-93,” Cities, vol. 11, no. 3, pp. 159- 180. Pugh, Cedric. 1994b. “The idea of enablement in housing sector development,” Cit ies, vol. 11, no. 6, pp. 357-371. Ramasubban, Radhika and Nigel Crook. 1996. “Spatial Patterns of Health and Mor tality,” in Bombay: Metaphor for Modern India, ed. Sujata Patel and Alice Thomer. Bombay: Oxford University Press. Reed, Sir Stanley and S. T. Sheppard (eds.). 1928. Indian Yearbook. Bombay: Ben nett, Coleman, and Co. Ltd. Renaud, B. 1987. “Financing Shelter,” in Shelter, Settlement, and Development, ed. Lloyd Rodwin. A study sponsored by the United Nations. Boston: Allen and Unwin. Reserve Bank of India. 1950. Cooperative Housing. Bombay. Reserve Bank of India. 1978. Finance fo r Housing Schemes: Report o f the Working Group on Role o f Banking System. Bombay. Riker, William H. 1980. “Implications from the Disequilibrium of Majority Rule for the Study of Institutions,” American Political Science Review, Vol. 74, No. 2, pp. 432-446. 2 9 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Risbud, N. n.d. Government Policies and Illegal Land Supply by Housing Coopera tives—Jaipur. Indian Human Settlements Programme (IHSP) Research. Robben, Pieter J. M. and Pieter A. van Stuijvenberg. 1986. “India’s Urban Housing Crisis: Why the World Bank’s Sites and Services Schemes are not Reaching the Poor in Madras,” Third World Planning Review, Vol. 8, No. 4, pp. 335- 351. Rogaly, Ben. 1998. “Combating financial exclusion through co-operatives: Is there a role for external assistance?” Journal o f International Development, Vol. 10, pp. 823-836. Rohe, W. M. and M. A. Stegman. 1995. “Converting Public Housing to Coopera tives: Lessons from Nashville,” in The Hidden History o f Housing Coopera tives, ed. A. Heskin and J. Leavitt. Davis, CA: University of California. Rohe, William M. 1995. “Converting Public Housing to Cooperatives: The Experi ence of Three Developments,” Housing Policy Debate, Vol. 6, Issue 2, pp. 439-479. Saegert, Susan and Gary Winkel. 1996. “Paths to Community Empowerment: Orga nizing at Home,” American Journal o f Community Psychology, Vol. 24, No. 4, pp. 517-550. Salamon, Lester. 1987. “Partners in Public Service: The Scope and Theory of Gov ernment-Nonprofit Relations,” in The Nonprofit Sector: A Research Hand book, ed. W. Powell. New Haven: Yale University Press. Sanyal, Bishwapriya. 1998. “Beyond the Theory of Comparative Advantage: Politi cal Imperatives of the Government-Nonprofit Relationship,” in Shelter and So ciety: Theory, Research, and Policy for Nonprofit Housing, ed. C. Theodore Koebel. Albany: State University of New York Press. Sazama, Gerald. 1996. A Brief History o f Affordable Housing Cooperatives in the United States. Department of Economics Working Paper No. 1996-09. Storrs, CT: University of Connecticut. . Sazama, Gerald. 2000. “Lessons from the History of Affordable Housing in the United States: A Case Study in American Affordable Housing Policy,” Ameri can Journal o f Economics and Sociology, Vol. 59, No. 4, pp. 573-608. Schenk, Hans, Charlotte Ten Holder, Ralph Mulders. 1989. “Open Developed Plots and Sites and Services,” Indian Geographical Journal, Vol. 64, No. 2, pp. 107- 126. 2 9 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. School of Architecture and Planning. 1993. Shelter Strategy for Madras, Trichy and Erode. Madras: Sponsored by Project Management Group (Tamil Nadu Urban Development Project), Government of Tamil Nadu. Schotter, Andrew. 1981. The Economic Theory o f Social Institutions. Cambridge, MA: Cambridge University Press. Scott, W. Richard. 1998 (Fourth Edition). Organizations: Rational, Natural, and Open Systems. Upper Saddle River, NJ: Prentice Hall. Shah, Tushaar. 1996. Catalysing Co-operation: Design o f Self-governing Organisa tions. New Delhi: Sage Publications. Sharma, R. N., S. Siva Raju, and I. U. B. Reddy. 1986. A Study o f Willingness and Affordability o f Occupants o f S.I.H. Scheme and Cessed Buildings to Avail Loan Facility fo r Repair or Reconstruction o f Their Tenements. Report pre pared for the Maharashtra Housing & Area Development Authority. Sharma, S. C. 1995. “Emergence and Evolution of HUDCO,” in The Corporate Story o f HUDCO, 1970-1995. Published by HUDCO to commemorate the Silver Jubilee. New Delhi. Shepsle, Kenneth A. 1989. “Studying Institutions: Some Lessons from the Rational Choice Approach,” Journal o f Theoretical Politics, Vol. 1, No. 2, pp. 131-148. Shirras, Findlay G. 1923. Report on an Enquiry into Working Class Budgets in Bom bay. Bombay: Government of Bombay (Labour Office). Skocpol, Theda and Margaret Somers. 1994. “The Uses of Comparative History in Macrosocial Inquiry,” in Social Revolutions in the Modern World, ed. Theda Skocpol. New York: Cambridge University Press. Skocpol, Theda. 1985. “Bringing the State Back In: Strategies of Analysis in Current Research,” in Bringing the State Back In, ed. Peter Evans, D Rueschenmeyer, and Theda Skocpol. Cambridge: Cambridge University Press. Skocpol, Theda. 1995. “Why I am a Historical-Institutionalist,” Polity, Vol. 28, pp. 103-106. Society for Development Studies. 1996. Housing and Urban Indicators: Manage ment Tools for Human Settlements, India Case Study. New Delhi. Socio Economic Research Foundation. 1990. Study on Real Estate Market—A Case Study o f Delhi. New Delhi. 2 9 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Staber, Udo H. 1992. “Organizational Interdependence and Organizational Mortality in the Cooperative Sector: A Community Ecology Perspective,” Human Rela tions, Vol. 45, No. 11, pp. 1191-1212. Stiglitz, Joseph E. 1990. “Peer Monitoring in Credit Markets,” World Bank Economic Review, Vol. IV, pp. 351-366. Strassmann, Paul W. 1982. The transformation o f urban housing: the experience o f upgrading in Cartagena. Baltimore: Published for the World Bank by Johns Hopkins University Press. Struyk, Raymond J., and R. Marisol Ravicz. 1992. Housing finance in LDCs : In dia's National Housing Bank as a model? Washington, D.C.: Urban Institute Press. Sykes (Colonel). 1852. “On the Census of the Islands of Bombay and Colaba, Taken on the 1st of May, 1849, by Captain Baynes, Superintendent of Police,” Jour nal o f the Statistical Society o f London, Vol. 15, No. 4, pp. 327-338. Talmaki, S. S. 1920. “Co-operation and the Housing Problem,” in R. B. Ewbank (ed.), Indian Co-operative Studies. London, Bombay, Madras: Oxford Uni versity Press. Talmaki, S. S. 1931. Co-operation in India and Abroad. Mangalore (India): Basel Mission Press. Task forces on Housing and Urban Development. 1983. Shelter fo r the Urban Poor and Slum Improvement. New Delhi: Planning Commission, Government of India. Thelen, Kathleen and Sven Steinmo. 1992. “Historical Institutionalism in Compara tive Politics,” in Structuring Politics: Historical Institutionalism in Compara tive Analysis, ed. Sven Steinmo, Kathleen Thelen, and Frank Longstreth. New York: Cambridge University Press. Thirunarayanan, R. 1993. Cooperative Housing in Tamil Nadu: Consultancy Study, Volume I. Sponsored by the Project Management Group (Tamil Nadu Urban Development Project), Government of Tamil Nadu. Chennai: Natesan Insti tute of Cooperative Management. Tiwari, Piyush, Kirit Parikh, and Jyoti Parikh. 1999. “Effective Housing Demand in Mumbai (Bombay) Metropolitan Region,” Urban Studies, Vol. 36, No. 10, pp. 1783-1809. 2 9 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Turel, Ali. 2000. Limits o f Home-ownership in Turkey. Paper presented at the Euro pean Network for Housing Research (ENHR) Conference, 26-30 June 2000. Gavle, Sweden. Turnbull, Shann. 1983. “Cooperative Land Banking,” in Land for housing the poor, ed. Shlomo Angel. Singapore: Select Books. Turner, Alan. 1980. Structure Plan for Madras Metropolitan Area, Volume 1 and 2. Madras: Madras Metropolitan Development Authority. Turner, Bengt. 1995. “Housing Policy in Sweden,” in P. Balchin (ed.), Housing Pol icy in Europe. London: Routledge. Turner, Bengt. 1997. “Municipal Housing Companies in Sweden: On or Off the Market,” Housing Studies, Vol. 12, No. 4, pp. 477-487. Turner, John and R. Fichter. 1972. Freedom to Build: Dweller Control o f the Hous ing Process. New York: The Macmillan Company. United Nations Centre for Human Settlements (Habitat) (UNCHS). 1989. Coopera tive Housing: Experiences o f Mutual Self-Help. UNCHS: Nairboi. United Nations Centre for Human Settlements (Habitat) (UNCHS). 1990. Finance fo r Shelter and Services. Nairobi: UNCHS. United Nations Centre for Human Settlements (UNCHS). 1996. An Urbanizing World: Global Report on Human Settlements. Nairobi: Published by Oxford University Press for UNCHS. Upadhyay, Shashi Bhushan. 1990. “Cotton Mill Workers in Bombay, 1875-1918: Conditions of Work and Life,” Economic and Political Weekly, 28 July, Vol. 25, pp. PE-87 to PE-99. Uphoff, Norman. 1986. Local Institutional Development. West Hartford, CT: Ku- marian Press. Vakil, Anna C. 1997. “Confronting the Classification Problem: Toward a Taxonomy ofNGOs,” World Development, Vol. 25, No. 12, pp. 2057-2070. Vakil, Anna C. 1996. “Understanding Housing CBOs: Compartive Case Studies from Zimbabwe,” Third World Planning Review, Vol. 28, No. 3, pp. 325-347. Vandewalle, Don, Linn Van Dyne, and Tatiana Kostova. 1995. “Psychological Own ership: An Empirical Examination of Its consequences,” Group and Organiza tion Management, Vol 20, No. 2, pp. 210-226. 2 9 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Varian, Hal. 1990). “Monitoring Agents with Other Agents,” Journal o f Institutional and Theoretical Economics, Vol. 146, pp. 153-174. Wenner, Mark D. 1995. “Group Credit: A Means to Improve Information Transfer and Loan Repayment Performance,” The Journal o f Development Studies, Vol. 32, No. 2, pp. 263-281. Williams, Joseph T. 1993. “Agency and Ownership of Housing,” Journal o f Real Es tate Finance and Economics, Vol. 7, pp. 83-97. Williamson, Oliver. 1985. The Economic Institutions o f Capitalism. New York: The Free Press; and London: Collier Macmillan Publishers. Williamson, Oliver. 1996. The mechanisms o f governance. New York: Oxford Uni versity Press. Wolcott, Susan. 1991. “British Myopia and the Collapse of Indian Textile Demand,” Journal o f Economic History, Vol. 51, No. 2, pp. 367-384. Wolcott, Susan. 1994. “The Perils of Lifetime Employment Systems: Productivity Advance in the Indian and Japanese Textile Industries, 1920-1938,” Journal o f Economic History, Vol. 51, No. 2, pp. 367-384. Wolcott, Susan. 1999. “Why nations fail: Managerial decisions and performance in Indian cotton textiles, 1890-1938,” Journal o f Economic History, Vol. 59, No. 2, pp. 397-423. Wolff, Henry W. 1920. “Introduction” in Indian Cooperative Studies, ed. R. B. Ew- bank. University of Bombay Economic Series No. 2. London, Bombay, and Madras: Humphrey Milford, Oxford University Press. World Bank. 1985. Bombay Urban Development Project, Staff Appraisal Report No. 4794-IN. Washington, D.C. World Bank: 1993. Housing: Enabling Markets to Work. World Bank Policy Paper. Washington, D.C.: The World Bank. Zachariah, K.C. 1966. “Bombay Migration Study: A Pilot Analysis of Migration to an Asian Metropolis,” Demography, Vol. 3, No. 2, pp. 378-392. Zachariah, K.C. 1968. Migrants in Greater Bombay. Bombay: Asia Publishing House. Zucker, Lynne. 1987. “Institutional Theories of Organization,” American Review o f Sociology, Vol. 13, pp. 443-464. 2 9 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1: Housing Cooperatives in New Delhi Unlike Mumbai and Chennai, housing cooperatives grew in New Delhi primar ily after India’s Independence. Housing cooperatives in New Delhi were modeled af ter those in Mumbai. Initially, cooperative laws in New Delhi were the same as those in Mumbai since the Bombay Cooperative Societies Act, 1925 was extended to New Delhi in 1949. An indigenous institutional framework governing cooperatives was established almost two decades later in 1972 with the Delhi Cooperative Societies Act. Thus, the institutional history of housing cooperatives in New Delhi is much shorter than that of Mumbai and Chennai. Building Cooperatives for land development (in the form of Cooperative House Building Societies) emerged in the initial years following independence. How ever, in the longer run, Tenure cooperatives (in the form of Cooperative Group Hous ing Societies) have predominated. These cooperatives cater mostly to high and mid- dle-income households, and hardly serve the low-income households. More recently, in the 1990s, entrepreneurial bureaucrats introduced New Multipurpose Cooperative Societies in slums as collective self-help mechanisms; however, growth of their activi ties has been limited. In Delhi, the institutional structures for administrative support are still young. The Registrar’s office plays a controlling role and few structures have evolved for providing administrative, legal, and procedural support. The structures for providing access to finance and land are, however, rather well defined. Financial support is pro vided through the Delhi Cooperative Housing Finance Corporation. The Delhi Devel opment Authority (DDA) provides land on a preferential (albeit erratic) basis. DDA’s role has been most significant in Delhi, since cooperatives’ growth has centrally de pended on the extent to which DDA allocated land to them. 299 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a. Endogenous Level Analysis: Growth of Housing Cooperatives Cooperative House Building Societies were initiated in 1947 in New Delhi to develop plots and allot them to members for individual house construction.1 However, their activities were limited in the first decade due to lack of effective leadership and access to finance. Land prices were also unaffordable for members (especially for those who belonged to low and middle income groups). The number of Cooperative House Building Societies grew from 3 in 1946-47 to 128 in 1955-56. The number of these cooperatives more than doubled over the next five years, so that there were 303 such cooperatives by 1960-61. The spurt in growth was in part due to the anticipation of allocation of land acquired by the local government through large-scale land acqui sition policies since the 1950s. Under the Land Acquisition Act, 1894, the local government planned to ac quire about 8,000 acres of land in 1961; of this, about 850 acres were earmarked for Cooperative House Building Societies and Industrial Cooperatives (Government of Delhi, 1961). However, in the same year, a ban was imposed on the registration of new cooperatives since the government could not allocate land to more cooperatives. Cooperatives’ activities received a boost in 1970, when registration of new housing cooperatives in the form of Cooperative Group Housing Societies was intro duced. DDA anticipated that it would allocate land to these cooperatives after fulfill ing the needs of the existing Cooperative House Building Societies, which were also given permission to opt for “group housing.” DDA’s scheme envisaged group hous ing for 50 to 120 households in apartment complexes to achieve higher residential density. Construction would be taken by cooperatives themselves, which provided an alternative to construction by DDA. The ownership of land and management of com 1 These cooperatives were also called Plot Holders Societies. 300 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mon spaces and services also rested with the cooperatives. These cooperatives were modeled after those in Mumbai. Table A-l. Growth of Housing Cooperatives in New Delhi, 1946-47 to 1995-96 Year Cooperative House Building Societies Cooperative Group Housing Societies New Multipurpose Cooperatives 1946-47 3a 1950-51 41a 1955-56 c J 00 < N 1960-61 303a 1965-66 303a 1970-71 306a 94b 1975-76 236° 170b 1980-81 595b 1985-86 230d 2,004d 1990-91 225d 2,012d 4d 1995-96 154e 1,944® 5 3 ® Sources', (a) Calculated from Loganathan (1995, p. 32); (b) Calculated from Logana- than (1995, p. 33); (c) Based on Loganathan (1995, p. 33) and Government of Delhi (1976); (d) Government of Delhi (1991); (e) Government of Delhi (1996) With the opening of the 1970 registration, 143 Cooperative Group Housing Societies were registered, before a general ban on registration was again imposed in 1973. The ban on registration was lifted and imposed twice more since then: the ban was lifted in 1979, and re-imposed in 1980; then, the ban was lifted again in 1983 and finally re-imposed in 1984 (Loganathan, 1995, p.32). Hence, the growth of Coopera tive Group Housing Societies has not been steady—their numbers grew in spurts as and when registration was opened. Table A-l shows the growth of housing coopera tives in New Delhi between 1946-47 and 1995-96. By 1995-96, there were 1,944 Co operative Group Housing Societies, with a total membership of 222,000; however, land had been allocated to only 1400 cooperatives by this year (Government of Delhi, 1996, p.6-7). Thus, although cooperatives received preferential treatment, allocation of land has been erratic. According to survey conducted by Socio Economic Research 301 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Foundation (1990), 44.4 percent of Cooperative Group Housing Societies obtained land within 5 years of application; another about 44.4 percent obtained land between 6 to 20 years. Further, due to the control over registration, and preference given to Co operative Group Housing Societies, the activities of Cooperative House Building So cieties waned over time: by 1995-96, the number of Cooperative House Building So cieties reduced to about half of that in 1960-61 (Table A-l). Cooperative Group Housing Societies are principally oriented towards high and middle income groups (especially salaried employees in the formal sector). The main reason for why cooperatives have not reached low-income households is DDA’s land pricing policy. Although DDA allocates land to cooperatives on a preferential basis at pre-determined prices (unlike to individuals where land is auctioned to the highest bidder), even such pre-set land prices are beyond the reach of low-income households. DDA’s policy has been traditionally to allocate land to such cooperatives that exhibit financial viability to construct and maintain group housing. There have been hardly any cross-subsidy mechanisms through which cooperatives of low-income households could obtain land for construction. Indeed, there has been much criticism of DDA as a public monopoly acting as a private land developer, so that much of its housing has been directed towards higher income groups (Maitra, 1991; Misra, 1986). A few feeble efforts were undertaken during the 1990s to use cooperatives as means of collective action and group credit. Experimental New Multipurpose Coop erative Societies were initiated under “Punervas” scheme in slums for all round devel opment of slum-dwellers (e.g. organizing them into a group for savings) (Gupta, Kaul, and Pandey, 1993, pp. 186-189). However, these cooperatives were initiated by entre preneurial government bureaucrats and have not been grass-roots efforts. Hence these cooperatives developed to the extent the bureaucrats were willing to be involved. As 302 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of 1995-96, about 53 New Multipurpose Cooperative Societies were formed; however, these cooperatives have not achieved any significant goals. b. Exogenous Level Analysis: Housing Market and Institutional Framework The state played a central role in the evolution of housing cooperatives in New Delhi. As the previous section shows, the growth in activities of housing cooperatives fundamentally depended on land allocation by DDA. DDA was formed in 19552 with an explicit objective to “promote and secure the development of Delhi according to plan and for that purpose the Authority shall have the power to acquire, hold, manage and dispose off land and other property, to carry out building, engineering, mining and other operations...” (DDA, 1996). In pursuing its objectives, DDA formulated two Master Plans—the first in 1961 and the second in 1990. The need for planning should be viewed in light of Delhi’s rapid population growth in the aftermath of World War II, resulting in higher housing costs, haphazard urban growth, and increase in slum population.3 The first plan avowed, “without large scale acquisition of land, long before need, planning cannot be fully effectuated” (DDA, 1961). Indeed, New Delhi was a pioneer in implementing large scale land acquisition policies, which were later re flected in the first national level urban land policy proposed by the High Level Urban Policy Committee in 1965 (Government of India, 1988, p. 213). The rationale behind land acquisition in New Delhi was manifold. First, it would bring undeveloped land in the periphery of city under government control at low prices. Second, public control of land would lead to a more orderly planning and development. Third, speculation 2 DDA was known as Delhi Developm ent (Provisional) Authority until 1957. 3 Delhi Municipal Corporation’s population surged 107 percent between 1941 and 1951 due to post-war rural-urban migration and influx o f refugees from Pakistan after Independence. The population continued to increase steadily since then to reach over 7 m illion in 1991. Slums accommodated about 33 percent o f the city’s population in 1991. 303 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. would be avoided when agricultural land is converted to urban use. Fourth, the poor would not be left to vagaries of the private market, and can be better served by the public sector. Under the scheme, serviced land was auctioned to the highest bidder. Housing cooperatives and a few other groups were exempt from such auctioneering, and were allocated land at pre-determined rates. Indeed, the large-scale land acquisition policies followed since then vested DDA with large tracts of land. Of the 63,201 acres envisaged for acquisition under the first master plan, DDA acquired 45,000 acres (71 percent) by the end of the plan pe riod in 1982 and 56,985 acres (90 percent) upto 1989. DDA became directly involved in the construction of housing and in land development since 1968 and became a sig nificant player in the local housing market. DDA provided as much as 64 percent of permanent housing in Delhi between 1970 and 1990 (Nath, 1995, p. 2196). As I highlighted in Section (a), the growth of housing cooperatives’ activities has fundamentally depended on land allocation by DDA. Indeed, few housing coop eratives have been formed outside the ambit of land allocation through DDA. Coop eratives could be hardly considered as autonomous and spontaneous organizations in Delhi due to the periodic bans and cooperative’s overt dependence on DDA. The overt control over formation and functioning of cooperatives restricted the scope of cooperatives’ activities. c. Cooperative Institutional Structure Although cooperatives in New Delhi were modeled after those in Mumbai, the institutional structures supporting the development of cooperatives differ. In New Delhi, the Registrar has immense power in terms of intervening in the decision making process of cooperatives. The immense power of Registrar has also resulted in graft, and it typically takes a long time to even register cooperatives (sometimes more than a 304 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. year). Indeed, although non-governmental organizations have strived to mobilize slum-dwellers and low-income households into groups for collective action and thrift savings, some of them do not consider cooperatives as primary vehicles for group formation due to bureaucratic interference.4 Unlike Mumbai, additional structures supporting the administrative and proce dural issues relating to cooperatives are weak. There are two plausible explanations for this. First, the history of cooperatives in New Delhi is much shorter, and the pro gress of cooperative’s activities has been erratic. Second, the number of housing co operatives is too few to give rise to such additional structures. The institutional structure for providing finance is better defined in New Delhi. The Apex Cooperative (Delhi Cooperative Housing Finance Corporation, DCHFC) has been instrumental in providing loans for construction to more than half the Coop erative Group Housing Societies. It lends money to primary cooperatives for housing construction; primary cooperatives, in turn, utilize the money on behalf of their mem bers (for construction, services, and so on). However, instead of the collective being responsible for repayments, the liability of repayment resides with individual mem bers. DCHFC’s loans hardly reach the low-income households. The institutional structure for providing land is better defined in New Delhi, since the DDA has an ex plicit (albeit erratic) policy of allocating land to Cooperative Group Housing Societies. Unlike DDA, the metropolitan development authorities in Mumbai and Chennai do not have an explicit strategy of allocating land to housing cooperatives. 4 Personal conversation held on 2 September, 1997 with Saurabh Pillay (Manager, Co-operative Programme) o f SHARAN, a non-governmental organization based in N ew Delhi for serving the urban poor. 3 0 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. d. Conclusion The institutional history of housing cooperatives in New Delhi is much shorter than that of Mumbai and Chennai since cooperatives were initiated only during the post-Independence period. The state played a central role in the development of coop eratives. The DDA plays a significant role in the housing market by virtue of its direct involvement in housing construction and land development. Cooperatives’ activities grew in Delhi to the extent DDA allocated land to them. New cooperatives could be formed only when the bans on registration were lifted. Cooperatives emerged princi pally as means for group housing, which was the main rationale for DDA to allocate land. In this, cooperatives have hardly catered to low-income households. Although new multipurpose cooperatives were started in slums for collective action and group credit, their activities have been limited. 3 0 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Faith -based organizations, international development agencies, and environmental management

PDF

Comparative study of organizational commitment in the public and private sectors: The case of transportation agencies in Thailand

PDF

Banking on the commons: An institutional analysis of groundwater banking programs in California's Central Valley

PDF

The social construction of brownfields

PDF

Business political influence on agency regulatory decision making: The efficacy of corporate political activity in Taiwan's CATV license awarding

PDF

Infrastructure investment and regional productivity of the manufacturing industries in Mexico

PDF

A configuration study of multiagency partnerships as practiced in Taipei City government

PDF

A comparative study of administrative corruption in ancient and modern China

PDF

A statistical analysis of the formation and location factors of high -tech centers in the United States, 1950--1997: An evaluation using quasi -experimental control group methods

PDF

Evaluation of policy alternatives to alleviate transportation-related air pollution

PDF

Governing public goods: how representation and political power in local and regional institutions shape inequalities

PDF

Essential features of federalism: A study of federal evolution in India's Capital Territory

PDF

Differences in the development of citizenship: Residents' motivations and capacities for participating in neighborhood associations in Los Angeles

PDF

Factors associated with length of stay on welfare: Toward a theory of context for addressing long -term dependency

PDF

Beyond regulation: special improvement districts, design review, and place-making in New Jersey

PDF

Decentralizing of public finance: Centralizing forces in developing countries

PDF

The social functions of private neighborhood associations: the case of homeowner associations in urban China

PDF

Strategy and ideology in nonprofit advocacy organizations

PDF

From “squatters” to citizens? Slum dwellers, developers, land sharing and power in Phnom Penh, Cambodia

PDF

Institutional analysis of factors critical to facilitating project management and implementation of Metro ExpressLanes

Asset Metadata

Creator Ganapati, Sukumar (author)

Core Title Institutional analysis of evolution of housing cooperatives in India

School School of Policy, Planning and Development

Degree Doctor of Philosophy

Degree Program Planning

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,sociology, social structure and development,Urban and Regional Planning

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Richardson, Harry W. (committee chair), Tang, Shui Yan (committee member), Verma, Niraj (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-482112

Unique identifier UC11340003

Identifier 3133271.pdf (filename),usctheses-c16-482112 (legacy record id)

Legacy Identifier 3133271.pdf

Dmrecord 482112

Document Type Dissertation

Rights Ganapati, Sukumar

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA