Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Developing countries in the digital era: state and business interactions for industrial upgrading
(USC Thesis Other)
Developing countries in the digital era: state and business interactions for industrial upgrading
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Developing Countries in the Digital Era:
State and Business Interactions for Industrial Upgrading
By Mariana Rangel Padilla
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment
Of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
POLITICAL SCIENCE AND INTERNATIONAL RELATIONS
December 2017
i
Abstract
This dissertation studies the adoption and design of industrial upgrading policies in
democratic developing countries. In doing so it highlights business-government interactions and
their outcomes. The research on the adoption of industrial upgrading policies tackles the following
questions: Who are the actors, interests and coalitions that drive the adoption of industrial
upgrading policies? Why do some countries (and provinces) adopt industrial upgrading policies
and others do not? Concerning the design, I address the questions: What accounts for differences
in the characteristics of industrial upgrading programs? How might variation in the quality of
government and business cohesion shape the measures adopted? Overall, we need a more
comprehensive understanding of the process of industrial upgrading policy underscoring state-
business interactions. This is the main contribution of my doctoral dissertation.
Developmental state literature revealed that meritocratic bureaucrats in East Asia were
capable of stimulating the transition into higher value-added activities. Yet these works largely
focused on economic transformations prior to the digital era and mostly under authoritarian
regimes. Thus, they are insufficient to explain the efforts to increase knowledge intensive activities
in a global and more open scenario. Some recent studies have focused in a few small, yet strategic
countries like Israel, Taiwan and Ireland. Meanwhile, business politics analyses, have highlighted
how firms actually shape the outcomes of economic policy-making processes such as structural
adjustment, trade liberalization and taxation. But they have paid scant attention to industrial policy
in developing countries.
Seeking to fill these gaps in the literature, this dissertation advances an analytical
framework that builds on the developmental state and on business politics theories. The framework
posits that bureaucratic quality and business cohesion generate diverse industrial upgrading
policymaking processes and outcomes. To test and refine the propositions, I conduct a set of
empirical studies. The analysis relies on a mixed-methods research strategy that conveys the
strengths of quantitative and qualitative techniques. More specifically, this dissertation combines:
1) a cross-national event history analysis of software policy adoption in Latin America (1995-
2012); 2) an in depth case study of Mexico, to process trace and find the causal mechanisms
underlying the adoption and design of the Program to Support the Software Sector (1990-2005);
and 3) two Mexican subnational cases, comparing Nuevo León and Puebla during two periods
ii
(2000-2007/ 2008-2015) to further refine the theory and test its applicability to the subnational
level.
The main conclusion of this dissertation is that not all states and all business sectors are
created equal: some may actually foster efforts to transition into knowledge intensive activities.
This challenges, once again but now in the digital era, the well-entrenched view against both, state
intervention in the economy and close business-government collaboration in developing countries,
which to many scholars and policymakers only leads to rent-seeking. Under certain conditions,
bureaucracies and business leaders, and not just free markets, may be the drivers of industrial
upgrading.
iii
Table of Contents
List of Tables iv
List of Figures v
Acknowledgements vi
Chapter 1. Introduction 1
Chapter 2. A Framework for the Study of Industrial Upgrading: Unpacking
Business-Government Interactions and Policy Outcomes
15
Chapter 3. The Adoption of Policies Towards the Software Sector: A Cross-
Country Quantitative Test, Latin America (1995-2012)
39
Appendix 1. Industrial Upgrading Policies for the Software Sector in Latin America 61
Appendix 2. List of Variables and Sources 62
Appendix 3. Robustness Checks Using Government Effectiveness 63
Chapter 4. State-Business Interactions in Industrial Upgrading Policy Processes:
The Case of Mexico’s Program to Support the Software Sector
65
Appendix 1. List of Interviewees 91
Appendix 2. AMITI's Membership Evolution. 1985-2016 92
Appendix 3. CANIETI's Membership Evolution. 1994-2016 93
Appendix 4. CANIETI Membership Distribution by Sector. 2013 94
Chapter 5. Industrial Upgrading Policies at the Subnational Level: The Cases of
Puebla and Nuevo León
95
Appendix 1. Prosoft’s Funds in Puebla and Nuevo León (2004-2015). 123
Appendix 2. Evolution of Prosoft’s Contributions by Actor. 124
Appendix 3. The Software Industry in Mexican States. Key Indicators. 2013 125
Appendix 4. Software Industry in Mexican States Ranking. 126
Appendix 5. List of Interviews Monterrey. July 2012 127
Appendix 6. List of Interviews Puebla. May 2014 128
Chapter 6. Conclusions and Research Agenda
129
References 135
iv
List of Tables
Chapter 2. A Framework for the Study of Industrial Upgrading: Unpacking
Business-Government Interactions and Policy Outcomes
15
Table 1. Varieties of Industrial Policies 32
Table 2. Summary of Models of Business Government Interactions 37
Chapter 3. The Adoption of Policies Towards the Software Sector: A Cross-
Country Quantitative Test, Latin America (1995-2012)
39
Table 1. Summary of Hypotheses and Variables 53
Table 2. The Effects of Quality of Government on Software Policy Adoption in
Latin America (1995-2012)
56
Table 3. The Effects of Sectoral Association on Software Policy Adoption in Latin
America (1995-2012)
58
Chapter 4. State-Business Interactions in Industrial Upgrading Policy Processes:
The Case of Mexico’s Program to Support the Software Sector
65
Table 1. Characteristics of AMITI’s Members in 2001 76
Table 2. The Mexican Case at a Glance 87
Chapter 5. Industrial Upgrading Policies at the Subnational Level: The Cases of
Puebla and Nuevo León
95
Table 1. Innovation Potential of Selected Cases: Nuevo León and Puebla 99
Table 2. Overview of Case Studies and Independent Variables 100
v
List of Figures
Chapter 1. Introduction 1
Figure 1. Research Design at a Glance 8
Chapter 2. A Framework for the Study of Industrial Upgrading: Unpacking
Business-Government Interactions and Policy Outcomes
15
Figure 1. Business Government Interactions and Policy Outcomes Unpacked 15
Figure 2. Patterns of State-Business Interactions 26
Figure 3. Pathways to Industrial Upgrading Policy 28
Figure 4. Synergistic Pathway 34
Figure 5. Business-led Pathway 34
Figure 6. State-led Pathway 35
Figure 7. Quiescent Pathway 36
Chapter 3. The Adoption of Policies Towards the Software Sector: A Cross-
Country Quantitative Test, Latin America (1995-2012)
39
Figure 1. Adoption of Software Industrial Policies in Latin America (1996-2012) 55
Figure 2. The Effects of Bureaucratic Quality on the Likelihood of Software Policy 57
Figure 3. Effect of Sector Association on the Likelihood of Software Policy 59
Chapter 4. State-Business Interactions in Industrial Upgrading Policy Processes:
The Case of Mexico’s Program to Support the Software Sector
65
Figure 1. Patterns of State-Business Interactions 67
Figure 2. The Policy Adoption Process of Prosoft 81
vi
Acknowledgements
Some consider doctoral studies, and especially dissertation writing, a lonely time. But it
was not my case. I actually had some of the best company I could ever wished for in this journey.
Thus, I wish to express my gratitude to the many people that inspired and pushed me forward in
multiple ways.
First, I would like to thank Prof. Gerardo Munck for being an amazing advisor and mentor.
Gerry was always available to discuss my research offering timely and detailed feedback on the
numerous drafts and ideas I sent him. I am very grateful for his generosity in devoting time and
energy for this project; for believing that theory building is worthwhile; and for sharing his
research design and methodology expertise. Overtime Gerry became not only a great example of
scholarly commitment but also a deeply esteemed friend for me and my family. It has been a great
honor to work with him.
I also wish to thank Prof. Carol Wise, who encouraged me to come to USC and made sure
everything was going well throughout these years. Her warmth and cheerfulness kept me
motivated. And I thank Prof. Jeff Nugent who kindly gave me his insights whenever I shared my
work and met with him to discuss it.
Other USC faculty and colleagues also provided valuable assistance at different stages of
the dissertation. Professors Nick Weller, Ben Graham and Morris Levy guided me through the
quantitative methods section. Prof. Jeffrey Sellers led a directed reading on Comparative Political
Economy where I deepened my understanding of existing literature. A research assistantship with
Prof. Wayne Sandholtz was very formative in terms of case study writing. And professors Hernán
Galperín and Manuel Castells gave me great feedback for the empirical section. POIR program
fellows, especially those participating in the Comparative Politics Group, David Somogyi, Kelly
Zvobgo, Brian Knafou, Youssef Chouhoud, Joey Huddleston and Tom Jamieson, were an avid
audience whose questions and comments enriched several chapters. Thank you all.
I owe a large amount of gratitude to my friends from the USC POIR Latin American Group:
Fabián Borges, Juve Cortés, Christina Faegri, Victoria Chonn, Nicolás Albertoni, Scott Wilbur,
Mauricio Rivera and Bárbara Zárate who thoroughly commented several iterations of my work.
Beyond providing top quality scholarly support they became a Californian extended family. We
not only exchanged ideas, methodological advice and bibliographic references, but also shared
vii
picnics by the sea, family outings, job searches and many other life projects that I will cherish
forever.
I thank USC’s Political Science and International Relations Ph.D. program directors and
staff, especially Prof. Saori Katada, Prof. Christian Grose, Veri Chavarin, Linda Cole and Fanny
Cisneros, for always ensuring that I had the necessary resources to complete my degree.
I want to acknowledge too, the multiple institutions that provided financial support for my
doctoral studies and for this dissertation’s fieldwork: the Fulbright-García Robles Scholarship, the
Conacyt Scholarship, the Tecnológico de Monterrey Scholarship, the USC Graduate School
Advanced Graduate Fellowship, the POIR Doctoral Studies Fellowship and the School of
International Relations Summer Fellowships.
My support network extended across the border and had a strong Mexican component. I
am deeply grateful to Prof. Víctor López Villafañe for showing me that academia was a worthwhile
alternative for professional development and inspiring me to pursue postgraduate studies. I also
thank Prof. Anne Fouquet for first giving me the chance to experience fieldwork and always
encouraging me to complete the Ph.D. And thanks to Prof. Lucrecia Lozano for sparking my
interest in Latin America and the ECLAC school of thought, and for endorsing my doctoral and
fellowship applications. I also wish to express my gratitude to the deans and directors at
Tecnológico de Monterrey for their support: David Garza, Alberto Bustani, Humberto Cantú,
Mauricio González, Gabriel Cavazos, Araceli González and Juan Carlos Cobián. And to my
longtime friends in Monterrey for all their backing along the way.
Throughout fieldwork in Mexico City, Puebla, and Monterrey I encountered very kind
persons who set aside some time of their busy agendas to talk about Prosoft and the policy making
processes, amongst them: Sergio Carrera, Juan Carlos Moreno Brid, Rocío Ruiz, Jesús Orta,
Teresa Carrillo, Alfredo Pacheco, Jesús de la Rosa, Javier Allard, Eloísa Talavera, and several
other policy makers and entrepreneurs.
Last but not least I wish to thank my family. My parents Oscar and Marcela for teaching
me the value of pursuing intellectual challenges and perseverance. The many conversations with
my dad about politics and economics strongly influenced my academic interests. And from my
mom I learned that it is possible to achieve a work-family balance. I also thank them for their
generous support especially during the dissertation writing period. My mom spent some months
in L.A. helping me with the kids and the household chores so that I could focus on my research.
viii
And both of my parents were wonderful hosts during my fieldwork trips to Mexico. The rest of
my family was an important source of love and encouragement: Andrea, Oscar, Rodrigo, Jaime,
Rosy, Linda and Roberto. Thank you all.
Finally, I am profoundly grateful to my nuclear family for being a place of inspiration and
making me keep things under perspective. My husband Jaime, offered unconditional support
through all these years. I do not have enough words to thank him for all what he did for me and
our kids. He was a cornerstone in this project. I also thank my son Emilio, who was an exceptional
companion during the time we spent in California. I hope some day he reads this “story” for which
he sweetly offered me help several times. And I thank my daughter Natalia, who joined us in the
middle of the dissertation writing stage, for bringing alongside a smile that made me (and the rest
of the family) joyful during the toughest times. I hope that somehow through my work I contribute
to create a better world for you Jaime, Emilio and Natalia.
1
Chapter 1.
Introduction
Latin American countries experienced high rates of growth and even managed to weather
the global financial crisis that hit the world in the first decade of the 21
st
century. So, after having
experienced debt, crisis and structural adjustment in the 1980s and 1990s this could be seen as
quite good news. In fact, policymakers have learned from past mistakes and devised prudent
macroeconomic policies such as accruing high levels of foreign reserves, regulating hot capital
inflows, and maintaining healthy fiscal balances. Nonetheless, the panorama is less bright if we
scrutinize more closely those activities that are fueling economic growth: exports of soy (Brazil
and Argentina), copper, iron ore (Chile, Peru) and oil (Venezuela).
Just as one century ago, Latin America is still relying largely on the primary exports and
importing manufacturing goods. Of course, many countries have already an industrial base that
grants them a leeway when the prices of commodities drop sharply. But the global context has
changed dramatically: we are living in an era where the production of knowledge has become
crucial. Latin America is caught in a middle-income trap
1
(Foxley 2012; Paus 2014). Innovation
intensive activities provide the highest value-added, and Latin America is lagging behind in this
realm.
2
In many ways, just as breaking away from a strong reliance on primary exports and
beginning a process of industrialization was seen as the core challenge for Latin America in the
1950s, today industrial upgrading is critical if countries aim to sustain high growth rates with better
equity. In fact, development experts have noted the risks of neglecting the industrial upgrading of
1
Middle-income trap is a situation where producers can no longer compete internationally in standardized, labor-
intensive goods because wages are relative too high, but they are unable to compete in higher value-added activities
on a broad scale, because productivity is relatively too low (Paus 2017, 4).
2
In 2017 the authors of the Global Innovation Index report asserted: “Although important regional potential exists,
the Global Innovation Index rankings of countries in Latin America relative to other regions have not steadily
improved. In recent years and in 2017, no economies from this region are identified as innovation achievers -Those
growing above the expected given their level of economic development measured as GDP per Capita-.” (2016, 36).
Appendices 1 and 2 show additional information on the Global Innovation Outputs ranking contents and data.
2
production into higher technology levels: “In order to avoid the resource curse, rents have to be
purposefully distributed against comparative advantages, fostering diversification of production in
knowledge intensive activities” (Cimoli, Dosi, and Stiglitz 2009, 556). International institutions
like ECLAC, the World Bank and OECD are also encouraging governments in the region to adopt
policies that increase innovation rates.
Some have argued that closer business-government collaboration is the most effective way
to approach industrial upgrading (Fernández-Arias et al. 2016; Rodríguez-Clare 2005; Rodrik
2007, 2008; Schneider 2015). But there is scant theoretical discussion of the process and the
potential outcomes of such collaboration in the digital era. We still don’t have a full understanding
of under what conditions do business-government collaboration for industrial upgrading towards
high-technology activities emerge? And what are the outcomes of such business-government
collaboration?
This dissertation studies the process of industrial upgrading in democratic developing
countries. To approach this topic, I will answer questions concerning the adoption and design of
industrial upgrading policies through a mixed-methods research strategy. And in doing so, I will
highlight the interactions between state and business. Regarding the adoption of industrial
upgrading, I tackle these general questions: Who are the actors, interests and coalitions that drive
the adoption of industrial upgrading policies? Why do some countries (and provinces) adopt
industrial upgrading policies and others do not? Concerning the design of industrial upgrading, I
will seek to answer the questions: What accounts for differences in the characteristics of industrial
upgrading programs? How might variation in the quality of government and business cohesion
shape the measures adopted?
To enhance policy advice, we need a better understanding of the politics of industrial
upgrading and especially of the role of business in this realm. As Schneider notes: “It makes little
sense to focus exclusively on political inputs like public opinion, voting, parties, and the
institutions that shape them if, in fact, policy outputs depend on other actors, like business, which
distort these inputs or twist later stages of policymaking and implementation or simply prevent
policies and reforms from ever getting on the agenda…Development theories and policies that do
not take into account the structures and strategies of business groups (and their roots in
oligopolistic markets) will remain abstract and ungrounded” (2014, 21–22).
3
The present chapter introduces this dissertation about the political economy of industrial
upgrading. The first section defines the dependent variable, the main research questions, and goals.
The second section outlines the theoretical framework. Section three explains the empirical
strategy. Section four contains a brief overview of each chapter. The last section synthesizes the
main contributions and conclusions of this dissertation.
1. Research Questions and Goals
1.1. Dependent Variable
Before going further, it is important to have a clear definition of the main dependent
variable: industrial upgrading policy. A broad definition of industrial policy consists of “policies
affecting infant industry support of various kinds, but also trade policies, science and technology
policies, public procurement policies affecting foreign direct investment, intellectual property
rights and the allocation of financial resources. Industrial policies, in this comprehensive sense
come together with processes of ‘institutional engineering; shaping the very nature of economic
actors, the market mechanisms and rules under which they operate , and the boundaries between
what is governed by market interactions, and what is not” (Cimoli, Dosi, and Stiglitz 2009, 1–2).
Chang (2002) defines industrial policy as interventions which seek to alter the production vector
of goods and services through the creation of new activities.
Industrial upgrading policies are defined here as those policies whose objective is the
creation or consolidation of high-technology sectors in which knowledge plays a critical role, for
instance software, biotechnology and aerospace. Traditionally, the main tools used by
governments have been fiscal incentives, trade protection, promotion of direct investments (state
or foreign-owned), and credits financed by national development banks. But in recent years new
strategies have emerged which involve closer business-government coordination such as fostering
clusters, stimulating innovation and coordinating collective action (O’Riain 2004; Peres 2009).
Moreover, industrial upgrading policies are layered, which means that they are created and
implemented at multiple levels (national, regional and even local level).
1.2. Research Questions
This definition of industrial upgrading lends accuracy to the overarching question of this
dissertation: What accounts for the success of developing countries in adopting industrial
upgrading policies for their high-tech sectors? However, to make this question more
4
analytically tractable, it is useful to disaggregate the analysis: here I focus on the adoption and
design stages of policy making. There are good reasons to anticipate that better theory and better
empirical results will be obtained essentially breaking down the examination of the process.
1.3. Research Goals
Despite the great value of prior developmental state works related to the topic of industrial
upgrading in developing countries, the issue of under what conditions is business-government
cooperation beneficial for the policy process of industrial upgrading is still unresolved. In other
words, we need to understand why and how can rent seeking shift into coordination that leads to
development, especially through activities with higher technological content.
This dissertation has two broad research goals. The first goal is to make a theoretical
contribution. So far we lack a detailed framework that allows us to understand why and how some
developing countries adopt industrial upgrading strategies and others do not, what types of policies
are being encouraged, and how business-government interactions lead to positive or negative
outcomes in more democratic settings. This dissertation constructs such a framework by
emphasizing the characteristics of business and state structures, and the potential outcomes of their
interaction. I systematize the existing literature on related matters, identify some hypotheses, adapt
them to the challenge of industrial upgrading, and identify the gaps that require theory generation.
The second goal is to offer empirical tests of conflicting or complementary arguments and
provide some guidance regarding how Latin America is (or is not) developing high tech industries
in the current knowledge-intensive era.
2. Theoretical Framework
The analysis of industrial policy has not been recently in the comparative political economy
research agenda.
3
This is partly because during the 1980s and until late 1990s most countries
adopted neoliberal models and government involvement in the economy was cut down. In this
dissertation, I build on the developmentalist chain of causal explanation
4
, especially on the
3
Two exceptions are Whitfield et.al. (2015) and Schneider (2015) but their focus is not on industrial policy towards
high-tech sectors. Meanwhile, Fernández-Arias et.al.(2016a) offers a compilation of 25 case studies of public-private
collaboration across several countries and industries in Latin America, but their approach is more policy-oriented
placing less emphasis on theory building and testing.
4
Institutional scholars, who grant the State a central role in economic development, have endeavored to understand
how institutions and non-institutional forces from the political and economic realm, such as the creative (or
destructive) action of politicians and economic entrepreneurs, and the structural factors such as power distribution,
5
developmental state literature and more recent business politics/varieties of capitalism scholarship
to generate and test a framework for the analysis of the political economy underlying industrial
upgrading processes in the digital era.
The developmental state works confronted the neoliberal view that predicted that economic
development would follow the removal of trade barriers and the reduction of state involvement in
the economy. Through the careful study of the trajectories of the newly industrialized East Asian
countries in the eighties and nineties, authors revealed how purposeful state action was needed to
stimulate industrialization (Amsden 2001; Evans 1995; Haggard 1990; Johnson 1982; Wade
1990). Furthermore, authors showed how countries used selective trade protection to protect infant
industries and promote exports of already developed firms (Wade 1990, Evans 1995). Evans
(1995) demonstrated that with “embedded autonomy”
5
, the state may facilitate structural
transformation of business and promote the emergence of new productive activities.
One shortcoming of the theoretical propositions of developmental state analyses is that
their empirical cases were mostly authoritarian regimes, so they overlooked that under democratic
conditions the policies for upgrading can become subject to different types of struggles. For
instance, existing congressional members will have diverging interests depending on the region
they represent and, in general, policymakers are constrained by the short term need to secure
victory in elections.
Moreover, given that this body of literature emerged mainly in the eighties and nineties
prior to the technological revolutions that dramatically changed the context of development
6
only
a few studies have assessed the efforts that developing countries are displaying to encourage high
technology activities in the digital era. Two recent works on the neo-developmental state (Breznitz
2007; O’Riain 2004) offer rich ideas about the emergence of high tech industries in small middle
income countries (Ireland, Israel, Taiwan) that are strategically related (to Europe or the United
lead to different development outcomes (Mazzuca 2015, 21). Two chains of causal explanations organize the
institutionalist literature: the smithian chain and the developmentalist chain. In the smithian stance (new institutional
economics, rational choice), the State is a precondition to the Market, a supplement that will ensure property rights
protection and public goods provision. In the developmentalist chain of explanation (structuralist, developmental state,
business politics, varieties of capitalism) the State is also an actor. One that will create an economic structure or sector
when not all the background conditions for growth are given.
5
Embedded autonomy consists of a situation where meritocratic bureaucrats are closely linked with the private
sector but preserve sufficient distance for renegotiation and monitoring of goals and policies for national economic
development.
6
For a thorough review of the transformations leading to the digital era see Castells (2000).
6
States). These merit more careful examination and exploration in countries where size is larger
and/or geography is not so favorable.
Overall, developmental state authors fleshed out the characteristics of government, but
placed less emphasis on the characteristics and structure of the business sector. Meanwhile, more
recently, comparative political economy has flourished with works studying how firms organize
and exert their power to impact policy decisions. Thus, I take these business politics works as a
second and complementary building block.
The business politics literature -which includes varieties of capitalism- shows how different
configurations of businesses, and to a certain extent states
7
, lead to diverse policy outcomes.
Nonetheless most of the analyses examine cross-sectoral policies at the national level and not
specific sectors. Examples of this approach focusing in advanced industrialized countries are:
Culpepper (2001); Hall and Soskice (2001); Hancké (2009); and Mares(2003). On developing
economies: Nolke and Vliegenthart (2009); and Schneider (2008, 2013a).
Included in this literature are also studies of structural adjustment and liberalization
processes in developing countries which offer valuable theoretical and empirical contributions.
These works show that varieties of states (authoritarian vs. democratic) and types of business
(weak vs. strong) generate distinct paths and outcomes to structural adjustment and liberalization
(Etchemendy 2011; Kingstone 1999; Özel 2015; Schneider 2004; Silva 1998; Thacker 2000). The
value of business coordination and the strategies that firms might pursue to shape the agenda of
policymakers has also been carefully theorized and examined in Schneider’s seminal work (Doner
and Schneider 2000; Schneider 2004, 2010, 2015; Schneider and Maxfield 1997), and in recent
taxation policy studies (Castaneda 2016; Fairfield 2015).
I take these two bodies of literature: developmental state and business politics, to generate
an analytical framework for the study of industrial upgrading policy. My general argument is that
two factors shape the prospects of industrial upgrading: the quality of the bureaucracy and the
business sector’ degree of organization. And we should consider how these two interact in a more
democratic context if we strive for a full understanding of the proximate causes of development.
This framework of analysis posits there are four models of business-state interactions: synergistic,
state-led, business-led and quiescent, generating varied policy outcomes.
7
Schmidt (2009) criticized the varieties of capitalism approach for not carefully considering the types of state
structures different countries have.
7
3. Empirical Strategy
To narrow the study of industrial upgrading strategies into a more manageable
phenomenon, I have chosen to focus most of the dissertation on the information technology (IT)
sector, more specifically the software sector. The IT sector is characterized by continuous change
and innovation. New products appear on the market frequently and solutions to old and new
problems are sought constantly. The costs of research and development in this area are often high,
so intense collaboration can allow firms, universities and government to spread these costs. The
software sector is strategic and can be directed towards aiding existing or new industries. Another
characteristic is that the level of job skills created in this sector varies according to the type of
service provided. It can range from testing and making minor adaptations to computer programs,
to developing full solutions and new software for high tech products. Given the omnipresence of
IT in the 21
st
century, we should observe efforts to promote it in almost every country. Also an
advantage of studying this sector is that it is comprised of small and medium enterprises, and other
sectors in the economy are like that so we need to have a better understanding of this.
One limitation of focusing on just one sector is that there is potential selection bias and
limits the generalizability of my findings. It may be the case that the dynamics observed in the
software sector are particular of this sector and not of others. One solution would be to conduct
similar analyses of other sectors like biotechnology and aerospace. On the other hand, the choice
of this sector also allows me to continue and build around a tradition of developmental state
scholarship that studied informatics (Breznitz 2007; Evans 1995; O’Riain 2004).
Throughout my dissertation, I concentrate empirically on Latin America. I will say more
about case selection in what follows. However, as a broad point I would highlight the following
consideration. I am concerned with democratic developing countries, because these countries have
not been studied much in the relevant literature and because I am concerned with the
generalizability of arguments stemming from other contexts. On the one hand, developmental state
studies focused considerably on non-democratic countries. On the other hand, studies of industrial
upgrading have focused mainly on big European countries (e.g. Germany); the only exceptions are
a few studies on Ireland, Israel, Taiwan and India (Breznitz 2007; O’Riain 2004). Thus, my
8
research on Latin America seeks to fill a gap in current research by focusing on the only region of
the developing world that nearly entirely has democratic regimes
8
.
I employ a mixed-methods approach to validate and refine the analytical framework of this
dissertation. First, I test the importance of business cohesion and quality of government controlling
for other variables through an event-history analysis of the adoption of policies to stimulate the
software sector in 17 Latin American countries from 1995-2012. Next, I undertake a qualitative
analysis at two different levels: an in-depth case study of the Mexican case at the federal level
from 1990-2005; and a set of comparative case studies at the subnational level of two Mexican
cases in two periods 2000-2007 and 2008-2015.
The input for the case studies came from field research conducted in Mexico (Mexico City,
Nuevo Leon and Puebla) in 2014 and 2016 where I interviewed 26 key actors. I also resorted to
primary sources. Business associations’ annual reports and policy proposals were fundamental to
understand the private sector stance and influence regarding the adoption and design of industrial
upgrading policies. And regional and national economic development plans; legislative acts; as
well as numerous reports and assessments conducted by the Ministry of Economy and other
independent agencies, provided information on the federal and regional governments’ position.
Secondary sources were valuable sources of information as well. Through national and regional
newspapers, I was able to reconstruct with detail the various policy processes that occurred in
Mexico.
Figure 1. Research Design at a Glance
8
Cuba is the obvious exception and, more recently, Venezuela.
Large-N Analysis
Chapter 3.
Cross-national
event history
analysis
17 Latin American
countries
(1995-2012)
Small-N Analysis
Chapter 5.
Comparative case
studies of
subnational units in
Mexico: Nuevo León
and Puebla
(2000-2015)
Small-N Analysis
Chapter 4.
In depth single
country case
study of Mexico
(1990-2005)
9
4. Overview of the Chapters
Chapter 2 lays out the theoretical framework that builds on developmental state and
business politics works. It presents the actors, preferences and strategies of the industrial upgrading
policy process. The central argument and general hypotheses that guide the empirical section are
there. Some of the more specific questions and hypotheses are introduced in the later chapters.
Chapter 3 addresses the following questions: Why do some countries adopt industrial
upgrading policies and others do not? And more specifically: Do state and business characteristics
influence the decision to adopt or not such policies? I argue that certain government and business
characteristics shape the adoption of industrial upgrading policies. My hypotheses are that
countries with bureaucracies insulated from political pressures recruited through a meritocratic
mechanism, and a formally organized business sector, are more likely to adopt programs to
stimulate high technology sectors than the rest of the countries. To test them I conduct the first
cross-country statistical study of the adoption of industrial upgrading policies in developing
countries through an event history analysis of the policies to stimulate the software sector in 17
Latin American countries from 1995-2012. After controlling for other economic, political and
international factors, I find that countries with higher bureaucratic quality are more likely to adopt
industrial upgrading policies. The presence of a sectoral association also has a positive effect on
the likelihood of adoption of industrial upgrading policies. Yet one limitation of the study is that
given sample size, the interaction effects cannot be explored. Thus, in subsequent chapters, to
further understand the phenomenon, I resort to qualitative research tools such as process-tracing
and structured comparison of cases.
Chapter 4 studies state-business interactions underlying industrial upgrading policies to
support the software sector in a Latin American country: Mexico. The two guiding research
questions are: What types of state-business interactions occur in industrial upgrading
policymaking? How do the outcomes (types of policies adopted) of such interactions vary? I argue
that the level of bureaucratic quality and the degree of business articulation generate a variety of
patterns of state-business interactions. These patterns lead to the adoption of different types of
industrial policies resulting in distinct upgrading outcomes. To illustrate the argument and further
build the theory I conduct an in-depth case study the adoption of the Program to Support the
Software Industry (Prosoft) in the early 2000s. The in-depth Mexican case reveals how a tight
group of high quality bureaucrats and a relatively cohesive business sector with clear goals,
10
generated a synergistic process of policy design and adoption. The outcome of such pattern of
collaboration was a program in which funds are allocated through clear criteria and a competitive
process.
Chapter 5 zooms into the subnational level and seeks to advance knowledge in two ways.
First, it offers an additional test of the analytical framework. And second, it refines the framework
by considering the specificities of the subnational dimension which has been given scant attention
in the political economy of industrial upgrading literature. The guiding research questions are:
How do business cohesion and bureaucratic quality shape the processes of adoption of industrial
upgrading policies at the subnational level? And are there other relevant actors that impact the
adoption and policy design? The empirical test consists of a set of structured comparative case
studies of two Mexican states: Nuevo León and Puebla, that vary on my key independent variables.
I assess the policy making process regarding various specific industrial upgrading initiatives; and
consider how bureaucratic quality and business cohesion affect the prospects of adoption of
industrial upgrading policies. I find that at the subnational level bureaucratic quality is harder to
achieve and the adoption of programs to stimulate high tech sectors is still highly dependent on
the leadership and support of the governor in place. But cases reveal that business cohesion can
actually become an important element in the adoption of either: collaborative public-private
initiatives; federal initiatives; or business-led programs to by-pass the lack of regional government
support.
Lastly, Chapter 6 presents the main contributions and conclusions of this dissertation as
well as the research agenda. In the next section I briefly anticipate the contributions and
conclusions.
5. Contributions and Conclusions
Analyzing industrial upgrading and high-tech sectors when countries have unresolved
problems like large levels of inequality, insecurity and violence might seem odd. Are there not
more basic and urgent tasks at hand? Yes, and no. I consider that one of the underlying reasons for
these problems stems from the lack of opportunities for human development, and the failure to
exploit the creativity and capacity of a young workforce. Moreover, moving into more intensive
uses of technology can stimulate other economic sectors like agriculture. After all technology can
have a cross-sectoral impact. The example of how Brazil has fostered biofuels is illustrative of this
11
point. Secondly, higher technology activities can be targeted to address some of the most pressing
problems. For instance, an entrepreneur of the software sector in Puebla, Mexico designed a system
to track the transactions of people in pawn shops (casas de empeño) where burglars often sell the
goods they steal. And in Nuevo León, Mexico a non-governmental organization, the Center for
Citizen Integration, created a crowd-source app to report problems (infrastructure, crimes) to local
authorities and follow up on their response.
The focus on business-government relations is another contribution of this project. The
prevailing notion is still that of rent-seeking. In developing countries, the main outcome of such
exchanges are often billionaire contracts accompanied by side-payments.
9
Nevertheless, history
shows us that business-government cooperation has been key to advance the living standards and
economic development of countries (For instance, the literature on the emergence of the welfare
state or the emergence of the automotive industry in Latin America). In theory, elected politicians
have an incentive to be concerned about economic growth and prosperity but we need to specify
those arrangements that constrain special interests.
Finally, a word about the value of this project for Latin American studies. Scholars like
Acemoglu and Robinson (2012, 9) underscore the importance of colonial legacies for
development: “Why are institutions of the United States so much more conducive to economic
success than those of Mexico or, for that matter, the rest of Latin America? The answer to this
question lies in the way the different societies formed during the early colonial period. An
institutional divergence took place then, with implications lasting into the present day.” The
panorama is grim. But these views underestimate the role of agency and the potential benefits of
coordination between business and government that I consider crucial. Technology is known as a
market failure given the high levels of risk and investment required, so it cannot be left to the
markets will. It requires public-private cooperation. What we need is a more accurate explanation
of when and why industrial upgrading policies are adopted and why they become successes or
failures. We cannot just leave our fate to what happened four hundred years ago. We need to find
ways to catch up on innovation capabilities and activities that are crucial for development.
9
The most recent example is the Odebrecht scandal where a network of corruption transcended Brazilian borders and
spread across countries in the region involving some of the presidents and high-rank authorities of Peru, Argentina,
Mexico amongst others.
12
The results of this dissertation, suggest that not all states and all business sectors are
created equal: some may actually foster efforts to transition into knowledge intensive activities.
This challenges, once again but now in the digital era, the well-entrenched view against both, state
intervention in the economy and close business-government collaboration in developing countries,
which to many scholars and policymakers only leads to rent-seeking (Bates 1981; Krueger 1974,
1990). Under certain conditions, bureaucracies and business leaders, and not just free markets,
may be the drivers of industrial upgrading.
13
Appendix 1.
Knowledge and Technology Intensive Outputs and Creative Outputs
By World Region, 2017
Source: WIPO, Cornell University, and INSEAD (2017)
14
Appendix 2.
Knowledge and Technology Intensive Outputs and Creative Outputs
By Income Group, 2017
Source: WIPO, Cornell University, and INSEAD (2017)
15
Chapter 2.
A Framework for the Study of Industrial Upgrading:
Unpacking Business-Government Interactions and Policy Outcomes
The analysis of industrial policy adoption has not been recently in the research agenda of
political scientists focusing in developing countries. This is partly because during the 1980s and
until late 1990s most developing countries adopted neoliberal models and government
involvement in the economy was cut down. Still, there is considerable research that is relevant to
an inquiry on industrial upgrading in democratic developing countries.
The objective of this chapter is to generate a framework of analysis to study the political
economy of industrial upgrading policies by building on developmental state and business politics
works. The main argument is that the level of bureaucratic quality and the degree of business
cohesiveness generate different patterns of state-business interaction. These patterns give way to
the adoption and design of a variety of industrial policies types that produce different upgrading
outcomes. The causal mechanisms are the strategies used by bureaucrats and firms to pursue their
interests which result in the types of industrial policies adopted (See Figure 1).
Figure 1. Business Government Interactions and Policy Outcomes Unpacked
State
(Bureaucratic quality)
Firms
(Business cohesion)
Patterns of
Business-
State
Interaction in
Policy
Making
Types of
Industrial
Policies
Industrial
Upgrading
Causal
mechanisms
Strategies used
by bureaucrats
and business
16
In the first section I introduce the building blocks from developmental state studies that
examined business-government interactions for industrial policy making in authoritarian contexts
and more recent literature on business politics. The second section presents the framework of
analysis and some general propositions that will be refined and tested in chapters 3, 4 and 5 of this
dissertation.
1. Building Blocks
1.1 The Legacy of the Developmental State Literature
The developmental state literature revealed how purposeful state action was needed to
stimulate industrialization through the careful study of the trajectories of East Asian and Latin
American countries (Amsden 2001; Evans 1995; Haggard 1990; Johnson 1982; Wade 1990).
Authors showed how governments used trade and investment restrictions to protect infant
industries and promote exports of already developed firms to encourage a transition into more
complex manufacturing with varying rates of success.
Yet early developmental state works were criticized for granting the state too much
attention while neglecting the private sector role. Thus, later works examined business-government
relations and explicitly focused on the interactions between state actors (executive and
bureaucrats) and firms. These studies found that indeed business and their organizations impacted
economic development outcomes (Doner 1992; Evans 1995; Haggard, Maxfield, and Schneider
1997; Kang 2002; Kohli 2004).
Today we know that business policy preferences and capabilities to solve collective action
problems were fundamental elements for the success or failure of development policies (Doner
1992). The importance of a bureaucracy that has embedded autonomy: close relationships with
business, exchange communication and preserve objectiveness is also part of the success stories
(Evans 1995). Thus in spite of the dominant public choice view predicting that the outcome of
close-business interactions always generates rent-seeking (Buchanan, 1980; Grossman &
Helpman, 2001; Krueger, 1990; Olson, 1982), case studies have revealed that under some
conditions: embedded bureaucracy and having firms organized into associations, business-
government cooperation can generate positive outcomes (Cammett 2007; Doner and Schneider
2000; Haggard, Maxfield, and Schneider 1997; Nugent and Sukiassyan 2009; Özel 2015).
17
Other studies have examined the effects that the balance of power between state institutions
and business elites have in the implementation of industrial and economic policies targeting
growth. They found that concentrated elites and a cohesive state generated more favorable
conditions (Kang 2002; Kohli 2004).
While prior studies represent an important stepping stone, few of them systematically
examined into detail the policy-making processes and the causal mechanisms underlying business-
government effects in outcomes. Furthermore, because most works were undertaken in the nineties
they focused on the industrial transition from agricultural to manufacturing or from heavy
manufacturing industries to light manufacturing industries.
10
Two recent works on the neo-
developmental state (Breznitz 2007; O’Riain 2004) offer rich ideas about the emergence of high
tech industries in small middle income countries (Ireland, Israel, Taiwan) that are strategically
related (to Europe or the United States). These merit more careful examination and exploration in
countries where size is larger and/or geography is not so favorable.
1.2. Recent Works about Business Politics
Comparative political economy studies scrutinizing business-government interactions in
various policy realms such as corporate governance, social welfare, and taxation provide a second
relevant building block. This business politics literature -which includes varieties of capitalism
works- shows how different configurations of businesses, and to a certain extent states
11
, lead to
different policy outcomes. Nonetheless most of the authors examine cross-sectoral policies at the
national level and not specific sectors For developing countries see: (Nolke and Vliegenthart 2009;
Sánchez-Ancochea 2009; Schneider 2013a). For advanced industrialized countries see: (Culpepper
2001; Hall and Soskice 2001; Hancké 2009). Specifically on social protection systems are:
(Culpepper 2011; Hacker and Pierson 2002; Martin and Swank 2012). And on corporate
governance see (Culpepper 2011).
Relatedly, the works on Latin American political economy make valuable contributions
theoretically and empirically to the comprehension of developing countries. These studies find that
different states (authoritarian vs. democratic) and types of business (weak vs. strong) generate
10
Ornston (2012) study of three northern European countries, Finland, Sweden and Denmark, and his concept of
creative corporatism, is an interesting approach. But he is more focused on economic performance and how these
countries have spurred innovation by promoting coordination between government, encompassing associations and
labor, rather than on specific sectoral industrial policies.
11
Schmidt (2009) criticized the varieties of capitalism approach for not carefully considering the types of state
structures different countries have.
18
distinct paths and outcomes to structural adjustment and liberalization (Etchemendy 2011;
Kingstone 1999; Özel 2015; Schneider 2004; Silva 1998; Thacker 2000). The value of business
coordination and the strategies that firms might pursue to shape the agenda of policymakers is also
carefully theorized and examined in recent taxation policy studies (Castaneda 2016; Fairfield
2015).
So, in the rest of this chapter I take these two building blocks – developmental state and
business politics- and their ideas about business-government relations to construct a framework
that will aid in a better understanding of industrial upgrading policy processes.
2. The Model: Industrial Upgrading Policy (Adoption and Design) Processes
2.1. Main Actors in Industrial Policy Making
Several actors may be involved in industrial policy making, yet the two most critical in the
adoption, design and implementation of sectoral programs are: government and business. The
following section provides further detail about the characteristics and classification of government
and business actors which are important to have a full understanding of the policy process. Two
general categories of state actors are non-elected and elected. While business players include
individual firms and business associations. Each of them, “with its preferences and incentives, and
within the constraints of the rules that frame its engagement, meet in different arenas to define
public policies” (Scartascini, Stein, and Tommasi 2010, 6).
2.1.1. State Actors
State actors can be classified into elected and non-elected, depending on whether voters
participated in their selection process. Elected actors are the president and the legislators, who
reach their seats through an electoral race and are held accountable to voters. Meanwhile non-
elected actors are ministers and bureaucrats. Ministers are appointed by the executive and/or the
legislative and may undergo a ratification process. Bureaucrats can be chosen by the ministers who
build a team to work with or through civil service career processes (competitive/merits-based).
Classic developmental state works stressed the role of state officials but mostly in
authoritarian contexts (Amsden 1989, 2001; Evans 1995; Johnson 1982; Wade 1990). More recent
studies have emphasized the role of elected politicians and the incentives they have to implement
industrial policy mostly deriving from firms’ campaign donations, and their support to the
19
incumbent government in exchange of favorable conditions (Whitfield et al. 2015)
12
. But they
underplay the potential critical role of meritocratic structures.
I contend that at the federal level the president and the legislative play a veto role and they
can start, amend or halt the proposals that stem from business-bureaucracy interactions.
Meanwhile non-elected actors, especially bureaucrats, play a crucial role in industrial policy
making. As Stein et. al (2010, 11) remark: “Ministers and other political actors like presidents and
legislators must rely heavily on the bureaucracy to convert policy ideas and laws into specific acts
of government… Having a technically competent and independent bureaucracy to which some
policy decision making and implementation may be delegated can facilitate intertemporal
agreements, particularly in policy areas that are prone to politicization and political opportunism.”
High-rank bureaucrats can also initiate and promote industrial policy, sometimes alongside with
business. And they may even work together to secure the support of elected state actors.
Elected State Actors I. The President
Role. Developmental state literature demonstrated how the leadership of the president was critical
to the enactment of policies (Evans 1995; Johnson 1982, 1998). But most of the cases examined
were set in authoritarian contexts. In a more democratic setting, even if the executive power has
the will to support certain strategic activities, it will also need the support of the bureaucracies and
the legislature (Scartascini, Stein, and Tommasi 2010).
Preferences. The president will support those programs that grant him political advantage- votes,
campaign funding etc. So, unless there are already major firms in high-tech sectors offering
significant resources; or firms in the sector are well organized; and/or his own background relates
to innovation intensive activities, the presidents’ sectoral policies will most likely continue to cater
12
Whitfield et.al. (2015,11) advance an approach they call the “Political survival of ruling elites”. They consider that
the main driver behind state elites’ policy choices and implementation, and their interaction with business is their
imperative to remain in power. Approaches in this group all argue that motivations, calculations and coalitional
strategies of ruling elites have significant influences on policy choices and changes. Thus, policy choice and
implementation cannot be separated from ruling elites’ strategies to ensure political survival. Economic policies are
rarely chosen for public or private interests, but for regime interests such as consolidating power and expanding
political support. Proponents of this approach argue that ruling elites always channel state resources to key
constituencies forming the coalitional base of their regime, and to powerful groups whose support they need to coopt,
in order to maintain political stability. One of the limitations of this approach is that the authors assume a unitary
government. Because in a divided government, each party will try to appeal and serve the key constituencies, so the
ruling elites decisions might not always be approved and thus it may not be the only channel through which business
and bureaucrats promote and adopt policies that are favorable to them. The case studies focus in Africa traditional
industries (agribusiness), and thus do not consider policies that may alter the status quo or involving relatively “weak”
or emerging sectors like software.
20
the traditional sectors such as the agribusiness, steel, automotive etc. Another way the executive
could be convinced to support new sectors is if within his closest circle of advisors there are
compelling proponents of industrial upgrading policy.
Elected State Actors II. The Legislature
Role. As countries move into increasingly democratic settings, more actors participate in
policymaking. Specifically, the congress can emerge as an actor because often, the policies
approved by the executive are subject to congressional ratification or budget allocation. Also,
legislators constitute a new entry point for policy proposals and for business to exert their influence
either through lobbying, electoral funding and/or resorting other means such as bribes and favors.
However, in the case of Latin America the congress is more of a veto player and perhaps an
increasingly important actor in the process of adoption, design and implementation. As Saeigh
(2010, 49) notes, “recent studies suggest that while legislatures in the region in general may not
be heavily involved in formulating and advocating policy change, they are nonetheless relevant to
policy outcomes. Legislatures in some countries are active in policymaking in the sense of mainly
being blunt veto players, blocking legislation proposed by the executive. Others, however are
involved in negotiating policy issues behind the scenes with the executive, or in amending or
reformulating executive, legislative initiatives.”
Preferences. Congressmen, as the executive, are also concerned about retaining voter’s support
and in theory they also have a party/territorial affiliation they must tend. The territorial dimension
may be of relevance especially in unequal countries, like Latin America. Legislators from
manufacturing regions that have institutions for innovation may advocate for industrial upgrading
policies while those from agricultural regions will probably not.
13
Non-elected State Actors I: Bureaucrats
Role. The role of bureaucrats in policy adoption, design and execution on industrial upgrading
processes has been documented and theorized extensively (Evans 1995; Johnson 1982; Kohli
2004; Rauch and Evans 1999; Schneider and Maxfield 1997). Evans (1995) concepts of
meritocratic bureaucrats and their embedded-autonomy offer the most refined version. In this
13
Whitfield (2015) et.al. argue that the strength of excluded factions is the key causal mechanism for explaining why
some ruling coalitions, seem to display a longer-term development vision (and thus enact industrial upgrading
policies). Weak excluded political factions give ruling elites a relatively high degree of stability, potentially allowing
it to undertake economic policies that require a longer time period to bear fruit. The degree of vulnerability of the
ruling elites shapes whether a pocket of efficiency emerges (endogenous rather than exogenous factor) (97).
21
view, high rank bureaucrats would collect information and devise plans to spur particular sectors.
Highly selective meritocratic recruitment and long-term career rewards create commitment and a
sense of corporate coherence. Corporate coherence gives these apparatuses a certain autonomy.
They are not, however, insulated from society as Weber suggested they should be. They are
embedded in a concrete set of social ties that binds the state to society and provides
institutionalized channels for the continual negotiation of goals and policies.
More recently, the idea of a meritocratic bureaucracy has been called into question. Some
authors argue that it is not a fundamental requisite to promote industrial upgrading and that there
may be alternative strategies such as creating specialized agencies or bringing-in business
expertise through the revolving-door mechanism (Breznitz 2007; Whitfield et al. 2015). But these
studies do acknowledge that the most successful transformation programs rely on at least some
bureaucratic agents (“pockets of excellence”) with strong capabilities to collect information and
think about the long term, which has spurred the promotion of high tech industries
Preferences. Bureaucrats have preferences on their own. In fact, principal-agent issues, where the
principals are elected politicians and agents are bureaucrats, have been subject to extensive
analysis in developed countries (Huber and Shipan 2002; McCubbins, Noll, and Weingast 1987).
Authors posit that agency costs are the costs created by agent’s trying to promote their own agenda
rather than that of the principal (Posner 2010). Some of the forms such agency costs take are:
slacking off, influence activities, and hoarding information to make one-self indispensable.
Bureaucrats will seek to adopt those policies in which the agencies they work for acquire more
funding and responsibilities because these elements secure their survival. In the theory of
bureaucratic politics, ministries push their narrow institutional interests in a bargaining game
(Allison and Zelikow 1971). Policy outcomes are not merely the result of a rational matching of
national interests and solutions, but are better seen as the product of internal conflict, compromise,
and negotiation among competing ministries. Thus, the importance of having a meritocratic
bureaucracy that is highly competent and has technical expertise to promote programs aimed at
industrial upgrading.
Beyond the principal-agent relationship amongst non-elected and elected government officials,
specific type of policies designed and adopted will also depend on the interactions between
bureaucrats and business, as will be discussed in the next section.
22
Non-elected State Actors II: Ministers
Role. Ministers from areas such as Economy, Industrial Planning or Science and Technology, can
play a role in setting up the agenda for industrial policy, bringing their connections with business
due to prior positions and backgrounds, as well as their technical expertise (Breznitz 2007; Evans
1995; Kang 2002; Schneider and Maxfield 1997). “Cabinet ministers play key roles in every stage
of policymaking. Together with the bureaucracies they head, ministers nearly have a monopoly on
the design of policy, with occasional input from political parties and/ or interest groups”
(Scartascini, Stein, and Tommasi 2010, 11).
Preferences. Ministers are not subject to election, thus do not have to be concerned about the
electorate support or campaign donors. Since the revolving door is a recruitment mechanism at top
levels ministers might bring in their own agenda favoring the status quo or change. This will
depend on their backgrounds and prior experiences. Nonetheless there are limits to this autonomy:
since they are appointed by the president they might have to be in line with the president’s goals.
Such independence/autonomy of ministers also depends on whether the president had the majority
in elections, or if he is trying to secure other parties’ support (ruling coalition). When he appoints
ministers from a different party we may see ministers with an agenda that may not coincide with
the president (Saeigh 2010).
2.1.2. Business Actors
On the business side, there are the individual firms which can vary by size (small, medium
or big) or capital ownership (foreign vs. domestic). And a related yet different actor are business
associations, where firms organize collectively. Associations generate their agendas and resources
to different degrees.
Firms
Role. Individual firms can play a passive or active role depending on several factors. A main reason
for being active is when firms face a pressing challenge or intensive competition like liberalization
(can be trade or FDI with the arrival of foreign products and firms). Another reason is when they
are aware and have examples of successful programs abroad (like that in India or Ireland).
Preferences. Firms’ main objective is to be profitable and have certainty (low risk). Growth and
expansion may be part of their strategy in securing its objectives. They will welcome programs
that are aimed for their protection, grant them incentives or funding. The main beneficiaries of
industrial policy are business (entrepreneurs) of the sector towards which government decides to
23
support. Therefore, there are incentives for firms to press for the adoption of policies and programs
favorable to their growth (Doner 1992).
Business Associations
Role. Business associations are a key actor (Doner and Schneider 2000; Schneider 2010; Schneider
and Maxfield 1997). They bring together firms in their attempt to foster coordination and
participate in policymaking in various ways. Most organize events in which they invite public
officials to meet their members and share information. Their leaders use media to voice their
concerns or announce their policy positions. Often, governments invite association representatives
to participate in policy councils. And some associations even have research departments that gather
data on sector performance and generate policy proposals (Schneider 2010, 223–24).
Firms may choose to organize and act collectively. Traditionally, it has been theorized that
firms will form business organizations in sectors where there is a small number of players (Olson
1965). However, other authors have shown that firms may attempt to organize even if there is a
large number of players (market fragmentation) to try to gain some of the selective benefits that
the state offers to them such as privileged access to policy making or funding (Schneider 2004).
Another reason firms will attempt to create or belong to an association is when individual firms
have not sufficiently developed their political relations to have enough power to influence policy
and their size or market share is insufficient to hold a fair amount of structural power.
Preferences. Business associations will look for public support for the sector. Support can come
in different modes: funding, training programs, creation of common standards, land or facilities.
During the policy adoption and design stages, they will want to participate through consultations,
and during the implementation phase, they may aim at being part of the assessment boards or even
play a role in the distribution of benefits.
2.2. Models of Business Government Interactions
The following section addresses the question of what types of business-state interactions
occur in industrial upgrading policymaking? I argue that there four broad patterns of business-
government relations: synergistic, state-led, business-led and quiescent. And two characteristics
of the main actors, bureaucrats and business generate these various models of business-government
interactions: level of bureaucratic quality and degree of business cohesion.
14
14
Scholars studying different policy realms, like social welfare schemes (Martin and Swank 2012); economic
liberalization (Schneider 2004, Etchemendy 2011); and taxation (Weyland 2000; Fairfield 2014; Castaneda 2015)
24
2.2.1. Level of Bureaucratic Quality
Bureaucratic quality is the extent to which government agencies possess Weberian standards of:
meritocratic recruitment, predictable careers or ladders, competitive wages, and technical expertise
(Dahlström, Lapuente, and Teorell 2012; Evans 1995; Evans and Rauch 1999). It ranges from high
to low. High quality bureaucrats possess relevant professional degrees; acquire technical
knowledge over time; expect to develop a long-term career and climb the organizational ladder;
and receive competitive wages. Low quality bureaucrats lack adequate studies for their positions;
are selected through nepotism; and usually have short term expectations, as soon as the politician
who appointed them finishes his period, they are replaced. Prior empirical studies have shown that
bureaucratic quality impacts economic policies and their outcomes (Amsden 2001; Evans 1995;
Johnson 1982; Kohli 2004; Rauch and Evans 1999; Wade 1990). Also where there is meritocratic
recruitment, bureaucrats will be less interested in bribes offered by business (Dahlström, Lapuente,
and Teorell 2012).
Bureaucratic quality shapes policy outcomes because depending on the level of
bureaucratic quality, bureaucrats will pursue different strategies when attempting to adopt, design
and implement a policy. The next section provides further details.
2.2.2. Level of Business Cohesion
Business cohesion is the degree to which firms in the same sector coordinate their policy
demands and their actions. It ranges from high to low. When there are high levels of business
cohesion, firms are organized into a sectoral association which: represents all types of firms; is
well organized and staffed, and uses its institutional strength for productive ends (Doner and
Schneider 2000, 263).
15
The association may include domestic and foreign companies, although
legislation in some developing countries bans foreign-owned firms from participating. Several
advantages result from a high level of articulation.
16
First, associations aggregate information that
have proposed a to focus on business and government interactions- to explain program variation in relatively similar
sets of countries. Fairfield (2015) found that cohesion was the most important business source of power explaining
variation in tax policies in a set of Latin American cases.
15
According to Doner and Schneider (2000), associations’ institutional strength depends in turn on internal factors:
high member density, valuable selective benefits (often delegated by governments), and effective internal mediation
of member interests and disputes. Also, external factors, especially competitive markets and government pressure.
16
Predominant theoretical treatments view business associations as rent-seeking and special interest groups (Olson
1965, 1982; Persson and Tabellini 2000). Yet, empirical research in a wide range of developing countries reveals a
broad range of functions and activities undertaken by business associations, many of which actually promote
efficiency and development (Cali and Sen 2011; Doner and Schneider 2000; Haggard, Maxfield, and Schneider 1997;
Nugent and Sukiassyan 2009; Özel 2015; Weyland 1997).
25
is useful for decision making. Also, they channel firm demands’ in an organized and coordinated
way, reducing conflict and limiting narrow interests. When government seeks to cooperate with
firms in a sector “cooperation is further facilitated by intra-organizational cohesion. Cohesive
organizations can make credible commitments because of their ability to control shirking among
their members” (Weyland 1997, 53).
In contrast, when there are low levels of business cohesion a sectoral association is missing
or weak, so each firm lobbies for its interests through its own means. Big businesses are in an
advantageous position because they have more resources to staff government-relations
departments; pay for lobbying firms and/or fund campaign elections. Meanwhile, small and
medium sized firms may face a hard time voicing their concerns unless they possess strategic ties
to politicians or bureaucrats.
2.2.3. Four Models of Business-Government Interactions
The two dimensions specified above (bureaucratic quality and business cohesion) generate
four broad models of business-government interactions.
17
These in turn lead to different policy
outcomes. It is important to have in mind that the relationship between business and government
is dynamic and some countries/sectors may manage to move from one category to another over
time. In this section I outline the main characteristics of each model. And in the following section,
the mechanisms and outcomes are specified in more detail.
I hypothesize that the combination of bureaucratic quality and business cohesion generates
four broad patterns of business-government interactions (See Figure 2). In turn, I argue that these
varied patterns lead to different policy processes and thus outcomes.
17
Kohli (2004) argues there are three types of state and part of his classification is based on how the state relates to
business. The cohesive capitalist would be close to the developmental quadrant because the state is so powerful that
it is able to lead private elites into its overarching goal of economic growth. The quiescent quadrant is similar to the
neopatrimonial state, because he mentions that weak states encounter weak private sectors and usually pursue
personalistic goals. Kang (2002) examines the power or strength of state and business. The strength of state depends
on the degree of coherence and fracture within the political leadership, and business strength is determined by the
number of firms in the sector (concentration). The main argument is that the relative strength of the state and the
business sector determines the form and level of money politics, which in turn has an impact on a country´s
development trajectory. State-business relationships can range from mutual hostages (when both actors are strong),
rent seeking, when there is a fractured state and concentrated business; predatory state if the state is coherent and
business is dispersed, and laissez-faire if state is fractured and business is dispersed. Furthermore, the author, in line
with the public choice school of thought, assumes a negative or rent seeking behavior from government intervention
when the state is fractured. Additionally, he implies that coherent states are only possible under authoritarian regimes.
26
Figure 2. Patterns of State-Business Interactions
Bureaucratic Quality
Low High
High Business-led
Process
Synergistic
Process
Business
Cohesion
Low Quiescent
Process
State-led
Process
A synergistic process emerges when both bureaucratic quality and business cohesion are
high. High quality policymakers are in a position of technical expertise and negotiation skills that
enables them to collect and assess information from several sources, generate long term
development plans and work closely with business without being captured by firms’ interests
(Etchemendy 2011; Evans 1995; Fairfield 2015). Business articulation is also relevant for policy
adoption and design. High business cohesion allows firms in a sector to better aggregate
information, set clear goals and negotiate with government as a block. Efforts to affect government
indirectly through associations are likely to be less distorting and narrowly rent seeking than those
made directly by firms in the form of unofficial payments to officials (Nugent and Sukiassyan
2009, 425). Once policy is approved, the government may even grant the associations the
responsibility of distributing the benefits derived from programs (Doner and Schneider 2000).
Theoretically, this is one of the best scenarios for industrial upgrading because both business and
states will work towards a common goal from a position where there is relevant information
available and management capabilities. Special interests are restrained. Adoption of programs is
more likely.
The business-led process is characterized by low levels of bureaucratic quality but high
levels of business cohesion. This results in the lack of formal dialogue and gives way to
uncoordinated actions from firms and bureaucrats. Since firms are well organized they might be
able to impose their views which may or may not be beneficial for the development of the sector.
27
In this context, efforts for industrial upgrading might be led by business associations but without
state support their reach will be limited.
The state-led process, is one where the quality of government is high but business cohesion
remains low, so the state is better able to impose its will without too much interference from the
firms. Since bureaucrats are selected through meritocratic mechanisms they have the skills and
capacity to extract the information needed to build their strategies. And they can also devise ways
to prevent capture from business. In a state-led setting, bureaucrats have to rely on their technical
expertise without having to negotiate with sectoral associations. They can also choose to generate
consultation mechanisms with a group of firms they may deem relevant. Prior studies revealed that
often, it is the most profitable firms (or those with largest market penetration) that are considered
(Evans 1995; Kang 2002). Governments have sometimes found it easier to first generate/force
business into associations and then make them their official counterparts in the policy making
process but this takes a longer time (Martin and Swank 2012; Schneider 2004; Wade 1990).
On the other extreme is the quiescent process where state-business interactions are
characterized by low levels of bureaucratic quality and low levels of business cohesion. On the
government side, low levels of bureaucratic quality generate an environment where elected-
government actors, impose their will more easily over bureaucrats, so they may push programs
that benefit their own interests/constituencies instead of trying to promote industrial upgrading.
And in the private sector, there are considerable asymmetries in firms’ access to policymakers.
Large firms often have political affairs departments as well as the resources to hire lobbying firms
and/or fund elections. These resources grant them easier access and more influence over
government than most of the small and medium sized firms. Furthermore, with no business
association in place, firms provide partial information to bureaucrats who must collect and
integrate it. But since bureaucrats have limited skills to assess the proposals, the results are most
likely to be either biased or limited.
After outlining the types of business-government interactions, the next question that arises
is how do the interactions and outcomes (types of policies adopted) vary? The various
configurations favor the use of certain strategies/actions (causal mechanisms) both from the
28
business and the bureaucrats which in turn generate different types of outcomes.
18
The next section
elaborates on the strategies of these two actors.
2.3. Causal Mechanisms: Pathways to Industrial Upgrading Policies
What are the specific causal mechanisms underlying business cohesiveness and
bureaucratic quality effects on the types of industrial policies adopted? That is, what strategies and
actions do actors from the state and the private sector take to promote the adoption and design of
programs for upgrading?
Figure 3. Pathways to Industrial Upgrading Policy
2.3.1. Bureaucrats’ Strategies
The strategies/causal mechanisms through which bureaucrats shape policy are:
1. Creation of formal channels of collaboration: entails interacting with business
organizations through councils or boards, usually meeting regularly in a predefined venue.
2. Self-acquired expertise on the subject: this is achieved through prior professional
training and experience. Also, when they seek technical advice from universities or
consultants.
18
To illustrate the models, we can take two case studies conducted by Cammett (2007). She describes that Tunisian
industrialists avoided collective lobbying efforts, instead focusing on firm-based upgrading or exit strategies, and
conveyed policy preferences largely through informal channels. As a result, the state dominated system of economic
policymaking in which firms were by and large, policy takers and state preemptively forged industrial policy, remained
relatively stable in Tunisia. In contrast, Moroccan producers organized powerful collective lobbying efforts through
producer associations and increasingly expressed policy goals through public channels such as the media and regularly
scheduled official business-government meetings. Collective action brought about shifts in modes of Moroccan
business politics in the 1990s. new forms of business representation and business-government relations permitted
expanded access to economic opportunities for a larger segment of industrial capital holders. At the same time, formal
business associations became important sites for business mobilization” (Cammett 2007, 6). The first scenario would
be an example of state-led setting and the second one of synergistic.
Causal mechanisms:
Strategies used by bureaucrats
and firms/associations
Patterns of Business-
State interaction in
policy making
Types of
industrial
upgrading
policies
29
3. Using the revolving door: in which the government hires bureaucrats coming from the
business sector bring with them technical knowledge and personal connections.
4. Informal interactions: with business organizations and/or specific firms (usually big
firms)
The level of bureaucratic quality impacts policy making through the selection of strategies/
causal mechanisms that government officials use to shape policy and the final outcomes. If the
bureaucrats have high quality then it is most likely that they will use formal consultations in which
they gather information, plan and assess the feasibility of the programs. Furthermore, meritocratic
bureaucrats will acquire or already possess highly technical knowledge that can be used to gauge
the different policy alternatives. This knowledge can also serve as a balance against business
information advantage (Culpepper 2011).
19
If the level of bureaucratic quality is low, public servants may choose to work with business but
mostly taking a passive or reactive position because business will have better information and use
its lobbying skills. They will opt for informal contacts rather than well-established mechanisms.
Or bureaucrats may also choose not to interact with business at all and make no attempts to adopt
a policy.
2.3.2. Business’ Strategies
Firms have a range of options through which they can influence government policies
(Schneider 2004, 2010). And businesses use these strategies depending on the situation and
position from which they are bargaining. These different strategies business implement are the
causal mechanisms that lead to particular outcomes. Their preference is for industrial policy but
the type of policy they will obtain is shaped by their interactions with government.
Schneider (2010, 221) proposes a framework where business have a portfolio of investment
in politics: “Businesspeople can invest in a range of political activities from business associations,
to financing parties and candidates to networking with government officials, to outright bribery.
In principle, rational business politics should balance their portfolio…where business concentrates
its political investment is largely a function of the perceived opportunities for influence offered by
the political system.”
19
Culpepper (2011) contends that businesses are capable of imposing their will in detriment of other options if there
is a large informational asymmetry between firms and bureaucrats.
30
The mechanisms that firms have to press for their interests are (Culpepper 2011; Fairfield
2015; Schneider 2004, 2009):
1. Direct access to policy makers: firms may have to high rank officials and might obtain
privileges through their political connections. It can be formal or informal
2. Lobbying: firms can choose to advance their interests by meeting the congress or with the
agencies in charge of promoting policy adoption and trying to convince them. Often this
entails hiring specialized government relations/public relations agencies to make their
message more convincing and professional.
3. Campaign finance: it may be possible that rather than working with bureaucrats, business
supports politicians and expect them to represent their interests during the policy making
process. Particularly if the congress has a crucial role in the adoption or implementation of
policy (see Boas et. al. 2014).
4. Revolving door: that is having former businessmen getting into politics and technocratic
positions so they have a more direct voice. And/or assuming office (direct power), specially
if there is no rule of law then they will have more leeway to implement the policies they
want.
5. Election into public positions: businessmen can also decide to enter the electoral race to
have a direct representation of their interests
6. Media: use of media to generate wider public interest or to put an issue in the agenda of
politicians whose support is needed for the policy approval.
7. Creating/Participating in a business association: firms may decide to join efforts and
overcome collective action.
8. Exiting (structural power): asset mobility, business can threat to move their assets if they
do not obtain state support. But this strategy is available only to certain types of industries
or with particular companies.
Business associations can resort to these same mechanisms. And we can expect that
business cohesion amplifies the effectiveness of the strategies chosen. Of the strategies available:
direct access to policy makers, lobbying, the revolving door and media, seem to be the preferred
actions. High cohesiveness positions business in a stronger bargaining position (Frieden and Stein
2000). In fact, studies have found that business that were better organized received more protection
from the states implementing structural reforms like steel companies in Argentina (Etchemendy
31
2011). And business cohesion derived in more favorable tax policies for the private sector in Chile
(Fairfield 2015).
20
There is also use of the revolving door. It occurs when prior business associations leaders
are invited to join the government ranks and make sector-wide declarations that attract other actors
(i.e. general public, ministers etc.) attention or support. Whereas campaign finance is not
frequently used by associations in Latin America (Schneider 2010) nor is the election into public
positions.
Fairfield (2015) argues that Schneider is confusing sources with power. Business have
structural and instrumental power. Structural power stems from the disinvestment threat. Sources
of instrumental power generate mechanisms. Thus, cohesion is a source of power and enhances
the effectiveness of lobbying or other forms of mobilization through two mechanisms: 1.
Strengthens economic elite’s bargaining position by increasing the cost of divide and conquer
strategies. If cohesion is weak policymakers may offer marginal concessions; and 2. Confers
legitimacy on economic elites’ demands.
Next section provides further detail into the types of industrial policies that are the outcome
of the different business-government interactions.
2.4. Types of Industrial Policies
Four types of policies can be identified according to their target population and allocation
criteria: particularistic, narrow, competitive and inclusive.
21
Some of these types of policies may
coexist in one same country so ideally, an index might capture the outcomes of business-
government interactions better. Table 1 summarizes the types of policies and their characteristics.
20
These empirical findings contradict skeptical views about the usefulness of sectoral associations in Latin America:
“The nature of political economies in Latin America also suggests that individual firms should demand input into the
policy process and be willing to pay for it. Latin American economies are dominated by multinational corporations
and highly diversified domestic conglomerates that are often controlled by single families. Sectoral business
associations tend to provide a weak basis for political influence because subsidiaries in a given sector have little
autonomy to coordinate with other firms and pursue collective policy preferences. Economy-wide business
associations are strong in a few countries but weak in many others, including Brazil” Schneider (2004, 2013) in (Boas,
Hidalgo, and Richardson 2014)
21
Shrank and Kurtz (2005) classify industrial policies as open economy and closed economy (import substitution)
using tariff protection. But countries are no longer using import substitution. And while exporters are an important
target population, in the case of IT policies also target players catering the domestic market. Other authors have
classified industrial policy by distinguishing the different instruments and domains of intervention(Cimoli, Dosi, and
Stiglitz 2009). Yet others have distinguished measures into two: functional (horizontal) open to all firms or sectors,
and selective (vertical) only for particular kinds of firms (Weiss 2010).
32
In particularistic policies only certain firms get state support, being funding, strategic
information, access to export advice. Firm eligibility is based on whether or not there are close
connections with bureaucrats (and politicians). The allocation criteria are non-competitive because
policies deployed are not subject to any sort of filter or assessment by a neutral or representative
board or council.
Narrow policies differ from particularistic because bureaucrats are able to impose certain
selection criteria, that may even be used as performance standards. For instance, the policy can
specify that only exporting firms are eligible to funding. Thus, bureaucrats will allocate public
support to firms that meet that criteria. And the participation in the program might be tied to
meeting particular goals or objectives. For instance, increasing firm exports by 10%.
Table 1. Varieties of Industrial Policies
Particularistic Narrow Competitive Inclusive
Target
population
One or few
firms (Usually
big companies)
Limited
(Only
certain
types of
firms
Comprehensive
(All firms are
eligible)
Comprehensive
(All firms
without
discrimination)
Allocation
criteria
Non-
competitive
Access only
through
personal/firm
connections
Non-
Competitive
but
restrictive
Firms have
to meet
certain
profile/
criteria (i.e.
exporting,
SMEs,
R&D) but
are not
subject to a
committee
assessment.
Competitive
All firms are
eligible. Firms
need to apply
following
specific
guidelines and a
committee/board
assesses and
awards the
resources.
Non-
competitive
Granted to
every firm in
the sector
Example Government
contract without
auction.
Export-
support
funds
R&D or training
funds
Industry wide
tax exemptions
33
Competitive policies are open to all firms in the sector, nevertheless the allocation of
support requires that firms meet specific deadlines, requisites (like presenting a business plan), and
all the candidates are subjected to a review process. This review process may rely on bureaucrats
but there may be some ad hoc committees with joint public-private participation.
Finally, inclusive policies are open to all firms in the sector and the allocation is not competitive.
Instead, all firms in the sector can benefit just by demonstrating they are located in the particular
sector targeted. An example of this is when government offers industry-wide tax exemptions.
A central argument of this chapter is that the configurations of business-government
interaction lead to the selection of different actions resulting in a variety of policies adopted. In
what follows I present four general propositions that will guide the analysis of the case studies in
the empirical section.
Proposition 1: Synergistic Relations Mechanisms and Outcomes
In a synergistic setting the mechanisms through which bureaucrats generate policy are most
likely formal. Since bureaucratic quality is high there are incentives to promote formal
consultations with business representatives rather than indirect or informal contacts. Bureaucrats
will try to extract information from business to better tailor the policies in a regular, ordered
fashion. This can be achieved through the creation of a board or council. Or holding frequent
meetings, i.e. monthly meetings.
Business having high levels of cohesion will probably opt to lobby through associations.
This facilitates information exchange with bureaucrats and ministers because rather than
contacting one by one firm or only the largest ones, associations collect and integrate the needs.
Interactions through associations, also levels the playing field for firms of different sizes. Finally,
associations can also resort to media, and make sector-wide declarations which will have a stronger
impact than a single, or isolated declarations of a few firms.
The outcome of business-government interactions might be the adoption of industrial
upgrading inclusive or competitive policies.
34
Figure 4. Synergistic Pathway
Proposition 2: Business-led Setting Mechanisms and Outcomes
When business has a high level of cohesion but bureaucratic quality is low there is an
asymmetric relation, so we can expect to see a sectoral association trying to find the best
interlocutor to voice/negotiate. Associations can attempt to lobby congress and the executive to
draw attention on their agenda. And also resort to informal channels such as personal contacts with
bureaucrats to try to involve at least some of them within the relevant ministries to support their
efforts. Low quality bureaucrats will not attempt to create formal channels of consultation and
since they have little to no expertise, they will rely on business proposals instead of generating
their own policy positions.
The policy outcome will most likely be biased towards a dominant or easy to target part
of the business sector since the executive and congress have so many constituencies to cater and
the bureaucrats will not have enough knowledge as to try to spread the program benefits to all
firms. They will opt for a narrow or a particularistic policy.
Figure 5. Business-led Pathway
Bureaucrats:
Formal consultations
Business:
Lobbying through associations
Synergistic Inclusive or
competitive policy
Bureaucrats: no formal consultation or
assessment, rely on business proposals
Business: lobby through associations,
formal and informal channels
Business-led
Narrow
or particularistic
35
Proposition 3: State-led Setting Mechanisms and Outcomes
In a state-led setting, bureaucrats have high quality but the business sector has low
cohesion. Thus, bureaucrats resort to their technical expertise and do not consult with sectoral
associations. They can also choose to generate consultation mechanisms with a group of firms they
may deem relevant. Prior studies revealed that often, it is the most profitable firms (or those with
largest market penetration) that are considered (Evans 1995; Kang 2002). Governments have
sometimes found it easier to first generate/force business into associations and then make them
their official counterparts in the policy making process but this takes a longer time (Martin and
Swank 2012; Schneider 2004; Wade 1990).
Figure 6. State-led Pathway
A possible consequence of low levels of business cohesion is that “policymakers may be
able to negotiate acceptance from particular sectors or subgroups by offering only marginal
concessions” (Fairfield 2015, 38). To some extent, this can be positive if for instance, the
government is able to impose selection criteria that makes the management of programs more
transparent. Another policy option is the narrow policy because it is more focused and also
provides reciprocal control mechanisms (see Amsden 2001). For example, choosing to promote
firms who export their products, rather than those catering domestic markets.
Proposition 4: Quiescent
When business cohesion is low and bureaucratic quality is also low, individual firms will
look for alternative channels of action. For instance lobbying the congress or the executive branch,
and through electoral contributions.
22
(According to Schneider, business associations by custom or
22
“Despite the flow of millions of dollars into electoral campaigns, there are several reasons to doubt that this flow
buys business a great deal of leverage in most policy processes. The first is a common collective action problem:
businesses contribute to individual legislators, who do not have much impact on policy except in collective votes,
Bureaucrats: self-acquired
expertise, revolving door,
councils
Business: participate in
consultations but individually
State-led
Competitive or
narrow policy
36
legal restrictions do not contribute to political campaigns in Latin America, and their influence
with legislators is likely to be less than that of major contributors, who tend to come from
individual firm (2010, 226).
There may be considerable asymmetries in access to policymakers because large firms with
political affairs departments or with the resources to hire lobbying firms will have an easier access
and more influence on government than small and medium sized firms. Also, because there is no
business association in place, firms provide partial information to bureaucrats who must collect
and integrate it.
The resulting policy, if any, will most likely be particularistic, that is it will only benefit to
firms who have preferential access to policy makers through their own lobbying efforts or contacts.
Figure 7. Quiescent Pathway
Table 2 synthesizes the patterns of business government interactions, the hypothesized causal
mechanisms and their most likely policy outcomes.
while it is nearly impossible for business contributors to coordinate to exercise collective pressure…Another problem
for contributors is that turnover is very high in most legislatures (100 percent in Mexico, by law) so that incumbents,
once in office, have weak incentives to heed their contributors” (Schneider 2010, 227–28).
Bureaucrats: rely on
information from specific firms
and contacts
Business: personal contacts and
firm lobbying
Quiescent
Particularistic or no
policy
37
Table 2. Summary of Models of Business Government Interactions
Bureaucratic quality
Low High
Business
cohesion
High Business-led
Causal mechanisms:
Business: lobbying
through an association
but also uses alternative
strategies such as direct
contacts
Bureaucrat side: no
formal consultation or
assessment, rely on
business proposals
Outcome: Narrow or
particularistic policy
Synergistic
Causal mechanisms:
Business: lobbying
through organization
Bureaucrat side:
Formal
collaboration/consult
ation that includes
variety of actors
Outcome: Inclusive
or competitive policy
Low Quiescent
Causal mechanisms:
Business side: firm
lobbying, personal
contacts, campaign
funding
Bureaucrat side:
selective consultation
Outcome:
Particularistic or no
policy
State-led
Causal mechanisms:
Business side: Firm
lobbying and
resorting to personal
contacts
Bureaucrat side: self-
acquired expertise,
selective and
informal consultation
Outcome:
Competitive or
narrow policy
38
3. Conclusions
This chapter has presented an analytical framework to guide the analysis of industrial
upgrading policies in democratic developing countries. It has first reviewed extant studies from
the developmental state and business politics literature and has brought their most valuable ideas
together into an analytical framework to explain industrial upgrading policy processes in the digital
era.
The framework goes beyond current explanations because it considers both actors: states
and businesses, and their interactions. Most of prior studies analyzing industrial transformation
have either carefully examined and theorized about the state characteristics (developmental state);
or on firm’s structures and their capacity to extract the policy that is most favorable to their
interests (business politics). The framework also contains a dynamic element, because it can be
used to understand the evolution of a single unit of analysis over time. Countries and regions can
actually move from one quadrant to the other depending on the conditions and strategies
implemented by policymakers. The general propositions presented here will be tested and further
refined through a mixed methods strategy in the following chapters (3, 4 and 5).
39
Chapter 3.
The Adoption of Policies Towards the Software Sector:
A Cross-Country Quantitative Test, Latin America (1995-2012)
If the 20
th
century challenge for developing countries was industrialization and the
consolidation of a manufacturing sector, the 21
st
century calls for the promotion of a knowledge
intensive economy. Still, developing countries seem to be having a hard time to upgrade their
industries and foster new activities. In fact, during the past decade many of them have even
experienced a deepening reliance on their primary sector caused mainly by the commodity-export
boom induced by China’s insertion into the global economy (Frenkel and Rapetti 2011). For
instance, in the Latin American region, just 10 out of 17 countries had implemented a sectoral
software strategy by 2012.
23
So it is puzzling that even if bureaucrats and businessmen are aware
of the need to encourage innovation-intensive activities, their efforts are sluggish at best.
In this context, this chapter addresses the following question: Why do some countries
adopt industrial upgrading policies and others do not? And more specifically: Do states and
business characteristics influence the prospects of industrial upgrading policy adoption? I
argue that government and business sector characteristics are fundamental and shape the adoption
of industrial upgrading policies
24
. My two hypotheses are: 1) Countries with bureaucracies
insulated from political pressures and with a meritocratic recruitment mechanism, are more likely
to set up programs to stimulate high technology sectors than the rest. And 2) Countries with an IT
23
Brazil and Costa Rica were first movers enacting an official strategy towards the Software sector in 1997 and 1999
respectively. Then during 2000-2005 Uruguay, Mexico, Venezuela and Argentina enacted their own initiatives. In the
period of 2005-2010 Peru, El Salvador and Colombia also adopted programs and Nicaragua has promoted its IT
services exports since 2012. See Appendix 1 for details.
24
Industrial upgrading policies are defined here as those policies whose objective is the creation or consolidation of
high-technology sectors in which knowledge plays a critical role, for instance software, biotechnology and aerospace.
Traditionally, the main tools used by governments have been fiscal incentives, trade protection, promotion of direct
investments (state or foreign-owned), and credits financed by national development banks. But in recent years new
strategies have emerged which involve closer business-government coordination such as fostering clusters,
stimulating innovation and coordinating collective action (O’Riain 2004; Peres 2009). Moreover, industrial upgrading
policies are layered, which means that they are created and implemented at multiple levels (national, regional and
even local level).
40
sectoral business association are more likely to adopt industrial upgrading policies than countries
lacking an IT sectoral business association.
In addressing these questions this chapter offers a first quantitative cross-country test
through an event history analysis of the determinants of policies to stimulate the software sector
in 17 democratic Latin American countries from 1995-2012. The software sector is strategic to
developing countries because software can be used across a wide variety of industries to improve
productivity, and generate much needed employment for the youth.
The results of the analytical effort presented in the following pages, suggest that not all
states and all business sectors are created equal and some may actually foster efforts to transition
into knowledge intensive activities. This challenges, once again but now in the digital era, the well-
entrenched view against both, state intervention in the economy and close business-government
collaboration in developing countries, which to many scholars and policymakers only leads to rent-
seeking (Bates 1981; Krueger 1974, 1990). Under certain conditions, bureaucracies and business
leaders, and not just free markets, may be the drivers of industrial upgrading.
This study must be taken as a first building block of a broader exploration of business-
government collaboration in industrial upgrading policy processes that seeks to answer the
following more general questions: under what conditions do business-government collaboration
for industrial upgrading emerge? And what are the outcomes of such business-government
collaboration? The statistical analysis in this chapter is complemented by an in-depth case study
and subnational comparative cases in chapters 4 and 5.
The next section presents some conceptual definitions and hypotheses. Section 2 discusses
the empirical strategy; and alternative explanations are considered when introducing the control
variables. Finally, section 3 presents the results and section 4 the conclusions.
1. Concepts and Hypotheses
Unlike other economic policy realms, such as trade and financial liberalization, where
there is a comprehensive understanding of the factors influencing policy makers’ decisions; and
several cross-country quantitative empirical tests have been conducted both for developed and
developing countries; the determinants of industrial policy adoption are much less established.
Still it is possible to draw from the broader literature about the political economy of economic
development as a starting point to derive testable hypotheses and further refine extant explanations.
41
My general argument is that two factors shape the prospects of industrial upgrading: the quality of
the bureaucracy and the business sector’ degree of organization. And we should consider them
both if we strive for a full understanding of the proximate causes of development.
1.1. The Quality of State Bureaucracy
The state has been identified as critical for industrial upgrading processes since several
decades ago (Gerschenkron 1962; Prebisch 1963). The scale of the task and the resources required
in the effort of transforming an economy, demand that the state actively engages in this endeavor.
One of the features that has been widely studied by developmental state scholars is the role of
bureaucrats and executive agencies.
A country’s adoption of industrial upgrading policies (and their success) is contingent on
having a meritocratic bureaucracy (Amsden 2001; Evans 1995; Johnson 1982; Wade 1990). When
bureaucrats are recruited through highly selective mechanisms and have long-term horizons, they
are able to overcome special interests. This type of government can achieve a close collaboration
with business but keeping enough distance to prevent capture, the so-called embedded autonomy
(Evans 1995).
With embedded autonomy, policy makers are in an improved position to extract
information about the needs of firms and devise long term plans to move the economy into higher
technology activities through the adoption of a set of industrial policies. Moreover, meritocratic
bureaucrats can establish control mechanisms, understood as a set of governmental institutions
that impose discipline on economic behavior. In prior successful cases, subsidies were allocated
to make manufacturing profitable but did not become giveaways. Recipients of subsidies were
subject to performance standards. The reciprocal control mechanism minimized government
failure (Amsden 2001, 8–9).
Finally, bureaucrats that have been long enough in a specific area, such as a Ministry of
Economic Development or a Science and Technology Ministry, accumulate technical knowledge
or specialization which allows them to better gauge business pleas and devise more adequate
incentive structures. They may also have opportunities to interact and engage in best practices
exchanges through regional and international institutions, and even travel to areas that are
considered as success stories that serve as reference points.
42
Therefore, there are reasons to believe that the quality of public servants a country has,
will impact the probability of industrial upgrading policies, and later on the design and
implementation stages.
Hypothesis 1: Countries with meritocratic bureaucracies are more likely to adopt industrial
upgrading policies than the rest.
Yet the validity of the developmental state emphasis on meritocratic bureaucracies has
faced three critiques from within institutionalist scholars: democratization, decentralization and
globalization (Breznitz 2007; Kang 2002; O’Riain 2004; Öniş 1991; Yeung 2014). The
democratization critique relates to the fact that most of the empirical support for the developmental
state theory comes from East Asian countries that had authoritarian or semi-authoritarian
governments (Onis 1991, 121). Whereas today many developing countries are attempting
business-government collaboration in somewhat more democratic countries. So politicians and
bureaucrats, are subject to pressures from business that use different channels to pursue their
interests such as lobbying, campaign contributions and media access (Schneider 2015).
25
The decentralization experienced by countries has generated new agencies where
technocrats
26
, not meritocratic bureaucrats, are in charge of decision making (Breznitz 2007;
O’Riain 2004)
27
. Thus, meritocratic bureaucracy may be a necessary but not a sufficient condition
25
As Schneider (2015, 61) notes: “The differences in political context are crucial to bear in mind when assessing the
relevance for Latin American experiences outside the region of industrial policy making and business-government
collaboration. There is a historical correlation between successful developmental states and authoritarian rule and
more specifically between performance standards imposed on firms and authoritarianism (Amsden 2001; Schrank
2013a). Even in more democratic contexts, developmental states in Asia counted on powerful insulated bureaucracies
with strong backing from ruling parties. These political and bureaucratic conditions of course do not exist in
contemporary Latin America”.
26
Technocrats are defined by Dargent (2014) following Collier’s definition as: “Individuals with a high level of
specialized academic training which serves as a principal criterion on the basis of which they are selected to occupy
key decision-making or advisory roles in large, complex organizations – both public and private” (Collier 1979, 403).
And these individuals are usually appointed by politicians to deal with complex issues that politicos lack the
knowledge to successfully address.
27
O’Riain (2004, 30) asserts that: “Embedded autonomy in the flexible developmental state is not guaranteed by a
coherent bureaucracy but by the flexibility of the state structure. The decentralization of state agencies enables them
to become deeply embedded in their clients/ constituencies, despite the fact that they are often dealing with a wide
range of individuals and organizations across widely dispersed networks. The agencies also retain a certain autonomy
despite their close relations to their constituencies as there tends to be relatively regular change in the goals,
composition and even existence of the agencies as development needs change… They remain potentially accountable
to national goals and bureaucratic norms and processes continue to provide valuable standards of accountability and
professionalism”.
43
for industrial upgrading policy adoption. In fact, technocrats, are prominent actors in some
developing countries. Dargent (2014), in his study of Latin America economic policy makers,
demonstrated that technocrats have autonomy vis-à-vis business.
28
These ideas about the role of
technocrats in industrial upgrading processes merit further analysis; unfortunately there are no
readily available measures to conduct a cross country study at this time. One could test whether
having technocrats in the top positions of agencies such as the Ministry of Economy, Ministry of
Science and Technology and/or the Ministry of Information Technologies, results in an increased
likelihood of industrial upgrading policy adoption.
The globalization processes that took place in the last two decades of the 20th century
transformed the scenario in which business-government collaboration takes place. One of the
weaknesses of prior developmental state work is that it understudied the business-side of such
collaboration. Today, it is crucial to understand the different configurations that the private sector
may have. For instance, multinational companies (MNCs) play a key role in several industries,
including IT, and they usually have the power to “exit” (Hirschman 1970) if they disagree with a
state’s policies so to a certain extent they can be considered as a binding force for government.
29
MNCs are also at the core of global production networks which generate a different dynamic
between business and government and even between domestic and foreign firms (O’Riain 2004;
Schneider 2013a; Yeung 2014).
Overall the presence of MNCs has been considered as an obstacle of business-government
collaboration because these firms receive a different treatment from government and they rarely
make linkages with domestic companies (Paus and Gallagher 2008). Internationalization of
domestic firms also raises new challenges for collaboration with their home and host states.
28
“Technocrats in Latin America, especially in the economic ministries possess more autonomy than it is often
assumed.… technocrats have sometimes used their power to transform the structure of an economy against the wishes
of dominate business groups, as happened in Colombia and Peru during recent decades” (Dargent 2014, 162).
29
The flexible developmental state is based on a double embeddedness in foreign capital and local, particularly
professional, networks of innovation (O Riain 2004). In the Irish case “These agencies were embedded in the emerging
technical professional class through personal and associational networks. Key individuals pursued careers which
spanned private, public and semi-public organizations, making the boundaries between these different spheres porous
and fostering increased interaction across those organizational boundaries. This in turn improved the agencies’
capacity to make and implement effective policy. The first National Software Director, Barry Murphy, had been the
Managing Director of Insight Software - an Irish company sold to Hoskyns from the UK in 1988” (O Riain 2004).
Breznitz (2007) criticizes developmental and neodevelopmental studies for arguing that their work posited that in
order to succeed in the developmental efforts in high-technology industries, the bureaucracy needs, first, to have
multiple ties with and embeddedness within, industry and finance, both locally and globally. But Breznitz argues that
there are many ways by which state and industry can interlink.
44
Developmental state works suggest that business would readily engage in the efforts deployed by
the government, but, what if firms are not too interested in collaboration or are so dissimilar (few
big companies and thousands of Small and Medium Enterprises (SMEs) that do not find common
ground to organize? Also, developmental state scholars often assumed that firms in similar sectors
organize similarly across countries, and this is not the case (Cammett 2007; Martin and Swank
2012). Others have noted that there is an interactive effect where state and business may have
multiple configurations and the key to successful collaboration is to reach a balance of power
situation between these two key actors (Breznitz 2007; Kang 2002).
In sum, there are reasons to go beyond the state-centric approach of developmental state
literature and consider business structures and inter-firm relations as well (Kang 2002; O’Riain
2004; Schneider 2013a, 2013b; Yeung 2014).
30
1.2. Business Sector’ Degree of Organization
Firms have incentives to promote or block industrial upgrading policies depending on the
benefits they may receive from such public interventions in the markets. The literatures on
collective action/special interest politics and varieties of capitalism provide valuable groundwork
into the study of business as an actor that shapes industrial upgrading policy adoption. A limitation
of these theories is that most of them stem from the United States context or the Western European
countries, and there have been few attempts to test if these models travel well to the Latin America
and other developing countries context.
31
The theory of collective action rests on the idea that organized groups are able to exert
more pressure on politicians than unorganized groups (Olson 1965). However, Olson (1982)
claimed that sector specific business groups would eventually stifle economic progress unless there
was an overarching encompassing organization that was able to generate consensus and a shared
interest. But Olson underestimated the power of emerging sectors and the interactions they may
30
For instance, O’Riaini has pointed to the importance of business structures: “Furthermore, there is a network of
industry and trade associations, universities, innovation and technology centers and other fora and groups which
provide an associational infrastructure for information-sharing, cooperation and innovation in Irish software. While
these bodies are outside the state or semi-autonomous from it, in most cases they have been founded through state
initiatives and underwritten by state guarantees and funding. Nonetheless they form a distinct layer of institutional
spaces and social networks between the state agencies and the companies in the industry. In this respect, they represent
a relation between the state and industry which is of a quite different nature than the relatively direct personal career
ties.” (2000, 175)
31
For a discussion on the limits of Rational Choice Institutionalism to study Latin America see Weyland (2002).
Schneider (2014) also remarks that Latin American political economists need to pay more attention to the role of
business in their analyses.
45
have with meritocratic bureaucrats who usually have a long-term view of policy and are to a certain
extent, shielded from corruption.
Scholars have further refined the theory by modeling the interactions between state and
firms in the policy-making process determining policy adoption through legislative bargaining,
lobbying and electoral competition (Grossman and Helpman 2001; Persson and Tabellini 2000).
Industrial policy falls in the realm of special interest politics because it is an economic policy
decision that creates concentrated benefits for a few well-defined groups, with the cost diffused in
society at large (Persson and Tabellini 2000, 157). Therefore, the political incentives to influence
the design of such policies are much stronger for the beneficiaries than for the majority bearing
the cost. Thus, we can expect that firms will organize into sectoral associations and press for the
enactment of programs that support their activities.
Hypothesis 2: Countries with an IT sectoral business association are more likely to adopt
industrial upgrading policies than countries lacking an IT sectoral business association.
Regarding the study of developing countries, the varieties of capitalism approach contends
that industrial upgrading policy adoption is not very likely to happen. This is because of the
existence of big business: “business groups in Latin America are larger, more internationalized,
and more diversified in political investments (media, think tanks, business associations, campaign
finance)… [and] these power asymmetries do not in principle compromise business-government
collaboration in passive industrial policy, but they do make active industrial policy and associated
performance standards more difficult and risky” (Schneider 2015, 61).
Other authors posit that the arrival of multinational companies stifles business-government
collaboration for industrial transformation because MNCs subsidiaries are subject to the decisions
taken at the headquarters (which are abroad and usually keep their innovation activities in the
developed world) (Nolke & Vliegenthart, 2009). But the varieties of capitalism approach for
developing countries is at an early stage; thus, there are many opportunities to test and refine it.
Perhaps in their attempt of creating a regional-based category for developing countries these
models rush into generalization and they have overlooked that within Latin America and Eastern
46
Europe, there are successful cases of cooperation to move into higher technological activities
(Hancké 2011; Sánchez-Ancochea 2009).
32
2. Model and Empirical Strategy
This section introduces the data and the empirical strategy used to test the hypotheses about
the influence of states and business characteristics’ in industrial upgrading. The sample comprises
17 countries from Latin America: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica,
Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru,
Uruguay, and Venezuela, in the period 1995-2012. Thus, the units of analysis are countries, which
are observed when they enter the sample (here the start point is 1995) and up to their exit from the
sample or failure (here, countries fail when they adopt a policy). The empirical strategy consists
of an event history analysis.
Latin America is an interesting region to study industrial upgrading policies in the context
of developing democracies. It contains substantial variation regarding the quality of bureaucracy
and business organization as well as human capital levels, resource dependency, manufacturing
sector, inflows of FDI, and trade openness. Plus, efforts for industrial upgrading date since mid-
1990s when a few governments, realizing the strategic importance of the sector, began a new wave
of programs to encourage the expansion of IT firms.
To narrow the analysis of industrial upgrading strategies into a more manageable
phenomenon, I concentrate on national-level policies to stimulate the information technology (IT)
sector, specifically software. The software sector is strategic because it offers potential benefits to
the whole economy and society: it can improve agricultural production, the development of
capabilities in the oil exploration, and the tackling of social and environmental challenges such as
insecurity and natural disasters. Another characteristic is that the level of job skills created in this
sector varies according to the type of service provided. It can range from testing and making minor
32
Sánchez-Ancochea (2009, 4) questioned the HME model by showing how Costa Rica has been able to overcome
some of the problems inherent to hierarchical relations through political action. He argued that the state played a
fundamental role in solving the negative complementarities by fostering investment in human capital and targeting
high-tech FDI. Hancké (2011, 2) also challenged the view that business coordination is exogenously given and handed
down through history under particular conditions: “If a political economy missed the rendezvous with business
coordination at a critical juncture, or destroyed existing or incipient forms of business coordination for whatever
reason, it seems to have lost this chance forever.” The author analyzed the reindustrialization and industrial upgrading
processes in Central Europe (Czech Republic, Poland, Slovakia and Hungary) to uncover the conditions under which
inter-firm strategic coordination emerged in certain geographic areas and industrial sectors.
47
adaptations to computer programs, to developing full solutions and new software for high tech
products. Given the omnipresence of IT in the 21st century and the relatively low entry barriers,
we should observe efforts to promote the software sector in almost every country.
One limitation of focusing on just one sector is that there is potential selection bias and limits the
generalizability of my findings. It may be the case that the dynamics observed in the IT industry
are particular of this sector and not of others. One solution would be to conduct similar analyses
of other areas like biotechnology and aerospace. On the other hand, the choice of this sector also
allows me to continue and build around a tradition of developmental state scholarship that has
studied IT (Evans 1995; Wade 2001; O’Riain 2004; Breznitz 2007).
2.1. Key Variables
The dependent variable, Industrial Upgrading Policy Adoption, refers to the policies
enacted with the purpose of creation or consolidation of new high-tech industries, specifically the
software sector. To generate it, I recorded whether there was a government-sponsored program to
support the software sector and the start year.
33
Usually there is a decree law or an official
announcement followed by the publication of the eligibility criteria. Thus, Software Policy is a
dichotomous variable indicating either the presence 1, or absence 0, of an industrial upgrading
policy for the software sector at the national level.
There are two main independent variables to test for the two hypotheses. The first
independent variable is Bureaucratic Quality. It is related to government characteristics and
operationalizes the concept of meritocratic bureaucracy. It measures three main aspects: 1. whether
countries possess a bureaucracy that has the strength and expertise to govern without drastic
changes in policy or interruptions in government services; 2. If bureaucracy tends to be somewhat
autonomous from political pressure and, 3. If it has an impartial mechanism for recruitment and
training. To operationalize it I use the ICRG indicator of bureaucratic quality which and ranges
from 1 to 4, where 1 is the lowest score and 4 the highest.
34
Although the variable is somewhat
coarse and based on expert assessments, it offers a comparable measure and has yearly coverage
for all the countries included in this study.
35
33
Policies may vary in scope and tools used. Some are broad that is they cover all companies in the sector, through
tax cuts or exemptions. Others are narrower because they subject firms to a competitive process, usually providing
credit to companies that present a specific project for exporting, training or conducting research and development.
34
The maximum score for a Latin American country in the dataset is 3.
35
An alternative measurement of this concept is the Quality of Government index (Dahlström et al. 2015) which
assesses the professionality and impartiality of bureaucracy. Unfortunately, so far only one country-year is available.
48
As a robustness check I use Government Effectiveness from the World Bank Governance
Indicators. Government Effectiveness captures perceptions of the quality of public services, the
quality of the civil service and the degree of its independence from political pressures, the quality
of policy formulation and implementation, and the credibility of the government's commitment to
such policies. It is available for the years: 1996, 1998, 2000, and annually from 2002-2012 for
every country in the sample.
36
For the missing years I interpolated the values when possible. This
variable is less focused on bureaucratic quality than the ICRG indicator, it has a .53 correlation
with ICRG’s Bureaucratic Quality.
The second independent variable is Sectoral Association. It is a dummy variable that
measures the presence of a software/IT sectoral association. Although this is a coarse proxy, it
offers some information about the firms’ cohesiveness in the sector. The presence of a sectoral
organization conveys whether business actors in the IT industry have overcome collective action
problems and are ready to lobby/collaborate with government institutions. To generate it, I coded
the presence or absence of an IT industry association by looking into the members of the Latin
American Federation of IT Associations(ALETI) and reviewing country newspapers, industry
reports and official websites.
37
2.2. Control Variables
Besides government and firms’ characteristics, other potential explanations highlight
features that may impact the decision to enact industrial upgrading policies. To subject my theory
to a more rigorous test, I include these other explanations as control variables. These are classified
into socio-economic, political, and international factors (For a quick overview see Table 1; for data
sources and descriptive statistical measures see Appendix 2).
Nonetheless, in the measurement validity tests the authors find its two components have a .64 and .74 positive
correlation respectively with the ICRG index of Bureaucratic Quality.
36
“When it comes to evaluating the growth- governance linkage, the policy biases embedded in these measures
become even more problematic. For example, one prominent school of thought has high- lighted the importance of
developmentalist policies and competent but interventionist bureaucracies for rapid economic development (e.g.,
Amsden 2001; Wade 1990) according to Amsden (2001)... such states are necessarily “disciplinary” of capitalists,
something that survey measures of business people’s opinions are likely biased against” (Shrank and Kurtz 2005). But
I think that with such bias, government effectiveness may be underestimating the effects of meritocratic bureaucracies.
37
Ideally this information could be supplemented with more detailed measures of industry size, structure (i.e.,
Herfindahl index of concentration) and share of foreign capital/ presence of major multinational corporations in the
sector.
49
2.2.1. Socio-economic Factors
Amongst the socio-economic factors that scholarly work has revealed as important are
those associated with the availability and reliance on natural resources, the existence of a large
pool of skilled workers and more generally higher income levels per capita.
Resource Dependence. The resource curse literature argues that natural resources reliance
generates rent seeking throughout business and government (Karl 1997, Shafer 1994). This hinders
the emergence of policies towards innovation/high-tech that may challenge the status quo. In fact,
natural resources possession may lead to a lack of meritocratic bureaucracy and also an absence
of sectoral associations of IT.
Prior studies have examined the mechanisms by which plentiful natural resources
disincentive the promotion of non-oil based economic activities through public policies (Karl
1997). In this view dependence on a particular export commodity shapes social classes, regime
types, and the very institutions of the state, the frameworks for decision-making, and the decision
calculus of policy makers. In oil dependent states, government officials have additional capacity
to extract unusually high income from their resource without added investment. Both public
authorities and private interests engage in rent seeking and “structures may create overwhelming
incentives for decision-makers to choose (or to avoid) a specific set of policies” (Karl 1997, 12).
Along the same lines, other studies have shown that state’s capacity to move forward depend on
the attributes of the leading sector through which it is tied to the international economy (Shafer
1994).
38
So we can expect that larger reliance on natural resources rents will have a negative
relationship with the likelihood of industrial upgrading policy adoption to stimulate the software
sector.
To operationalize this concept, I use two measures. The first one is Commodities Exports
is a variable that partly captures reliance on resources and potential rent-seeking. It is the
percentage of commodities exports of the total amount of exports: agricultural, food, minerals and
oil exports. The second is Natural Resources Rents a variable from the World Bank Development
Indicators which is defined as the total natural resources rents are the sum of oil rents, natural gas
rents, coal rents (hard and soft), mineral rents, and forest rents serves as an alternative.
38
Economists argue that resource dependence dampens economic development through the Dutch Disease: that is
favorable price changes in one sector of the economy, for example, petroleum, causes distress in other sectors – for
example, agriculture or manufacturing (Corden 1984; Roemer 1983). Sachs and Warner (2001) claim that there is a
significant effect of the possession of natural resources on growth and on manufacturing exports growth.
50
Skilled Workers Availability. A critical input for the Information Technology sector is human
capital. Scholars argue that in developing countries there is a low-skill equilibrium trap where
governments have fiscal constraints to invest heavily in public education and skill formation, and
companies do not want to bear the costs of training. Even MNCs demand less skilled labor and
workers do not invest in technical skills (Soskice and Schneider 2009). If policy makers consider
the chances of success of the programs they promote we can expect that industrial upgrading policy
adoption, if any, will only be likely after some preconditions are present: high amounts of human
capital/skilled workers and the appropriate infrastructure, which in this case is internet access.
39
The variable Human Capital from the Macro IPE dataset is an available proxy for this
concept. It measures the skilled-workers availability based on percentage of tertiary schooling
attained. It comes from the World Bank based on Barro-Lee. This indicator is highly correlated
(.85) with another variable from CEPAL that measures the percentage of population with
minimum high-school education (more than 12 years of education).
40
Income levels. Gross domestic product (GDP) per capita is associated with better quality of
government although the relationship is subject to reverse causality. In fact Evans and Rauch
(1999) demonstrated that higher levels of meritocratic bureaucracy led to higher rates of GDP per
capita growth. Given the disparity of GDP per Capita levels in the region I use the natural log of
the variable available from the World Bank dataset.
My concern with the GDP per Capita variable is that it is too general, and it may be
capturing several of the more specific factors associated with software policy adoption such as
39
“Conventional explanations of why some countries have been more successful in sustaining high technology
investments have focused on infrastructure and education, but though important these aspects of industrial policy do
not take us far enough: institutions that induce and compel rapid learning are needed. Critical conditions for success
include: a capacity of the state to monitor and make judgments about performance capacity to relocate the subsidies
and assets of non-performers but this in turn requires critical political capacities: The organization of power in society
must be compatible with the rent management that state institutions are trying to implement (Khan and Blankenburg
2009, 34).
40
Barro and Lee argue that their estimates of educational attainment provide a reasonable proxy for the stock of
schooling capital for a broad group of countries. However, Hanushek and Woessman (2012) argue the school
attainment does not take account of the skills and experience gained after formal education. The measure does not
directly measure the skills obtained at schools and, specifically, does not take account for differences in the quality of
schooling across countries. Unfortunately, their dataset does not cover the years/countries in the present study. The
results of the Science and Math PISA tests would be a better alternative, but the information for Latin American
countries is only available for 7 countries in the region since 2002.
51
human capital and bureaucratic quality. In fact, GDP per Capita has a relatively strong correlation
with government effectiveness indicators (.51). And when included in the regressions it generates
very high variance inflator estimators. So most likely bureaucratic quality is a mechanism through
which GDP per capita stimulates industrial upgrading. In Jaccard and Jacoby’s (2010) terms,
bureaucratic quality might be a partial mediator of the impact of GDP per capita. Human capital
could be another partial mediator of the GDP per capita effect on the adoption of software policy.
By including these two more specific indicators (bureaucratic quality and human capital)
considered mediators in the quantitative tests, instead of the coarser GDP per Capita I seek to
refine extant theories. As Jaccard and Jacoby note, “the main idea is that you can expand a
theoretical framework that has direct causes by turning direct causal relationships into indirect
causal relationships through the specification of mediators” (2010, 147). Because to date,
economic theories have shown how countries with higher GDP per capita levels move into
knowledge intensive activities but we do not know exactly through which mechanisms or under
what conditions do they adopt policies to stimulate innovative sectors such as information
technologies.
2.2.2. Political Factors
The effects of political variables over policy-making decisions are well theorized and tested
(Cox and McCubbins 2000; Mainwaring and Shugart 1997; Tsebelis 2002). Studies of economic
policies such as trade liberalization and capital controls also account for the impact of political
institutions and executive ideology. Industrial upgrading policy is usually launched by the
executive and then is subject to a ratification or budget allocation process in the congress. Thus,
the characteristics of these institutions shape the adoption likelihood, especially the executive
ideology.
41
Institutional studies of policymaking have centered on executive discretion—including the
constraints derived from the distribution of partisan power across branches because all Latin
American countries are presidential (Murillo and Martínez-Gallardo 2007). So, I include the
executive ideology as a control.
41
“Latin America’s political institutions suggest that firms should seek particularistic policy benefits. Most countries
combine presidentialism with proportional representation. Presidentialism allows for greater volatility in policies of
interest to businesses, giving them an incentive to cultivate direct ties to politicians. Meanwhile, proportional
representation makes legislative coalitions the norm, exerting a check on executive power and offering a variety of
parties and politicians that firms can pay off to gain policy influence” (Boas, Hidalgo, and Richardson 2014, 416–17).
52
Executive Ideology. Scholarly research on the determinants of trade and financial liberalization
has found that the partisanship of the executive influences policy decision-making (Alesina, Grilli,
and Milesi-Ferrett 1993; Brooks and Kurtz 2007; Kastner and Rector 2005; Quinn and Toyoda
2007). Since industrial upgrading has been much less studied, there is no straightforward
expectation on how a left leaning or right leaning government may affect the adoption process.
The traditional support base for left parties has been manufacturing workers. This
constituency may fear the “loss of employment” that is usually associated with the introduction of
new technologies and prefer programs to support the industries in which they work rather than
those fostering industrial upgrading. On the other hand, right governments may be more business
friendly and enact favorable policies to the private sector. But politicians may also oppose
industrial upgrading if it harms their supporting base of firms concentrated in the agricultural
and/or traditional manufacturing sectors.
42
The effect of government ideology on industrial
upgrading policy is ambiguous.
To operationalize this variable, I resort to presidential ideology from the Baker (2016)
dataset. This variable codes the ideology of the executive from 0 to 16 The data contains
information from 1995 until 2016. The measure relies on fine-grained ideological distinctions
between parties from across the left-right spectrum. Where zero is extreme left and sixteen extreme
right.
2.2.3. International Factors
International factors are another relevant to an explanation of industrial upgrading.
Foreign Direct Investment. The amount of foreign direct investment stocks that a country possess
will have a negative effect on business-government cooperation and the adoption of industrial
upgrading policies (Nolke and Vliegenthart 2009; Paus and Gallagher 2008; Schneider 2013b).
Ideally the variable employed should be the FDI stocks/ GDP, nevertheless data is very
incomplete. For this reason, I use FDI Inflows as a percentage of GDP from the World Bank.
42
Work on the adoption of market reforms in Latin America has emphasized a decline in partisan policy preferences
(Murillo and Martínez-Gallardo 2007, 123). Their research also finds evidence pointing to the apparent death of
ideology in Latin America during the 1980s and 1990s. But political competition was central to policy adoption.
53
Table 1. Summary of Hypotheses and Variables
Variable Measurement Relationship
Dependent
Variable:
Software
Policy
Presence of software industry sectoral program.
1 if there is a program, 0 otherwise.
Independent
Variables:
State
Quality of
Government
Bureaucracy Quality: high scores are given to countries
where the bureaucracy has the strength and expertise to
govern without drastic changes in policy or
interruptions in government services. Bureaucracy
tends to be somewhat autonomous from political
pressure and to have an impartial mechanism for
recruitment and training.
+
Business Sectoral
Association
Presence or absence of a software industry association
in the country
+
Controls
Socio
Economic
Factors
Natural
Resource
Rents
Commodities
Exports
Human
Capital
GDP per
Capita
Total natural resources rents are the sum of oil rents,
natural gas rents, coal rents (hard and soft), mineral
rents, and forest rents.
Is the percentage of commodities exports of the total
amount of exports: agricultural, food, minerals and oil
exports.
Skilled workers availability based on % of tertiary
schooling attained
Natural Log of Gross Domestic Product per Capita
Political
Factors
Executive
Ideology
The ideology (partisanship) of the executive power. It
ranges from 1 to 16, from extreme left (1) to extreme
right (16).
International
Factors
Foreign
Direct
Investment
Net Foreign Direct Investment inflows as a percentage
of GDP
54
2.3. Empirical Strategy
To test for the probability of enacting an industrial upgrading policy over a given period of
time, I resort to event history analysis. Event history analysis allows the researcher to explore the
factors that affect the probability that a country will choose a particular policy, given that it has
not done so up to that point in time.
The event history analysis, also known as survival-time analysis, or hazard model considers
a set of cases as being in “risk” until they experience the event under study (also known as failure).
Thus, the independent variable conveys the information of duration time (Box-Steffensmeier and
Jones 2004, 70). I examine the binary time-series-cross-section (BTSCS) data using standard logit
analysis for grouped duration data (Beck and Katz 2011). A limitation of this technique is that it
considers the observations to be temporally independent. Beck, Katz and Tucker. (1998) cautioned
about estimating standard logit on BTSCS data under conditions of temporal dependence because
it produces standard errors that are likely to be wrong and t-values that are inflated. The authors
recommended the use of cubic splines to account for potential time dependence. However, a
simpler alternative is the use of a cubic polynomial approximation by including time (t), time
squared (t
2
), and time cubed (t
3
) in the model as Carter and Signorino (2010) advice. I follow this
latter approach.
43
Robust (Huber-White) variance estimation with clustering by country is
employed to account for any possible heteroscedasticity. To avoid endogeneity issues and temporal
ordering, the main independent variables, the socio-economic and international variables have a
one-year lag (Box-Steffensmeier and Jones 2004, 111).
To aid the understanding of the phenomenon of interest, Figure 1 displays a graphical
depiction of the failure rate over time. It shows the evolution of the adoption of industrial
upgrading policies in Latin America from 1995, the first year observed, to 2012 which is the last
year covered by this study. The number at risk quantities in the bottom account for the number of
countries in the group that have not adopted a software sector policy. By 2012 which is the last
observed year, 10 out of 17 countries had implemented this type of industrial upgrading policy.
43
Since it is a small sample and to deal with possible multicollinearity, the time variable (t) is demeaned (t-mean(t))
before squaring (t
2
) and cubing it (t
3
). And t
3
is divided by 1000 to avoid numerical instability which is a potential
threat (Carter and Signorino 2010, 283).
55
Figure 1. Adoption of Software Industrial Policies in Latin America (1996-2012)
Analysis time = 0 in 1995
3. Results
In the theoretical section, it was hypothesized that countries with meritocratic
bureaucracies (higher levels of Bureaucratic Quality) are more likely to adopt industrial upgrading
policies than the rest. Table 2 shows the results of regressions testing the hypothesis and controlling
for different factors such as socioeconomic (models 2-3), international (model 4) and political
(model 5).
Bureaucratic Quality is positive and significantly related to Software policy adoption and
keeps its significance when controlling for other socioeconomic, political and international
variables in all specifications.
The models including GDP per capita and Human Capital presented severe collinearity
issues. As I argued before, GDP per capita is a variable that is too broad and can obscure the effects
of other factors. That was the case because the models which included GDP per capita presented
very high variance inflation indicators. I include Human Capital in the model 2 to show how
bureaucratic quality preserves its significance but the model also has high variance inflation
indicator.
56
Table 2. The Effects of Bureaucratic Quality
on Software Policy Adoption in Latin America (1995-2012).
DV: Software Policy Adopted
(1) (2) (3) (4) (5)
Bureaucratic 2.345** 2.193* 2.152* 2.148* 2.293*
Quality
t-1
(2.93) (2.48) (2.52) (2.46) (2.37)
Human 0.565
Capital
t-1
(0.49)
Commodities -0.0223* -0.0236** -0.0202*
Exports
t-1
(-2.40) (-3.03) (-2.09)
FDI
0.0430 0.0359
Inflows
t-1
(0.34) (0.25)
Executive 0.0697
Ideology
t-1
(0.70)
Constant -8.017*** -9.066** -6.229* -6.312* -7.859+
N 193 176 192 192 192
Pseudo-R 0.172 0.156 0.205 0.206 0.212
Robust Standard Errors Clustered by Country
+ p<0.10 * p<0.05 ** p<0.01 *** p<0.001.
Time controls included but not shown.
57
Regarding the control variables Commodities Exports shows a consistent negative and
significant effect. The political variable Executive Ideology had a negative yet insignificant relation
with software policy adoption when included in the tests. That Bureaucratic Quality significance
level survives even after controlling for the Executive Ideology supports the idea that, once
bureaucratic agencies have secured a certain amount of technical knowledge and autonomy, they
can act regardless of the political whims that developing democracies experience. The finding that
Executive Ideology effect is not significant is in line with the literature on the adoption of market
reforms in Latin America which has emphasized a decline in partisan policy preferences (Stokes
2001, Weyland 2002, Stallings and Perez 2000 cited in Murillo and Martinez 2007). The
international variable Foreign Direct Investment inflows is negatively associated with software
policy but it did not attain significance levels.
To better illustrate the effects of Bureaucratic Quality on the probability of Software Policy
adoption one can look at the predictive margins graph. It seems there is a threshold at the value of
2 and after that threshold it is more likely that countries implement industrial upgrading policies.
The sample size is small so confidence intervals are wide.
Figure 2. The Effects of Bureaucratic Quality on the Likelihood of Software Policy
Based on estimates from model 5, other values held at the mean. Adjusted predictions with 90%Cis.
58
As a robustness check I ran the regressions excluding each country at a time. The results
hold well. Another robustness check is to use Government Effectiveness which is a continuous
variable instead of Bureaucratic Quality. The results are very similar and the variable accounting
for the quality of government has a stronger effect (see Appendix 3).
Table 3 presents the results of the second hypothesis which tests whether the presence of a
Sectoral Association impacts the likelihood of software policy adoption. Models 1 and 3-5 reveal
that in fact, sectoral association presence has a positive and significant effect on the probability of
adoption. Bureaucratic Quality remains positively related but its significance drops below the
threshold in all the models except model 2.
Table 3. The Effects of Sectoral Association
on Software Policy Adoption in Latin America (1995-2012)
DV: Software Policy Adopted
(1) (2) (3) (4) (5)
Sector 2.282* 1.530 1.692+ 1.772+ 2.010+
Association (2.12) (1.30) (1.92) (1.89) (1.85)
Bureaucratic 1.976* 1.547 1.500 1.625
Quality (2.21) (1.58) (1.48) (1.51)
Commodities -0.0002* -0.0003** -0.0002+
Exports (-2.07) (-2.67) (-1.80)
FDI 0.0906 0.0944
Inflows (0.75) (0.72)
Executive 0.111
Ideology (1.53)
Constant -0.810 -4.018 -4.342 -5.139 -4.730
(-0.21) (-0.80) (-0.86) (-1.06) (-0.95)
N 193 193 192 192 192
pseudo R-sq 0.115 0.206 0.253 0.260 0.273
Robust Standard Errors Clustered by Country. + p<0.10 * p<0.05 ** p<0.01 *** p<0.001.
Time controls included but not shown.
59
In the robustness checks using government effectiveness as a proxy for the quality of
government, the positive and significant relation holds better even when including sectoral
association (see Appendix 4). The effect of Commodities Export is negative and significant
(models 3-5). Meanwhile, foreign direct investment goes in the expected direction but is not
significant. The executive ideology does not seem to have a clear effect.
Figure 3 presents a graphic depiction of the effects of sector association on the likelihood
of software policy adoption at different levels of Bureaucratic Quality.
Figure 3. Effect of Sector Association on the Likelihood of Software Policy
Based on estimates from model 5, other values held at the mean. Adjusted predictions with 90%Cis.
One of the limitations of this analysis is that there are insufficient observations to examine
the interaction effects between Bureaucratic Quality and Sectoral Association. But the results
suggest further study is needed to fully flesh out how is it that business and government traits shape
the industrial upgrading policies.
60
4. Conclusions
This chapter’s central argument is that government and business sector characteristics
impact the adoption of industrial upgrading policies. The two main hypotheses were that countries
with bureaucracies insulated from political pressures and with an impartial recruitment
mechanism, are more likely to set up programs to stimulate high technology sectors than the rest.
And countries having a business sectoral association are more likely to adopt industrial policies
for upgrading than others with less organized firms.
In generating and testing these hypotheses the chapter has made a bi-fold contribution:
First it expanded and refined the political economy theories of industrial upgrading in democratic
developing countries by drawing on prior works from the developmental state and varieties of
capitalism literature. And second it provided a first quantitative cross-country test through an event
history analysis of the policies to stimulate the software sector in 17 democratic Latin American
countries from 1995-2012.
The main takeaway from the event history analysis is that the quality of state bureaucracy
and the level of organization of the business sector matter for industrial upgrading policy adoption.
However, this is a very limited sample, with only 9 events out of 210 observations. We should try
to test these hypotheses for a larger set of developing countries. Also, further exploration of how
business and government characteristics shape the adoption of policies through in-depth case
studies is necessary.
Finally, the relative importance of the quality of state bureaucracy and business sector
organization suggests that cooperation between meritocratic bureaucrats and organized business
may facilitate the adoption, design and implementation of policies to stimulate transition into
higher technological levels of production even when the long-run structural context is unfavorable
(i.e. low levels of human capital or reliance on natural resources). Future research should also
attempt to assess the design and implementation stages of these programs
61
Appendix 1.
Industrial Upgrading Policies for the Software Sector in Latin America
Note: I recoded Venezuela because they do have some policies towards the industry starting 2001.
Table 1. Industrial Upgrading Policies for the Software Sector in Latin
America
1997-2015
SW
Association
Specific SW
Industry
Program.
Year of
Adoption
Legal instrument Scope Agency
Program
renewal or
transformation
Brazil
1976 1997 Law 10.176/2001
Domestic and
International
Development
Bank
BNDES/Softex
2004, 2011
Costa Rica 1998 1999
Signed Memorandum
with IDB
Domestic and
International
Procomer 2006
Uruguay
1989 2000
Decree 84/1999.
Executive Decree
386
Domestic and
International
Diprode 2007
Mexico 1945/1997 2002*
Announced Ministry
of Economy. Diario
Oficial de la
Federacion.
Domestic and
International
Ministry of
Economy
2007, 2014
Argentina
1982 2004 Law 25.922
Domestic and
International
Ministry of
Science
(ANPCYT)
The 2004 law
was set to last
until 2014
Chile 1993 2004
National Digital
Agenda
Domestic and
International
Ministry of
Economy/Corfo
2013
Peru
2000 2006
Registered
Trademark
CreaSoftware 1996.
International PromPeru
El
Salvador
2010 2007
Legislative Decree
No. 431.
International N.A.
Colombia 1998 2009
Tax Law 2009.
Estatuto tributario
2009
Domestic and
International
MINTIC 2015
Nicaragua None 2012 Tax Law 822. International N.A.
Honduras 2009 None Free Zone Law. Ley
de Zonas Libres,
couldnt confirm.
International
Bolivia 2005 None
Ecuador 1995 None
Paraguay 2005 None
Venezuela 1983 None
Panama 2004 None
Guatemala 2005 None
Source: Elaborated by the Author with information from Rovira &Stumpo (2011); Tigre &
Marques(2009) and Government and Business Associations Internet Websites
62
Appendix 2.
List of Variables and Sources
Obs. Mean Std.
Dev.
Min. Max. Source
Software
Policy
210 0.04 0.20 0.00 1.00 Presence of software industry sectoral
program. Coded by the author with
information from Rovira and Stumpo
(2013), Tigre and Marquez (2009) and
countries’ official sources.
Bureaucratic
Quality
193 1.81 0.60 1.00 3.00
Bureaucracy Quality. (International
Country Risk Guide (ICRG) 2017).
Government
Effectiveness
192 0.80 0.48 0.00 2.50
World Development Indicators. World
Bank.
Sector
Association
193 0.52 0.50 0.00 1.00 Dummy indicating the presence of a
software sector association. Elaborated by
author with data from ALETI, newspapers
and official websites.
Sectoral
Association
Age
193 4.08 5.61 0.00 21.00 Years since the creation of the software
sector association. Elaborated by the author.
Human
Capital
176 2.39 0.31 1.68 2.91 Barro and Lee (2012)
Commodities
Export
192 72.88 20.16 6.53 95.91
World Development Indicators (2017)
Natural
Resources
Rents
193 1.35 1.10 -1.34 3.68
World Development Indicators (2017)
GDP per
Capita (log)
193 7.80 0.60 6.81 8.95
World Development Indicators (2017)
Executive
Ideology
193 13.56 4.33 2.65 18.80
Baker and Greene (2016)
FDI Inflows
(% of GDP)
193 3.57 2.96 -2.50 17.13 World Development Indicators (2017)
63
Appendix 3.
Robustness Checks Using Government Effectiveness
The Effects of Government Effectiveness on Software Policy Adoption in Latin America.
1995-2012
DV: Software Policy Adopted
(1) (2) (3) (4) (5)
Government 2.558** 2.800*** 2.840** 3.362* 3.376*
Effectiveness
t-1
(3.04) (3.53) (3.26) (2.42) (2.49)
Human -0.966
Capital
t-1
(-0.65)
Commodities -0.0312*** -0.0268** -0.0254*
Exports
t-1
(-3.82) (-2.98) (-2.54)
FDI
-0.153 -0.149
Inflows
t-1
(-0.85) (-0.85)
Executive 0.0355
Ideology
t-1
(0.51)
Constant -5.150*** -2.984 -3.667** -3.937* -4.564+
(-5.41) (-0.85) (-2.88) (-2.48) (-1.93)
N 192 175 191 191 191
Pseudo R-sq 0.178 0.171 0.244 0.256 0.258
Robust Standard Errors Clustered by Country
+ p<0.10 * p<0.05 ** p<0.01 *** p<0.001.
Time controls included but not shown.
64
Appendix 3 (Continued).
Robustness Checks Using Government Effectiveness
The Effects of Sector Association on Software Policy Adoption in Latin America. 1995-2012
DV: Software Policy Adopted
(1) (2) (3) (4) (5)
Sector 2.282* 1.593 1.724* 1.660* 1.754*
Association (2.12) (1.33) (2.33) (2.16) (2.01)
Government 2.065* 2.584* 2.926+ 2.838+
Effectiveness
t-1
(2.12) (2.39) (1.77) (1.75)
Commodities -0.0003** -0.0003** -0.0003*
Exports
t-1
(-3.26) (-2.99) (-2.45)
FDI
-0.0884 -0.0748
Inflows
t-1
(-0.44) (-0.37)
Executive 0.0594
Ideology
t-1
(0.83)
Constant -4.167*** -5.697*** -4.887*** -5.020*** -5.975**
(-4.39) (-6.30) (-4.72) (-3.85) (-3.28)
N 193 192 191 191 191
Pseudo R-sq 0.115 0.213 0.303 0.306 0.309
Robust Standard Errors Clustered by Country
+ p<0.10 * p<0.05 ** p<0.01 *** p<0.001.
Time controls included but not shown.
65
Chapter 4.
State-Business Interactions in Industrial Upgrading Policy Processes:
The Case of Mexico’s Program to Support the Software Sector
The event history analysis on the adoption of policies to stimulate the software sector
presented in Chapter 3 confirmed that in a more democratic setting and in the digital era,
meritocratic bureaucrats and sectoral associations shape the possibilities of adoption of industrial
upgrading policies. Nonetheless while the quantitative analysis reveals the presence of correlation,
we are unable to sort out causation. The number of observations is limited thus testing for
interactions is not feasible. This chapter complements and expands the inquiry through an in-depth
case study of Mexico, where a program to support the software sector (Prosoft) was devised and
approved in the context of increasing democracy in 2003. Information for the case comes from
twelve semi-structured interviews with key actors involved in the policy making process
44
; and
newspaper articles and reports from 1993 until 2004.
The two research questions addressed will be: What types of state-business interactions
lead to the adoption of industrial upgrading policies? And, what are the mechanisms through
which bureaucrats and businesses shape the likelihood of adoption and policy design? My
general argument is that both states and business are crucial to the development process. More
specifically I contend that the level of bureaucratic quality and the degree of business cohesion
generate a variety of patterns of state-business interactions. In turn, these patterns shape the
processes of adoption and design of industrial upgrading policy.
The chapter is organized as follows: Section 1 presents the concepts, argument, and the
research strategy, to situate the Mexican case within the analytical framework. Section 2 provides
some background about the software sector in Mexico. Section 3 gauges the levels of bureaucratic
quality and of business articulation existing prior to Prosoft. Section 4 examines the pattern of
business-government relations and its mechanisms during policy making. Section 5 moves from
44
Appendix 1 provides a list of the interviewees.
66
the policy process to its outcomes. The conclusions and further research steps are discussed in
section 6.
1. Concepts, Argument and Research Strategy
The three central concepts in the analytical framework of this dissertation are industrial
upgrading policies, bureaucratic quality and business cohesion. After going over their definitions,
in this section I present the argument and the research strategy.
Industrial upgrading policies are those strategies seeking the creation or consolidation of high-
technology sectors in which knowledge plays an essential role, for instance software,
biotechnology and aerospace.
Bureaucratic quality is the extent to which government agencies possess Weberian
standards of: meritocratic recruitment, predictable careers or ladders, competitive wages, and
technical expertise (Dahlström, Lapuente, and Teorell 2012; Evans 1995; Evans and Rauch 1999).
It ranges from high to low. High quality bureaucrats possess relevant professional degrees; acquire
technical knowledge over time; expect to develop a long-term career and climb the organizational
ladder; and receive competitive wages. Low quality bureaucrats lack adequate studies for their
positions; are selected through nepotism; and usually have short term expectations, as soon as the
politician who appointed them finishes his period, they are replaced. Prior studies have shown that
bureaucratic quality impacts economic policies and their outcomes (Amsden 2001; Evans 1995;
Johnson 1982; Kohli 2004; Rauch and Evans 1999; Wade 1990). Also where there is meritocratic
recruitment, bureaucrats will be less interested in bribes offered by business (Dahlström, Lapuente,
and Teorell 2012).
And business cohesion is the degree to which firms in the same sector coordinate their
policy demands and their actions. It ranges from high to low. When there are high levels of
business cohesion, firms are organized into a sectoral association which: represents all types of
firms; is well organized and staffed, and uses its institutional strength for productive ends (Doner
and Schneider 2000, 263).
45
The association may include domestic and foreign companies,
although legislation in some developing countries bans foreign-owned firms from participating.
45
According to Doner and Schneider (2000), associations’ institutional strength depends in turn on internal factors:
high member density, valuable selective benefits (often delegated by governments), and effective internal mediation
of member interests and disputes. In addition, external factors, especially competitive markets and government
pressure.
67
Several advantages result from a high level of articulation. First, associations aggregate
information that is useful for decision making. Also, they channel firm demands’ in an organized
and coordinated way, reducing conflict and limiting narrow interests. When government seeks to
cooperate with firms in a sector “cooperation is further facilitated by intra-organizational cohesion.
Cohesive organizations can make credible commitments because of their ability to control shirking
among their members” (Weyland 1997, 53).
In contrast, when there are low levels of business articulation and a sectoral association is
missing or too weak, each firm lobbies for its interests through its own means. Big businesses are
in an advantageous position because they have more resources to staff government-relations
departments; pay for lobbying firms and/or fund campaign elections. Meanwhile, small and
medium sized firms may face a hard time voicing their concerns unless they possess strategic ties
to politicians or bureaucrats.
I hypothesize that the combination of bureaucratic quality and business cohesion generates
four broad patterns of business-government interactions (See Figure 2). In turn, I argue that these
varied patterns lead to different policy processes and thus outcomes. These patterns have already
been discussed extensively in Chapter 1, so here I just review the two extremes: a synergistic
process and a quiescent process.
Figure 1. Patterns of State-Business Interactions
Bureaucratic Quality
Low High
High
Business-led
Process
Synergistic
Process
Business
Cohesion
Low
Quiescent
Process
State-led
Process
A synergistic process emerges when both bureaucratic quality and business cohesion are
high. High quality policymakers are in a position of technical expertise and negotiation skills that
enables them to collect and assess information from several sources, generate long term
development plans and work closely with business without being captured by firms’ interests
68
(Etchemendy 2011; Evans 1995; Fairfield 2015). Business articulation is also relevant for policy
adoption and design. High business articulation allows firms in a sector to better aggregate
information, set clear goals and negotiate with government as a block. Efforts to affect government
indirectly through associations are likely to be less distorting and narrowly rent seeking than those
made directly by firms in the form of unofficial payments to officials (Nugent and Sukiassyan
2009, 425). Once policy is approved, the government may even grant the associations the
responsibility of distributing the benefits derived from programs (Cammett 2007; Doner and
Schneider 2000; Haggard, Maxfield, and Schneider 1997).
In a synergistic process, we can expect the presence of two mechanisms leading to policy
adoption. The first one is the information collection and assessment: bureaucrats consider business
input and other sources of information to generate a policy proposal; and the second one is a formal
collaboration platform between business and government that facilitates exchanges and
negotiation with other actors.
Theoretically, this is one of the best scenarios for industrial upgrading because both
business and states will work towards a common goal from a position where relevant information
is available and there are management capabilities. Special interests are restrained. Adoption of
programs is more likely and these will probably be competitive, that is funding will be awarded to
the best proposals under clear selection criteria.
On the other extreme is the quiescent process where state-business interactions are
characterized by low levels of bureaucratic quality and low levels of business cohesion. On the
government side, low levels of bureaucratic quality generate an environment where elected-
government actors, impose their will more easily over bureaucrats, so they may push programs
that benefit their own interests/constituencies instead of trying to promote industrial upgrading.
And in the private sector, there are considerable asymmetries in firms’ access to policymakers.
Large firms often have political affairs departments as well as the resources to hire lobbying firms
and/or fund elections. These resources grant them easier access and more influence over
government than most of the small and medium sized firms. Furthermore, with no business
association in place, firms provide partial information to bureaucrats who must collect and
integrate it. But since bureaucrats have limited skills to assess the proposals, the results are most
likely to be either biased or limited.
69
This combination results in the absence of two crucial mechanisms, there is a lack of an
information collection and assessment on the side of the bureaucracy and no formal dialogue with
business. None or uncoordinated actions from firms and bureaucrats emerge. In a quiescent pattern
of business-government interactions attempts of industrial upgrading are difficult. Thus, adoption
of programs is unlikely or when policies are adopted these will probably be particularistic, that is
only benefiting a few firms with close ties to bureaucrats or politicians.
In what follows I undertake an in-depth case study to process trace how the level of
bureaucratic quality and business cohesion shape the policy making processes towards the
software sector. The quantitative analysis in Chapter 3 found a positive and significant relation
between the level of bureaucratic quality and the likelihood of adoption of industrial upgrading
policies. The levels of business sector articulation showed a positive and significant relationship
as well. But quantitative studies are limited: they cannot fully sort the direction of causality and
have measurement issues.
An in-depth case study expands the analysis in several ways and is suitable for theory
building (George and Bennett 2005; Lieberman 2005). First, it allows for a finer-grained
measurement of the concepts of bureaucratic quality and business cohesion. Furthermore, through
process tracing one can explore the direction of the relationship between bureaucratic quality and
industrial upgrading. An in-depth case allows for the identification of the mechanisms through
which bureaucrats and business sectors shape the adoption and design of policies. As Beach and
Pedersen note: “In what we term mechanism-centered designs, mechanisms are traced to gain
better understanding of the causal relationship between X and Y within a bounded population,
resulting in mechanistic evidence that sheds light on causal processes. Here, the theorized
mechanisms are front and center in our analysis, with the analytical focus on assessing how the
causal arrow(s) in-between X and Y actually works in particular cases” (2016, 4). Additionally, a
case might also reveal whether additional actors played important roles.
Regarding case selection, Beach and Pedersen (2016) suggest that if we are interested in
tracing a causal mechanism linking X and Y, we want to trace it in cases where it could have been
present, at least in theory. Similarly Seawright (2016, 75) recommends selecting extreme cases on
the independent variables as a valuable and underrated strategy for exploring hypotheses about
causal paths.
70
As was shown in Chapter 3, during the past two decades, several Latin American countries
have enacted variegated policies to stimulate their software sectors under diverse business-
government collaboration patterns. Ten out of seventeen countries studied in the region adopted a
sectoral policy for the software industry between 1995 and 2012: Argentina, Brazil, Colombia,
Costa Rica, Chile, El Salvador, Mexico, Peru, Uruguay and Venezuela.
46
Of these I focus here on
the case of Mexico.
Relative to the universe of cases in Latin America, Mexico has extreme values on my key
explanatory variables: high business cohesion and relatively high bureaucratic quality. My
expectation, in brief, is that in the Mexican case I should observe a synergistic pattern of business-
government collaboration. The Mexican case of software policy adoption is especially puzzling
because a policy was embraced in a country where neoliberalism is deeply entrenched. In fact
politicians and policymakers have rejected most kinds of industrial promotion of sectors since the
mid 1980s (Moreno-Brid 2013).
2. Background: The Early Years of the Software Sector in Mexico
The origins of the software sector in Mexico are partly related to the evolution of the
computer and electronics industry, yet a major difference is that the software sector was never
favored with a specific industrial policy prior to the year 2001. In the late 1970s, the National
Industrial Development Plan targeted the computer and electronic parts sector as strategic for the
first time. The Ministry of Trade and Industrial Support (SECOFI) in 1981 devised the “Program
to Promote the Manufacturing of Electronic Computer Systems”
47
with the objective of generating
a local production of mini and micro computers, promoting exports and acquiring technological
autonomy in this sector (Mochi 2006, 72). A market reserve was set to protect the electronics and
hardware segments requiring companies to spend 4.5% of the total net sales on R&D. All imports
were subject to discretionary permits and high duties (Noriega 1993a, 2).
The economic liberalization that took place in Mexico after the mid-eighties swept away
all industrial policies including the electronics sectoral program. A decree issued by SECOFI in
April 1990 opened the borders to international products, eliminating the need of import permits
46
For a thorough description of the evolution of the IT sector throughout Latin America, see Rovira and Stumpo
(2013) and Bastos and Marquez(2009).
47
The Spanish name was “Programa para la Promoción de la Manufactura de Sistemas Electrónicos
Computacionales.”
71
for new equipment and treating information technology (IT) imports under the same uniform
conditions applied to all imports (stable duties that never exceed 20%) (Ibidem).
Firms in the software sector sprouted in the eighties due to the increasing domestic demand
for PCs and the gradual diffusion of computers into public and private institutions’ activities.
Microsoft set up operations in Mexico in 1986 and Oracle in 1987. There were also Mexican
companies offering tailored solutions for small and medium enterprises and large domestic
conglomerates created their own informatics departments to cater their needs. About 60 to 70% of
the software sold in Mexico was foreign (Mochi 2006, 79). In the early nineties, there was a
constant increase in the demand of IT goods and services. The IT market showed an average yearly
increase of over 30 percent. The software segment was not subject to market reserve regulations
yet there were tariffs and permits for imported products that were also removed by SECOFI in
March 1990 (INEGI 1993). Firms organized around a sectoral organization, the National
Association of Computer Programs Industry (ANIPCO) in 1985.
High-rank bureaucrats and businessmen from the information technologies sector in
Mexico recognized the need of adopting an industrial policy to promote this strategic sector.
48
A
Program for Informatics Development was announced by the executive power in May 1996, during
Ernesto Zedillo’s administration (1995-2000)(Diario Oficial de la Federación 1996).
49
INEGI was
designated as the coordinating institution, and other agencies like the Ministry of Trade and
Industrial Promotion (SECOFI) and the Ministry of Education would participate. The private
sector was invited to a consultative forum around the governmental plan but it was not thoroughly
involved in the design or adoption process. Overall, the initiative was vague, no specific funds or
resources were allocated for the objectives so the program failed to materialize. As an industry
leader remarked: “the National Program for Informatics Development is full of good intentions, it
lays out the goals, but it does not explain how” (E. Lopez 1998b).
48
The National Institute for Statistics, Geography and Informatics (INEGI) was the main government agency in charge
of the “informatics policy” until the late 1990s. Yet the majority of its actions were related to the adoption and use of
information technologies in the public administration; and the creation of copyright laws (Noriega 1992). INEGI
organized a consultative forum intellectual property and the copy right laws in 1996 (Castro 1996). In it specialists
from the public sector, the private sector and academia expressed their opinions and made policy proposals about the
legal and administrative framework of informatics. Some participants advocated for a strategy to support the sector.
But debates mostly centered around copyright and no specific instruments were developed after the meetings.
49
The program outlined six objectives: to promote the use of informatics in the public and private sector; to stimulate
human capital formation; to promote research on informatics; to spur the development of the informatics industry; to
consolidate the juridical dispositions and to foster the development of data networks infrastructure
72
It was only until the arrival of President Vicente Fox from the National Action Party (PAN) in
2000 that a process of adoption and design of a sectoral policy materialized in the Program to
Support the Software Sector (Prosoft).
3. State and Business Characteristics Prior to Prosoft
3.1. Bureaucratic Quality: Highly-skilled Bureaucrats at the Ministry of Economy
In this section I gauge the quality of Mexican bureaucrats in charge of the definition of
policies towards the software sector in the early 2000s. I specifically examine the type and level
of studies they possessed; how they reached the positions they held, their prior experience on the
IT subject and their professional trajectory. The analysis reveals that “pockets of efficiency”
developed in the agencies where the software program was designed.
The public sector organizations in charge of the Information Technologies policies during
the 1990s and early 2000s were: the National Institute for Statistics, Geography and Informatics
(INEGI) and the Ministry of Trade and Industrial Promotion (SECOFI). Inside these agencies there
were attempts to design strategies to facilitate the insertion of Mexico into the global Information
Technologies market and encourage the growth of the domestic software sector.
50
But a more
ambitious and comprehensive effort did not emerge until 2001.
When Vicente Fox launched the initiative to promote the software sector in 2000, the
leadership was bestowed upon SECOFI/Ministry of Economy. There, a small group of highly-
skilled bureaucrats had acquired technical knowledge about the IT sector during the late nineties.
First there was Rocio Ruiz, an economist who had a long trajectory inside the ministry. She entered
SECOFI in 1965 and gradually moved up the ladder. Interviewed people who have worked with
her consider her a civil career servant, even if she did not go through the official process because
in Mexico Civil Service Laws were passed until 2003. Ruiz had close contact with the IT sector
50
Archival documents reveal that top bureaucrats at INEGI were aware of the need for an information technologies
sectoral policy and supported it (Noriega 1992, 3). Even if they had the technical knowledge they lacked the influence
to develop such policies. INEGI bureaucrats had a deep comprehension of the IT trends and possessed relevant studies
and experience on the matter. Noriega, the Director of National Informatics Policy from 1989-1994, held an Actuarial
Sciences bachelor degree and a Masters in Mathematics from University of Wisconsin. He had worked as head of the
artificial intelligence group for two years at IBM Scientific Center in Mexico and as a consultant in Artificial
Intelligence based projects prior to joining the agency. When Noriega left INEGI in 1995 Alfredo Bustos with a
bachelor´s degree in computing and a Ph.D. in statistics by LSE succeeded him.
73
during Zedillo’s administration because she led business-government consultations related to the
Information Technology Agreement (ITA).
51
Two other SECOFI bureaucrats who later on played a central role were Sergio Carrera and
Jesus Orta. Both Carrera and Orta had a Bachelor degree on Economics and a Master’s degree in
Business Administration. Carrera joined SECOFI ranks in 1995, and Orta in 1998 by invitation of
an under-secretary. Around 1999-2000, Carrera and Orta were assigned the task of developing
guidelines for electronic commerce. They delved into the main global trends of the Information
Technologies sector and were well aware of the important role of the software industry. This small
yet highly competent group of bureaucrats would become key players in the design and
implementation of Prosoft.
After the arrival of PAN into power, SECOFI became the Ministry of Economy. With this
reorganization came a replacement of the Secretary in charge of it. Luis Ernesto Derbez substituted
Herminio Blanco.
52
Yet a crucial decision was to keep second and third level bureaucrats in place.
The newly arrived Minister of Economy, Derbez promoted Rocio Ruiz as undersecretary of
Industry and Commerce in 2001. In such a position, she would be in charge of guiding the
industrial upgrading policy process for the software sector.
Rocio Ruiz, assigned Sergio Carrera and Jesus Orta, who were already knowledgeable on
IT-related topics, with the task of designing the sectoral program. Both bureaucrats did not have
prior experience in the private sector, but they were interested in electronic commerce. While
working in a project to stimulate electronic commerce in the former SECOFI, they realized that
the information technologies sector was strategic and something more comprehensive was required
(GVT3 Interview 2016; GVT4 Interview 2016).
Carrera and Orta recruited Claudia Ivette Garcia. Garcia came from the private sector, and
was an IT engineer that had worked in Motorola’s facilities in Puebla. She had some expertise in
designing public programs (incubators for the IT sector) during Vicente Fox’ governorship in
51
Mexico did not join the ITA but unilaterally liberalized many of the sectors (BS3 Interview 2016).
52
Herminio Blanco held a Ph.D. in Economics by the University of Chicago. He was the chief negotiator of NAFTA
from 1990-1993 and head of SECOFI from 1994-2000. Derbez held a Ph.D. in Economics from Iowa State University.
Prior to his appointment, he worked for fourteen years at the World Bank in the structural adjustment area. In 2003
Derbez was succeeded by Fernando Canales Clariond, a lawyer from a prominent business family in the steel sector
that had served as Nuevo Leon’s Governor from 1998-2003.
74
Guanajuato.
53
“It was a great team…within the Ministry of Economy we trained top quality people.
It was a time where top positions changed but the lower ranks rose. The secretary was gone but
the rest kept moving upwards. So we had very skilled cadres.” (GVT1 Interview 2016).
Some bureaucrats that left SECOFI when the PAN administration took over, migrated to
the private sector to organizations related to the IT sector. In fact, a few of them were hired by the
National Chamber for Electronics, Telecommunication and Information Technologies
(CANIETI). Rogelio Garza, who had been personal assistant of the Minister Blanco and Deputy
Director General of International Trade Negotiations at SECOFI, was recruited as head of the
government relations area and later became general director of CANIETI. Alfredo Pacheco former
colleague of Carrera and Orta at SECOFI, was hired by CANIETI too. Others like Graciela
Gutierrez joined the government relations area in IBM’s Mexico office (BS3 Interview 2016).
Thus, the revolving door became an important, though not the main, mechanism for business-
government cooperation.
3.2. Business Cohesion: Well-organized Firms at AMITI and CANIETI
Next, I explore the organization of firms in the software sector and the activities of the
sectoral associations prior to Prosoft’s adoption. On the private sector side, software firms
generated increasing levels of cohesion through the creation and modernization of business
organizations like the Mexican Information Technology Industry Association (AMITI) and
CANIETI. This allowed them to produce relevant information for policy makers and mobilize to
secure resources for the industry.
Mexican business has traditionally displayed high levels of organization (Schneider 2002).
Mainly a legacy of corporatism during the PRI-regime and also as a result of the economic shocks
and subsequent liberalization that took place during the 1980s and 1990s (Pastor and Wise 1994).
Nonetheless, in 1996 the Chambers and Associations law changed making affiliation no longer
mandatory. In the late nineties most encompassing industry organizations weakened (Shadlen
2004) and others reconfigured (Luna, Matilde; Salas-Porras 2012). Despite this general trend, the
main information technologies sectoral associations -- AMITI and CANIETI -- managed to keep
or increase their members and strengthen throughout the past two decades. Other associations
53
Garcia worked for the Ministry of Economy for fourteen years (2001-2014) and then became a consultant to the
World Bank. At the Ministry of Economy, she was promoted as Director of Digital Economy when Carrera left in
2010.
75
related to the IT sector with narrower coverage and more limited impact in the definition and
implementation of public policy are the Mexican Association for Open Source Software
(AMESOL) and National Association for Technology Information Dealerships (ANADIC).
Overall, prior to Prosoft’s launch, the software sector firms had slowly established cohesion, that
is they were well organized and prepared to become an active interlocutor with bureaucrats and
politicians.
AMITI
Software firms, had its first sectoral organization in 1985 with the creation of ANIPCO
which became AMITI in 1996. Some of the founding members were Hewlett-Packard Mexico,
IBM Mexico, Agora Internacional, BMG Entertainment, Consorcio Red Uno, Epson Mexico, EDS
Mexico y Digital Equipment. By 1993 ANIPCO had 165 members (INEGI 1993, 8). In its early
years the association generated a study about the Software Industry in Mexico and pressed for
governmental support to develop software exports: “In Mexico, while the government has had a
lukewarm attitude towards software export, (ANIPCO) has been actively negotiating support and
developing export skills and experience…Among the instruments promoted in Mexico by
ANIPCO there are direct subsidies and loans for participation in software fairs, and some financial
and commercial advise” (Noriega 1993b, 5). ANIPCO also participated in Copyright Law
consultations and the Informatics Program during the Zedillo administration.
To have a more comprehensive coverage of the different types of IT companies ANIPCO
evolved into AMITI in November of 1996.
54
“ANIPCO was too specialized on computer programs
and the Information Technologies industry did not have a representative body. There was
CANIECE, but CANIECE was too focused on the telecommunications and electronic equipment
industries. So, we decided that if we wanted an association that was more representative ANIPCO
had to evolve. That is why we created AMITI, encompassing software and information technology
services” (BS3 Interview 2016).
Shaping industrial policy was a central goal for AMITI since its beginnings: “We do not
aim at making politics for the sake of politics, we want to do industrial policy, to collaborate with
54
According to ANIPCO last president, Jesús de la Rosa, the association was not generating the interactions it should
create amongst firms in the sector (Lopez/Reforma 1996). There was a large turnout at the first AMITI meeting in
1996. “The main room at the Industrial Club in Mexico City was completely full… the largest companies IBM, HP,
Oracle, Unisys, most of them foreign were there, and there were also several Mexican companies… ANIPCO
embraced Mexican software developers, and they were there too. There was a strong support and plenty of resources
for the new association” (BS3 Interview 2016).
76
all other existing chambers to spur the development of the software sector,” declared its first
president, Jesus de la Rosa, who was also IBM’s legal advisor (E. Lopez 1996). Henceforth, the
preparation of policy proposals was one of AMITI’s main tasks.
55
AMITI’s leaders would actively seek to get involved in the design and follow up of
governmental actions: “Our objective is to scrutinize it closely to achieve the objectives that are
beneficial for our sector… the times are over in which the government set up the plans, now the
industry has to do it” (E. Lopez 1998a). Nonetheless, the Mexican government led by neoliberals
had no interest in industrial policy during those years (Moreno-Brid 2013).
Though the exact figures of ANIPCO/AMITI’s membership and its evolution are not
available on an annual basis, available sources reveal that the association gathered a diversity of
firms in terms of size. Also, its membership grew steadily. (See Table 1 below and Appendix 2).
Table 1. Characteristics of AMITI’s Members in 2001
Firm type Size
(Employees)
Average number of
employees (estimated)
Number of Firms
Micro <15 7 63
Small 16 to 100 60 117
Medium 101 to 250 175 14
Large 251 to 1,000 600 11
Corporate More than 1,000 1,500 1
Source: AMITI (2001, 5)
In 1991 it reported 165 members out of 500 estimated firms (INEGI 1993). Ten years later,
in 2001 AMITI had 206 members out of 1,600 firms in the sector (Ministry of Economy). To some
government officials, those figures were as a sign of industry dispersion (Alvarez quoted in
Vizcaino 2001).
56
But compared to membership in other chambers, like CANACINTRA which
55
AMITI staff is small therefore they have relied on the board members’ as well as external consulting firms like
IMCO and Select to draft the policy proposals. AMITI also offers several services to its members like: organization
of workshops and training sessions; export promotion trips; guidance for funding applications to Prosoft and other
public programs; and certification processes.
56
Other accounts mention that the sector had 400 firms. It was in this context that Alvarez Gurza, Bancomext executive
director, made his remarks.
77
represents SMEs from all industries and had a coverage of about 5% of the total firms by the end
of the 1990s (Shadlen 2004), AMITI’s 12% coverage was well above the national trends.
57
Membership diversity is an asset for AMITI: large firms and SMEs, as well as domestic
and foreign owned; hence, the government considers it a representative body. Moreover, AMITI’s
associates generate 70% of total IT sales in Mexico. Since its early days two large sets of members
comprise this organization.
58
The first set is a permanent base of large companies. It is very
difficult that they drop and they have joined little by little. They are approximately 35% of the
membership. The second set is made of small and medium sized firms that usually enroll because
they want to access the services AMITI offers like training, certification or advice for preparing
Prosoft application paperwork. Once they get the benefits they drop. Thus, the attrition level of the
other 65% of the membership is sometimes high.
AMITI’s members also encompass both domestic and foreign owned companies. “Large
MNEs have an interest in developing the industry, developing public policy, training programs
and human capital. And the other firms generally seek specific benefits” (BS1 Interview 2016).
Policymakers at the Ministry of Economy consider that the presence of MNE’s alongside domestic
firms in the negotiation tables of Prosoft was beneficial because they pushed the agenda of
information technologies at the higher government levels and kept pulling SMEs when lobbying
was needed to materialize the program (GVT1 Interview 2016).
Lastly, the organization’s board is composed by the CEOs of the 20 largest domestic and
foreign companies in the sector which facilitates the generation of consensus.
59
“The board meets
every month and its participants cannot send other people to replace them. So it is the cream of the
cream. They also have an offsite yearly session where they do annual planning.” (BS1 Interview
2016). AMITI’s staff has remained small throughout the years, eight persons in total. It does not
have a lobbying area, so the executive director and board members personally meet with
policymakers. AMITI has also developed close coordination with CANIETI, especially when
dealing with legislators.
57
In 2006 AMITI had 260 members (Paramo 2007) from 19 states out of a total of 2,125 firms in the IT sector (INEGI
2004 Economic Census). In the past decade, its membership levels were around 300 members out of 2,336 firms (BS1
Interview 2016; INEGI 2012 Economic Census).
58
Javier Allard has been AMITI’s Executive Director for 15 years. He joined the organization after working for Skytel
(satellite communications company) and IBM.
59
Board members are: Accenture, Alestra, Avantare, Aspel, Brainup2.4. Systems, Cisco, Compusoluciones, HP,
Deloitte, Hitss, IBM Mexico, Indicium, Intel, Indra, Microsoft, Oracle, Prosistemas, SAP Mexico, Softtek, T-Systems.
78
CANIETI
The National Chamber for Electronics, Telecommunications and Technology Information
(CANIETI) is the official sectoral organization therefore, government consults it in matters related
to telecom, electronics and information technology policies.
60
It was created in 1957 as CANIECE
during the corporatist period (1940s-1970s) when the Mexican government fostered business
chambers and implemented Import Substitution Industrialization developmental policies. For
several decades the chamber encompassed radio, television and electronic companies. Until the
late 1990s it neglected the flourishing IT sector. In fact, as was mentioned in the previous section
that was one of the reasons why ANIPCO evolved into AMITI: to give some voice to IT firms that
were not being represented in the official bodies. CANIECE leaders grew aware of the importance
of the information technologies and in 1997 the organization evolved into CANIETI embracing
the new sectors like hardware, software and information technology services.
61
Unfortunately, membership evolution data for CANIETI is incomplete. Some sources
suggest that CANIETI had around 700 members in early 2000s (Osterroth 2002). By 2016 it had
1400 members and IT firms comprise 40% of the total (Email exchange with Ruiz 2016).
62
CANIETI established the Executive Commission for the Promotion and Development of
the Industry in charge of generating the means to negotiate with the government a development
program for the IT sector since 1998 (E. Lopez 1998b). Another of CANIETI’s strategies was to
recruit former policymakers to consolidate its department of government relations. In 2000 when
changes occurred at SECOFI/Ministry of Economy, Rogelio Garza and Alfredo Pacheco former
SECOFI employees, were hired to lead the lobbying efforts (BS5 Interview 2016). When the
government opened a space for policy input, CANIETI was ready to provide it. Additionally, Jesus
60
In Mexico presidents used corporatist laws to create chambers that strongly supported the expansion of the
developmental state. Something similar happened in Brazil (Schneider 2004 28). Membership into the corresponding
sectoral associations was mandatory to all Mexican firms. CANIETI (back then CANIECE) was created by the PRI
government in 1957 as the authorized consultative organization for the radio, TV and electronics sectors. In 1995 it
incorporated the informatics sector and in 1997 it evolved into Information Technologies. Since the Law of Chambers
was passed in 1996, membership to chambers is no longer mandatory. But the legislative still has to consult chambers
for law changes.
61
In 1997 CANIECE became CANIETI incorporating the informatics sector. In 2007 the organization changed its
name to be Camara Nacional de la Industria Electronica, de Telecomunicaciones y Tecnologias de Informacion.
CANIETI was organized into four sections: electronics, telecommunications, information technologies and services.
62
Appendix 3 presents a rough overview of the number of associates with data compiled from annual reports and
written communications with CANIETI staff.
79
de la Rosa, former AMITI president, was invited to lead CANIETI in 2000-2002. This created a
bridge and eased articulation between the two organizations.
AMITI and CANIETI Cohesion
The existence of AMITI and CANIETI facilitated mobilization and pressure when dealing
with government. Overall there has been a collaborative relationship between AMITI and
CANIETI, despite the fact that they both strive to attract firms in the IT sector. In their interviews
AMITI leaders remarked that at some point CANIETI wanted to absorb them, but they resisted.
They claim that CANIETI has more diverse interests because it not only represents the software
sector but also electronics and telecom. Therefore, it is preferable to keep a separate association
that veils for the specific interests of the sector.
Although these two organizations share members, there is also some differentiation and
complementariness in their scope and reach. AMITI is a two-tier organization – big firms aiming
at coordinating and pressing for a national strategy and SMEs looking for specific benefits.
CANIETI had a well-established consultative process with SMEs and a wider national coverage
through regional and local CANIETI’s offices.
63
Additionally, CANIETI’s consultative status with
government and its lobbying department made it a valuable partner for AMITI.
AMITI and CANIETI agreed to join efforts in their pursuit for government support to the
IT sector since 1997. They signed a collaboration agreement in December 1997 to exploit the
latter’ status as official consultation organization (E. Lopez 1997). Both organizations
acknowledged the existence of a National Program for Informatics Development in 1998 but
deemed necessary more specific actions to spur the sector. They agreed to lobby the Executive and
Legislative powers for a program to foster the development of the IT industry in Mexico (E. Lopez
1998b). It would take a few years but their work finally paid off in 2002.
Industry observers claim that prior to Prosoft AMITI and CANIETI did not work so
closely. Carrera (2015, 65) asserts that: “Between 1999 and 2001, AMITI and CANIETI, each
pushed for the establishment of programs or measures to support growth in the IT industry in
63
CANIETI has developed a wide presence throughout the country with five regional offices: Center- for Guerrero,
Puebla, Tlaxcala, State of Mexico, Mexico City; Center-South; Northeast for Nuevo Leon, Coahuila, Tamaulipas, San
Luis Potosi; North got Chihuahua; Northwest for Baja California, Sonora and Sinaloa; West for Jalisco, Nayarit,
Colima, Aguascalientes, Guanajuato; Southeast for Yucatan and Tabasco. It also has representation offices in
Aguascalientes, Campeche, Coahuila Durango, Guanajuato, Guerrero, Mexicali, Sinaloa, Sonora, Tabasco,
Tamaulipas and Quintana Roo. The organization’s nation-wide coverage has been crucial for the operation and
survival of Prosoft.
80
Mexico. The close collaboration between CANIETI and AMITI consolidated when Prosoft
negotiations began. “Prosoft made their agendas converge, they bought the idea of adding vectors
and they have kept up their collaboration. They saw the advantages of being constructive” (GVT4
Interview 2016).
Actors involved explained that in the mid-1990s and early 2000, the IT industry
representation was stronger in AMITI. And CANIETI continued to be very focused in the
telecommunications sector (BS1 Interview 2016; BS2 Interview 2016; BS3 Interview 2016; BS4
Interview 2016; BS5 Interview 2016). After the collaboration agreement was signed in 1997 things
were attuned and collaboration increased gradually. When De la Rosa, former AMITI president,
became president of CANIETI in 2000 he reached out to AMITI and began working on more
projects together (BS3 Interview 2016; BS4 Interview 2016).
Besides CANIETI and AMITI there were also other associations like the AMECE
(Asociación Mexicana de Comercio Electrónico), AMESOL (Asociación Mexicana de Software
Libre) and ANADIC (Asociación Nacional de Distribuidores de Cómputo), but these were
secondary given their narrower focus (hence lower membership levels).
4. The Policy Process: Synergistic State-Business Interactions
This section examines the pattern of state-business interactions that took place during the
industrial upgrading policy process. It focuses on the adoption and design stages. High quality
bureaucrats and well-articulated business sector generated a synergistic process that derived in
the emergence of a Program to Support the Software Industry (Prosoft).
Mexico underwent a historical turning point in 2000. For the first time in 70 years the PRI
lost the presidential election to an opposition party: the PAN, a conservative party with roots in
the business sector. The PAN also achieved a slight majority in the Chamber of Deputies.
64
The
National Development Plan (2001-2006) presented by Vicente Fox’s government included the
software industry as one of twelve strategic sectors. In 2001, an analyst remarked: “The Fox
administration is strongly committed to the promotion of e-commerce and its related information
technologies. This active involvement contrasts with the hands-off policy adopted by the two
64
In 2000, of a total of 500 seats, the PAN held 213 seats, PRI 211, PRD 51. Remaining seats were held by other
small parties.
81
preceding administrations regarding information technology (IT)” (Palacios 2001, 15). For the first
time, the Mexican government made a specific pledge towards the software industry.
The Ministry of Economy (Formerly SECOFI) was assigned the task of developing the
software sector industrial policy. The seed fell in fertile ground, because inside this government
agency and in the private sector, high quality bureaucrats and a well-articulated business sector
were ready to start collaboration. Figure 2 offers a quick overview of the policymaking process
that will be described into detail in the next pages.
Two mechanisms were fundamental in the synergistic policy design and adoption process:
the professionalized collection and assessment of information and the formal collaboration
forums. Prior to the beginning of the policy process, the industry associations had paved the road
for a sectoral program by drafting several policy proposals that were building blocks for the
Ministry of Economy. Bureaucrats collected available information and generated a draft of the
program when President Fox issued the order to support the software sector. Formal discussion
groups to further develop and polish the specific objectives and instruments of the program were
established. When the proposal was ready, bureaucrats designed the rules of operation, that is the
allocation mechanisms.
A last key step was securing a budget because even if the president had launched the
initiative and gave his support to a sectoral policy, the Ministry of Economy alongside CANIETI
and AMITI had to lobby the congress and the Ministry of Finance for the funding. Once they won
it the program was implemented with the help of associations and local economic development
agencies. A selection committee with members from state, business and academia selects the
grantees every year.
82
Figure 2. The Policy Adoption Process of Prosoft
4.1. Mechanism 1: Professionalized Information Collection and Assessment
Prior to the start of the process of policy adoption of Prosoft, during the nineties, industry
associations had prepared and lobbied for sectoral policies, paving the road and providing valuable
input. For instance, in 1999 AMITI met with legislators to highlight the need of policies to support
industrial development and take advantage of business opportunities, like India and Ireland had
done (Aguilar 1999). AMITI also advocated for legislation to regulate e-commerce and in 2000
the Chamber of Deputies unanimously approved the new rules on e-commerce (Garcia and
Taboada 2000).
AMITI issued a document entitled “Scheme for Government Support for the Software
Industry” in 2000 and sent it to President Zedillo as well as to the presidential candidates running
for the 2000 election including Vicente Fox (BS2 Interview 2016; BS4 Interview 2016). In it
AMITI’s members acknowledged that some government support for exporting firms through
Bancomext already existed. Yet they strived for a more comprehensive effort. Amongst the
requests included in the proposal were: to offer financial support to fund the capital needs of IT
firms, funding for the creation of technology centers to encourage quality certification and training.
To increase the promotion of Mexican IT firms in foreign markets. AMITI’s scheme added
conditionality based on two criteria: firm size and quality level. There were some actions targeted
at supporting firms large enough to export and other actions aimed at improving the productivity
Mechanism 1: Professionalized information collection and assessment
1. Industry provided specific data and proposals on what they considered was needed- related to
human capital, funding, quality standards
2. Bureaucrats gathered and assessed the requests. They generated a baseline program.
Mechanism 2: Formal collaboration groups established
3. Business, government and academia met in discussion groups to further develop the specific
objectives of the program.
4. The rules of operation were designed by bureaucrats
5. Business and government jointly lobbied the congress for budget.
6. Every year, associations and local economic development agencies promote the Prosoft funds
7. A state-business-academia selection committee was established in order to select grantees
83
of SMEs in the domestic market.
65
This proposal became a building block for policymakers (GVT2
Interview 2016).
Bureaucrats from the Ministry of Economy led the design stage but carefully considered
private sector input. A long-standing SECOFI bureaucrat, Rocio Ruiz, was promoted as under-
secretary of Industry and Trade and within her responsibilities was the preparation of the program
to promote the software sector. Ruiz, assigned Sergio Carrera and Jesus Orta who were also
knowledgeable on IT related topics with the task of designing the sectoral program.
The two main sources of ideas for developing the program were an analysis of the Indian
case (which included few trips to that country to gather firsthand information) and the policy
proposals that AMITI had presented to President Fox. As one of the main policymakers recalls
“There were some attempts from the industry, AMITI, CANIETI, these associations and chambers
had requested the government to develop a support program. So, we thought, let’s build on that.
Bring them, and start talks, two or three months of ideas and then we were asked to make a formal
program. That is how we drafted Prosoft. First, I drafted an unofficial program titled “Program for
Digital Economy Development” but the Minister of Economy at that time (Derbez) did not approve
it. He thought it was too comprehensive…it encompassed actions to encourage electronic
commerce, the software industry, the use of IT to increase firms’ productivity. Instead he
encouraged us to focus only in the software industry” (GVT2 Interview 2016).
The study and subsequent trips to India allowed bureaucrats to have a clearer sense of the
type of strategies that a successful case had followed. Despite the presence of industrial upgrading
policies in the Latin American region that predated Mexican efforts, like those in Brazil (1997);
Uruguay (1999); Costa Rica (1999) and Chile (2000) these programs were not a reference in the
design of the Prosoft (GVT2 Interview 2016; GVT4 Interview 2016).
4.2. Mechanism 2: Establishment of Formal Collaboration
The Ministry of Economy orchestrated a synergistic process of industrial policy making in
which business and academia participated.
66
On the one hand, Orta and Carrera had a good sense
65
In the past 15 years AMITI has continued to advocate for policies to support the IT sector through documents like:
“Visión Mexico 2020” in collaboration with CANIETI in 2004 and the “Digital National Agenda” in 2010 in
collaboration with CANIETI, AMPICI and ANIEI which served as a base for the Mexican National Digital Agenda
issued by Felipe Calderon’s government in 2011 (Mexican National Digital Agenda 2011). The most recent document
was “Road Map 2025” in 2015. See AMITI Website: www.amiti.org.mx
66
Rocio Ruiz required that the program was designed in consultation with business (GVT1 Interview 2016; GVT2
Interview 2016).
84
of what the public policy should look like because of their prior research on the trends in the IT
sector and the policies that had been implemented in other countries. On the other, business and
academia were invited to join discussion groups to delineate the specific objectives of the program.
In 2002, working groups with business, government and academia representatives were set
up to develop an industry plan. The Ministry of Economy presented a first policy draft that was
enriched by the working tables (GVT4 Interview 2016). A dialogue was established: “We asked
business associations, AMITI and CANIETI, what do you need? And we offered them resources
to meet their demands…and it was not only money, we designed several policies to stimulate the
sector like training, certification programs, clusters” (GVT1 Interview 2016).
From 2002 until July 2003 the Ministry of Economy coordinated 19 working groups with
the participation of 118 representatives of firms, educational institutions, business organizations,
state governments and federal government (Méndez 2003).There were eight strategies and each
strategy was assigned to a particular association or agency that would be in charge of summarizing
the agreements and following up. Policymakers coordinated the logistics. The result was a program
that carefully considered the demands of the sector.
While the content of the Program was widely discussed with the industry through
negotiation tables, the Rules of Operation were specified by bureaucrats from the Ministry of
Economy (GVT2 Interview 2016).67
Their objective was to offer impartial opportunities and reach
all sorts of firms, regardless of their size or ownership origin.
The Executive power was a key player, yet the initiative required the support of another
actor to crystalize the sectoral policy, the Congress. The President and the Minister of Economy
provided the first impulse that was needed to launch the collaboration efforts. And through their
orders, highly skilled bureaucrats were able to design a program in close consultation with a well-
articulated business sector. Even if the order to set up a sectoral policy had come from the President
and was supported by the Ministry of Economy, the program faced a hard time to secure the funds
it needed. A reluctant Minister of Finance and the lack of awareness of the relevance of the
software sector amongst Mexican legislators required further coordination between business and
government.
67
The rules of operation have been subject to external review and have been modified over the years to better serve
Prosoft’s objectives.
85
The government area in charge of allocating the funding for Prosoft was the Legislative,
specifically the Chamber of Deputies. The formal collaboration established between the Ministry
of Economy and the software sector organizations generated trust and common shared goals.
Together they mounted a lobbying campaign to generate awareness and secure the money (BS5
Interview 2016; BS4 Interview 2016). Some legislators supported them but the majority had to be
convinced.
Ministry of Economy policymakers admit the crucial role played by business organizations
in this stage. The conditions were favorable for lobbying. When the PAN won the Presidency, the
traditional party discipline of the PRI-era was relaxed, a plurality of actors, like industry
associations increased their lobbying activities at the congress (Luna, Matilde; Salas-Porras 2012).
For this effort, the Ministry of Economy relied on CANIETI and AMITI’s resources and know-
how (GVT2 Interview 2016; GVT1 Interview 2016).
At the Congress, few legislators were cognizant of the relevance of the software sector in
industrial upgrading efforts. Eloisa Talavera (PAN-BCN) was one of them. She held a bachelor’s
degree in Informatics and represented Baja California Norte, a state where the electronics industry
was relatively strong. The proximity to innovation hubs in California also made her sympathetic
to industry requests. Other legislators like Eloy Cantu Segovia from Nuevo Leon, where the
Monterrey IT cluster is based, also supported the initiative. Having leaders such as Juan Molinar
(PAN-Chih) and Gustavo Madero (PAN-Chih), one coming from academia and the other from the
private sector in the Finance Committee facilitated the process. “They understood what the project
was about and they opened us the door with the Finance committee. They worked with the Finance
Ministry until they approved the fund” (GVT3 Interview 2016). The prior work and the clarity of
the project that the Ministry of Economy put forward was essential. So was the presence of
business organizations like CANIETI and AMITI. “We could team up” remarked a legislator
closely involved in the process. The committee of Science and Technology made a forum to
position the IT industry amongst policymakers. Because not all legislators understood what was it
about.
A group of legislators from the Science and Technology Committee introduced a bill in
December 2003 calling for resources to fund Prosoft. The legislators warned: “Prosoft is a strategic
program to stimulate the IT sector in Mexico. But its goals will be dead letter if the program does
not receive enough financial resources to provide decisive support to the software industry in
86
Mexico” (Cámara de Diputados de México 2003).
68
That year they secured US$12.8 million to
start Prosoft. The next year, in 2004, when the budget was to be lobbied, they had more allies. By
2008 federal funds mounted to US$65 million but given the way the program is structured the total
resources mobilized by Prosoft accrued US$180 million (Brown Grossman and Domínguez
Villalobos 2015). The local component was essential. Governors from states that were benefiting
from Prosoft joined the efforts to increase the program’s budget.
5. From Policy Process to Outcome
5.1. Assessing the Pattern of State-Business Interactions and its Outcomes
The analysis of state and business interactions behind Prosoft, as well as the interviews
with the main actors involved in it, reveal the existence of a synergistic process (GVT1 Interview
2016; BS3 Interview 2016; GVT4 Interview 2016). A top-level policy maker from the Ministry of
Economy asserts: “Business actors jumped in to promote the program decidedly. Without them we
would have not been able to do anything because they were deeply committed…I have not seen a
sectoral program that started from zero and achieved so much with a joint strategy between the
public and private sector. Prosoft will be remembered as a success story because there has not
been, neither before nor afterwards, such a coordination and such growth in the sector, and I know
because I have been involved in sectoral development for a long time” (GVT1 Interview 2016).
Similarly, a bureaucrat that was part of the negotiation team said: “Top level decision makers at
the Ministry of Economy were willing, open to let the policy flow. That was a great advantage.
They helped us with what we proposed. For instance, they received investors, promote the topic
and negotiated for resources. The case is an example of how to make a policy happen. Not
everything was easy. We also faced restrictions. We failed to include tax exemption and there were
very professional discussions but the Ministry of Finance refused to grant fiscal exemptions
because they considered those were not useful” (GVT4 Interview 2016).
Another actor from the Ministry of Economy claimed that “Prosoft had a great success and
created an ecosystem. I do not think that there is a sector, where all actors were aligned: public
and private universities, industry, CANIETI, AMITI, AMIPICI, AMESOL, all were together.
68
The legislators from the Science and Technology Committee were: Eloisa Talavera Hernández(PAN-BC), Julio
César Córdova Martínez (PRI-SON), Víctor Manuel Alcérreca Sánchez (PRI-QR), Omar Ortega Alvarez (PRD-
JAL)(Cámara de Diputados de México 2003).
87
Every actor related to the software industry development was there, aligned, and I think that
allowed the program to grow and increase its budget and impact. And if it did not grow more was
because they (Ministry of Finance) did not want to give it more”. (GVT2 Interview 2016). On a
similar line: “The exchange of information between business and the Ministry of Economy was
possible because there was trust and what was agreed turned into actions” (GVT4 Interview 2016).
The outcome of this synergistic process of business government interactions was a competitive
program. Prosoft’s target population encompasses all sizes of firms. Associations and regional
governments have become sources of support especially for SMEs who do not have the skills, time
and resources necessary to apply. Regarding the allocation criteria: firms apply with a project
during the call for proposals. A board composed of business, government and academia analyzes
the projects and awards the funding.
The neoliberal ideology deeply entrenched in the Ministry of Finance played a decisive
influence in the final design of Prosoft. Despite the fact that both the Ministry of Economy
bureaucrats and sectoral organizations were advocating for fiscal exemptions, they failed to
include them in the industrial policy because the Ministry of Finance was adamant and refused to
grant any kind of widespread prerogative to the firms in the sector (GVT4 Interview 106; GVT2
Interview 2016; GVT1 Interview 2016; BS3 Interview 2016). One of the impacts of this decision
is the somewhat limited size of the program.
Table 2 synthesizes the main features of the Mexican case in terms of prior levels of
bureaucratic quality and business cohesion; the mechanisms underlying the policy process and the
outcomes.
88
Table 2. The Mexican Case at a Glance
Prosoft Policy Process (2000-2003)
Business-
Government
Characteristics
Bureaucratic
Quality
High
Bureaucrats held Economics bachelor’s degree. There was a
career ladder at the Ministry of Economy.
Business
Cohesion
High
Firms were organized into sectoral associations. The
associations were well staffed and had resources. They
represented different types of firms.
Process of
Business-
Government
Relations
Synergistic
Business and government collaborated closely in the
definition of the industry needs and the development of a
program to support firms.
Mechanism 1
Professional
collection and
assessment of
information
Both business and bureaucrats generated their own
information as a first step, then bureaucrats were capable of
conducting independent assessments to select the optimal
policy options.
Mechanism 2
Formal
collaboration
Meetings between business and bureaucrats were
institutionalized. There were round tables for 8 strategic
topics. These groups met following a specific program and
reaching concrete agreements. There was a secretary in each
table that generated a summary of the agreements and was
responsible of the follow up.
Outcome
Adoption
(Yes/No)
Yes
The policy was issued by the executive in 2002 and congress
allocated the funds needed to begin in 2003.
Type of Policy Competitive
Every year a call for projects is issued by the Ministry of
Economy. Any firm in the software sector can submit a
proposal. The proposals are reviewed by a committee
comprised of bureaucrats from the Ministry of Economy,
business sector and academia representatives.
6. Conclusions
This chapter has examined the patterns of business-state interactions leading to the
adoption of industrial upgrading policies and the mechanisms through which bureaucratic quality
and business articulation shape the likelihood of adoption and policy design. The in-depth Mexican
case reveals how a tight group of high quality bureaucrats and a relatively cohesive business sector
89
with clear goals, generated a synergistic process of policy design and adoption. The outcome of
such pattern of collaboration was a program in which funds are allocated through clear criteria and
a competitive process.
A relevant finding is that highly-skilled bureaucrats and a cohesive business sector with
clear goals were critical players in the policy adoption and design process. Yet two other important
actors were: an executive power and the legislators with political will to support the sector. These
two were crucial to launch and materialize Prosoft. In fact, congressional approval was key to
securing the funding needed to implement the program and ensure its survival despite political
alternation. The synergistic pattern of business-government allowed firms and bureaucrats to
overcome opposition from the Ministry of Finance and the lack of awareness inside the congress
to secure the funding needed.
The fact that firms are using business organizations, which are legacies of the corporatist
era and are seen as too rigid by some scholars, might appear counterintuitive in an era of flexibility.
Yet, when targeted towards productive goals, such as increasing firms’ participation in policy
making through more institutionalized channels, their effects can be positive. Prosoft’s design
includes private sector funding so the government is encouraging private investment in a country
where the rates of this type of investment are low. Additionally, business cohesion through
associations can also help overcome the lack of trust amongst different companies and size of
companies, and as trust increases more collaboration for innovation will be achieved.
Another interesting finding is related to the effects of the presence of foreign-owned
multinational companies and large Mexican firms in the software sector. These two actors are seen
as a potential obstacle for industrial upgrading (Schneider 2013a). However, this case study reveals
that incorporating MNEs and large domestic firms into business associations allowed the sector to
have more resources to advocate for government support for all types of firms, including SMEs.
To further refine and test the analytical framework it would be good to undertake more case studies
of the efforts to foster the software sector at the national level in other Latin American countries.
Costa Rica seems interesting as a small economy, lacking a large manufacturing base and lower
levels of business articulation in the software sector. Argentina and Brazil, countries that have also
implemented policies and programs to support the software sector, are two other interesting cases
that merit exploration.
90
An alternative approach is to conduct subnational analysis. This is possible for the Mexican
case because 23 out of 32 states have devised policies to foster the software sector in the past
fifteen years. Chapter 4 presents comparative cases of Nuevo León and Puebla from 2000-2015.
Later on, testing the theory with empirical cases from industrial upgrading efforts in different
sectors such as aerospace or biotechnology and/or innovation policies could also be relevant.
The chapter has contributed to refine the analytical framework of state-business
interactions in democratic settings. And it has also added to the case studies documenting policies
to foster industrial upgrading in Latin America.
69
Overall, more research is required on the political economy of industrial policy and on whether
developing countries, including Latin American ones, can overcome structural limitations, a lack
of fully meritocratic bureaucracies, and actor’s short-term objectives (like raising profits and
winning elections); through state-business collaboration that stimulates more knowledge-intensive
activities to escape the middle-income trap.
69
ECLAC and the Interamerican Development Bank have documented cases of business-government collaboration
in Latin America and/or the emergence of the software sector in the region, but their approach is more empirical,
rather than attempting theory building. For example (Fernández-Arias et al. 2016b; Tigre and Marques 2009).
91
Appendix 1.
List of Interviewees
Nov 2016 GVT1 Subsecretary for Competitiveness. Ministry of
Economy. (2000s). Subsecretary for Industry
and Trade. (2000s) Ministry of
Economy/SECOFI.
Nov 2016 GVT2 Ex Director of Digital Economy Area.
Ministry of Economy (2000s). Sub director
Trade Promotion. SECOFI. (1990s)
Nov 2016 GVT3 Legislator. Mexican Chamber of Deputies.
Science and Technology Committee. (2000s)
Nov 2016
May 2014
GVT4 Ex-Director of Domestic Trade and Digital
Economy. Ministry of Economy/SECOFI.
(late 1990s-2000s)
May 2014 GVT5 Sub director. Domestic IT Market. Ministry
of Economy. (2000s)
Nov 2016 BS1 Director. Mexican Association for
Information Technology Industry (AMITI).
(2000s)
Nov 2016 BS2 Head of the Software Development
Commission of AMITI. (late 1990s-early
200s)
Nov 2016 BS3 Director. Government Relations. IBM Mexico
Ex-president of AMITI (late 1990s);
Ex-president of Chamber for National
Electronics, Telecommunications and
Information Technologies Industries
(CANIETI) (early 2000s)
Nov 2016 BS4 President. CANIETI (2000s).
Vice-President. Corporate Affairs. Hewlett
Packard. (2000s)
Nov 2016 BS5 Executive Director. CANIETI. (2000-2010s)
May 2014 ECLAC Regional joint director. ECLAC Mexico.
(2000s)
92
Appendix 2.
AMITI’s Membership Evolution
1985-2016
Source: Elaborated by the author with information compiled from newspapers and
AMITI’s presentations
Note: In the year 2002, there was a conflict between members of the
telecommunications section that led to the exit of part of its members like Alcatel, Nextel
and Ericsson, and around other 35 more companies. They founded the National Association
for Telecommunications (ANATEL) (Celis/Reforma 2004). But since the association is
organized into four sections, the conflict in the telecommunication sectors, did not cause a
major impact on the IT membership, in fact it might have granted a larger weight to the
sector
93
Appendix 3.
CANIETI’s Membership Evolution
1994-2016
Source: Elaborated by the author with data from CANIETI Annual Reports and Email Exchanges 2017 with
Brenda Ruiz, Affiliation Coordinator in CANIETI.
94
Appendix 4.
CANIETI Membership Distribution by Sections
2013
Source: Elaborated by the author with information from CANIETI Email Exchanges with Ruiz 2017.
This figure shows the distribution of CANIETI affiliates in 2013 according to the four sections.
Information Technologies comprises 41% of the membership; Services firms make 26%,
Electronics 21% and Telecom 12%. This may also reflect the number of firms in the sectors,
because IT is characterized by having a large number of SMEs whereas Telecommunications is
highly concentrated in a few big companies. CANIETI director of external relations declared that
detailed information regarding the distribution of the members by sectors is not available for the
1997-2013 period and for the years 2013-2016 the distribution has remained very stable.
95
Chapter 5.
Industrial Upgrading Policies at the Subnational Level:
The Cases of Puebla and Nuevo León
Existing explanations of the political economy of industrial upgrading mostly focus at the
national level. Meanwhile, in the past two decades developing countries undertook thorough
decentralization measures that generated new policy-making settings with potentially different
actors and goals than those at the national level. Decentralization has also made the implementation
of national level policies more complex because subnational units have a larger leeway in deciding
whether they want to support federal plans or not. Thus, it is important to investigate whether the
processes that lead to the adoption of industrial upgrading policies at the subnational level are
similar to those at the national level and, if they differ, how they differ.
This chapter focuses on industrial upgrading policies at the subnational level and seeks to advance
our understanding of the subject in two ways. First, it offers an additional test of the analytical
framework presented in Chapter 1. Second, it refines this framework by considering the
specificities of the subnational dimension. The guiding research questions are: how does business
cohesion and bureaucratic quality shape the processes of adoption of industrial upgrading
policies at the subnational level? And are there other relevant actors that impact the adoption
and policy design?
The chapter is divided in four sections. The first one presents the argument, methods and
case selection criteria for studying two Mexican states: Nuevo León and Puebla. These cases vary
on the key independent variables: bureaucratic quality and business cohesion. The second part
examines the case of Nuevo León, considering its evolution from 2000 to 2015. The third part
examines the case of Puebla, tracing its evolution also from 2000 to 2015. In each of the case
studies, based on field research, I assess the role of business and government in the policy process
regarding various specific industrial upgrading initiatives and consider how the main explanatory
variables affect the prospects of adoption of industrial upgrading policies. The last section offers
96
a comparison of the cases of Nuevo León and Puebla, and outlines the main findings and
conclusions.
1. Argument, Methods and Case Selection
Decentralization has allowed subnational units to generate their own objectives and
programs so their economic development plans might differ from those devised at the federal
level.
70
As a result subnational actors can influence the adoption and implementation stage of the
national policy making process by obstructing, delaying, or reshaping national policies (Monaldi
2013). Provinces can also choose to build upon the national strategy and expand it according to
their own context.
Furthermore, even if at the national level democracy has been achieved and there are
checks and balances, the provinces of democratic developing countries have varied
democratization levels. In some subnational units, the decision making processes are still heavily
dependent on the executive power and governors have a large influence in the programs adopted
(Gibson 2005; Giraudy 2015; Snyder 2001). Relatedly, subnational bureaucratic structures of
developing countries are more deeply intertwined with political cycles and electoral competition.
Civil service career structures that could generate meritocratic bureaucrats at the subnational level
are weaker than at the federal level (Moscovich 2016; Somuano and Nieto 2017). And since there
are no long-term career prospects, some bureaucrats decide to leave their position to run for a
publicly elected post or job at the federal level (in Mexico these type of bureaucrats are called
“chapulines”- grasshoppers). Likewise, well qualified bureaucrats have to quit when there are
changes in administration, especially if an opposition party wins elections.
Given this context, in the following analysis of the subnational level policy-making
processes I expect that business cohesiveness, as in the federal level, will be critical in the adoption
of programs to support the IT sector. Meanwhile, bureaucratic quality, will still be important.
Executive leadership, defined as the support that the executive, in this case the governor, displays
towards the sector in public acts and official reports, may be a variable that becomes more salient.
70
In most Latin American countries, mayors are now elected rather than being appointed by the national government.
The four federal countries – Argentina, Brazil, Mexico, and Venezuela, along with Colombia, Paraguay and Peru, also
elect regional (provincial) authorities.
97
As has been theorized in Chapter 1, the combinations of business cohesiveness and
bureaucratic quality generate distinct processes of decision making. Here I restate the working
hypotheses stemming from this framework and that will guide the analysis:
Hypothesis 1: Regions with higher levels of bureaucratic quality and higher levels of business
cohesion will develop synergistic policy processes.
Hypothesis 2: Regions with lower levels of bureaucratic quality and lower levels of business
cohesion will experience quiescent policy processes.
Hypothesis 3: Regions with lower business cohesion and higher bureaucratic quality are more
likely to display state-led policy processes.
Hypothesis 4: Regions with higher business cohesion and lower bureaucratic quality are more
likely to display business-led policy processes.
More than looking for a definitive test, my goal is to refine the framework following
Munck’s (2004, 107) advice that hypothesis testing is best seen as an iterative process that interacts
with the development of theory, rather than as a process in which theory is more nearly treated as
static. I undertake an in-depth subnational comparison of two cases in Mexico: Nuevo León and
Puebla. I use process-tracing to examine how the level of business cohesion and bureaucratic
quality shape the industrial upgrading policy processes. The information for the case studies comes
from 15 semi-structured interviews with key actors conducted in Monterrey and Puebla during the
summers of 2012 and 2014.
71
I also collected additional information from primary sources like
official economic development plans and program presentations, organization’s websites, and
secondary sources like local newspapers. Additionally, I subdivide the cases into two periods
generating four causal-process observations (Collier, Brady, and Seawright 2004). This strategy
also helps illustrate the varieties of industrial upgrading trajectories and reveals the dynamic
component of the analytic framework.
71
Appendices 5 and 6 contain the full list of interviewees for the case studies.
98
As Fairfield (2010, 22) notes, whereas systematic cross-case observations buttress
correlation-based causal inference, causal process observations can provide inferential leverage
independently of their relationship to a larger cross-case dataset by providing information about
“context or mechanism” (Collier, Brady, and Seawright 2004, 253). Accordingly, process-tracing
allows the researcher “to identify and analyze the temporal sequence through which hypothesized
explanatory variables affect outcomes” (Munck 2004, 111).
Qualitative methods can also reveal key factors absent in statistical analyses or that are
hard to operationalize. That executive ideology was not a determinant of industrial upgrading
policy adoption in the event-history analysis does not rule out the possibility that the executive
might have an impact in the decision-making process. Thus, I will pay special attention to the role
of the governors’ actions throughout the cases.
Mexico is an ideal country to analyze because the 32 Mexican states have displayed
variegated efforts to establish a program to support the local software industry since the early
2000s. A 2004 assessment revealed that there were three broad categories: 9 regions had set up a
program (Nuevo León, Jalisco, Guanajuato, Puebla, Campeche, Morelos, Sinaloa, Aguascalientes,
Yucatán); 6 had a declared intention to set up one (Baja California, Baja California Sur, Distrito
Federal, Durango, San Luis Potosi, Veracruz); and 17 had not yet shown any interest (Estado de
México, Coahuila, Sonora, Tamaulipas, Chihuahua, Hidalgo, Querétaro, Michoacán, Colima,
Guerrero, Tlaxcala, Chiapas, Quintana Roo, Tabasco, Nayarit, Zacatecas and Oaxaca) (Ruiz Durán
2004, 28). The adoption of Prosoft’s federal program
72
in 2004 has encouraged some of the states
that were missing a plan to develop sectoral policies. By 2008, 23 out of 32 states displayed efforts
to support the software sector.
73
72
“PROSOFT emerged during the 2000-2006 as a Ministry of Economy program to support the software sector. The
normative of Prosoft is defined by high rank bureaucrats from the Ministry of Economy. In between the Ministry of
Economy and recipients are the Promotion Organisms that can be subnational unit governments or business chambers
and associations. Through them, firms present their proposals to compete for funding. Each states starts an evaluation
and approval process by receiving requests, selecting and defining which projects will be granted the funds” (Stezano
and Padilla Pérez 2013, 17)
73
The type of programs adopted in Nuevo León, Jalisco, Guanajuato, Puebla, Campeche, Morelos, Sinaloa,
Aguascalientes, Yucatán differed. Nuevo León and Jalisco had a formal industry program with clear objectives and
economic development institutions in charge. Puebla and Hidalgo had initiatives linked to the expansion of e-
government, that is providing services through the use of internet portals. Aguascalientes and Morelos were
encouraging the emergence of clusters by creating research centers and encouraging business to organize. Meanwhile
Campeche, Sinaloa and Yucatán, had programs under construction (Ruiz Durán 2004).
99
In the following pages, I focus on Nuevo León and Puebla because both had relatively
similar levels of innovation capacity in the early 2000s: a large pool of skilled human resources,
universities with research capabilities (See Table 1 below for details). Also, both states ranked
amongst the top 10 contributors to the national economy in 2000, Nuevo León 2nd and Puebla 9th,
having a predominantly manufacturing economic structure. The two states were eligible for
PROSOFT’s federal funds since the beginning of the program in 2004. In the political realm, both
states were ruled by a PRI governor in 2003. All these similar characteristics allow for the control
of factors not included in the analysis.
Table 1. Innovation Potential of Selected Cases: Nuevo León and Puebla
Nuevo
León
Puebla
National
Total
Sources of
innovative
technological
information
Other
characteristics
Number of Universities (2002)
145 343 7,688
Number of Universities offering IT
Programs (2002)
13 45 531
Students in Computer and Systems
Programs (2001)
9, 267 9, 098 177,110
National Researcher System (SNI)
researchers (2004)
303 455 10,904
Research Centers (2004)
10 6 213
Land (sq. km) 64,220 34,290 1,964
Population (million inhabitants) (1999) 4 5 97
GDP (US billion) (2000) 40 20
581
Source: Elaborated by the author with data from Dutrénit et al. (2013); Medina (2004); Ruiz
Durán (2004)
The selected cases vary in the main independent variables: business cohesion and
bureaucratic quality. To measure business cohesion, I assess whether there was a sectoral
association exclusively for the IT sector; if software firms were organized within a more general
business association; if they had constituted clusters; and the number of members belonging to
these types of associations. Unfortunately, most associations do not have a historical record of the
number of members so I rely on facts obtained from interviews with former directors and
secondary sources. To gauge bureaucratic quality I examine the professional background of the
100
bureaucrats working at the regional Minister of Economic Development, if there was a unit
devoted to knowledge-intensive sectors, and whether the state has enacted a Civil Service Career
Law.
74
In the first period of analysis from 2000-2007 Nuevo León had a well-organized business
sector (medium-high) and medium bureaucratic quality. Meanwhile Puebla had medium-low
levels of business organization and medium bureaucratic quality. By 2008-2015, Nuevo León’s
business sector had a high level of business cohesion, medium-high bureaucratic quality. And
Puebla had a medium level of business cohesion and medium-low bureaucratic quality.
Table 2. Overview of Case Studies and Independent Variables
Cases Nuevo León Puebla
Period 1
2000-2007
Period 2
2008-2015
Period 1
2000-2007
Period 2
2008-2015
Independent
Variables
Level of
business
cohesion
Medium-high High Medium-Low Medium
Bureaucratic
quality
Medium
Medium-high
Medium
Medium-low
While this is a limited test of the theory, it helps validate the importance of business-
government interactions and reveal the differences between the national and the subnational levels.
A next complementary step could be to conduct a quantitative analysis of the determinants of
software policy adoption in the 32 Mexican states. Endogeneity is a threat to validity of this study
because industrial upgrading policy processes may shape the levels of business association but by
using process tracing it is possible to distinguish the sequence in which the relationship between
the main independent variables and the dependent variable appeared.
75
74
Alternatively, I could take as a proxy the Corruption and Good Government Index produced by Transparencia
Mexicana. The index measures the acts of corruption in exchange of public services that people experienced in the
previous year and it is produced for every Mexican state. It comes in a scale ranging from 0 to 100 where the lower
the number the lower the corruption levels.
74
There are observations for 2001, 2003, 2007 and 2010. Nevertheless,
while this is a handy measure because it allows for inter-state comparison, one of the limitations is that the types of
services included are more related to transit, public registry, health, schooling and utilities.
75
Falleti (2006) identifies this type of process tracing as “theory-guided process-tracing” (TGPT) method. She notes
that according to Aminzade (1993: 108), the researcher has to provide “theoretically explicit narratives that carefully
trace and compare the sequences of events constituting the process” of interest… By comparing sequences, we can
101
2. Nuevo León Case: Institutions to Preserve Achievements
Nuevo León’s software industry emerged in the late 1980s and 1990s when local small and
medium enterprises were created to cater the needs of a growing market for IT solutions in the
manufacturing and services sector. Alongside, large Mexican-conglomerates based in Monterrey
like CEMEX, VITRO and ALFA created their own IT areas to facilitate the adoption of frontier
technologies for their operations. As an example, Softtek was founded in 1982. This company
would become one of the few large Mexican software firms with operations at the international
level.
2.1.First Period (2000-2007): Attempts to Encourage the Software Sector and the Prosoft
2.1.1. Business cohesion: Well-organized firms at AETI, CANIETI-Noreste and ANADIC
The software sector in Nuevo León comprised 323 firms in early 2000 (INEGI 2004).
These firms were organized around three main industry associations: ANADIC-Monterrey,
CANIETI-Noreste and AETI. ANADIC-Monterrey (Asociación Nacional de Distribuidores de
Computación) had been created in 1998 and by 2004 it had 55 members that were mostly retailers
of computer equipment and packaged software. CANIETI-Noreste
76
(National Chamber of the
Electronics and Information Technology) opened its regional office in Nuevo León in December
2001 (J. M. Lopez 2001). By 2006 it had 70 members of which 40 were from the IT sector. AETI
(Asociación de Empresarios de Tecnologías de Información) was founded in 2002 by sixteen local
software entrepreneurs and attracted 67 businesses that same year. One of AETI’s first activities
was to generate a census of software firms in Nuevo León in collaboration with the Chamber of
Commerce (CANACO) (Valencia 2002).
Industry leaders belonging to these business organizations inspired by the Indian and Irish
cases, approached the regional government to request support for the sector. During Fernando
Canales’ (1997-2003) administration they contacted the regional Ministry of Economic
Development (SEDEC) and they had some talks regarding the possibilities of collaboration
between firms and government but these were unfruitful (Interview with BSM1 2012). In May
2003 sector associations CANIETI and AETI formally presented their demands to the two main
determine whether there are typical sequences across [cases] … and can explore the causes and consequences of
different sequence patterns” (2).
76
CANIETI was created by the PRI government in 1957 as the authorized consultative organization for the radio
sector. In 1995 it incorporated the electronics sector and in 1997 and IT firms. But for several decades its offices
were centralized in Mexico City.
102
candidates of the gubernatorial race: Natividad González Parás (PRI) and Mauricio Fernández
(PAN) (NA 2003). These lobbying efforts eventually translated into an industry program.
2.1.2. Medium Bureaucratic Quality
The heads of the ministry of Economic Development in Nuevo León, usually stem from
the private sector or have strong ties to the main conglomerates through family kinship. Ministers
are appointed by the governor, usually after consultation with business. Lower ranks change
frequently, especially after shifts at the top level. Regarding qualifications, the type of bureaucrats
working in the Ministry are people holding bachelor´s degree in International Trade, Law,
Economics and International Relations. They have either prior experience in the business sector or
are hired by firms after they finish working at the ministry. No Civil Service Career Law was in
place from 2000-2007.
The Ministry of Economic Development underwent several leadership changes during the
analyzed period which translated into high turnover rates at the lower ranks. When Fernando
Canales was governor the minister of Economic Development was Jorge Arrambide. But in 2003
a new administration took over and a new minister was designated. Eloy Cantú, a Ph.D. in Political
Science and Law headed the ministry from 2003-2005. Cantú left his position to become Senator
for Nuevo León at the Federal Congress and he was replaced by Gustavo Alarcón who only lasted
eight months. The governor substituted Alarcón with Alejandro Páez, a mechanical engineer with
an MBA belonging to the opposition party PAN. Páez had private sector experience and led
SEDEC from 2006-2010. With the changes in the leadership also came several bureaucrat
movements (Olvera 2006). For instance, Edmundo Cepeda who was in charge of Industrial
Groups, software included, worked at SEDEC from 2005-2006 and then migrated to the private
sector.
The creation of an area exclusively devoted to deal with firms in the high technology
sectors called “Direction for the Support to New Economy Firms” around 2005 allowed that at
least some of the people working there were familiar with the issues and processes. The area
comprised a team of six persons and it has survived government changes.
Another factor that has contributed to bureaucratic quality despite the absence of a civil
service career is that the regional Ministry of Economic Development hired Rogelio Flores from
2005-2013 to head the Office for Support to New Economy Firms (Dirección de Fomento a las
Empresas de la Nueva Economía). Flores a lawyer with a master’s degree in public administration
103
by Universidad Autónoma de Tamaulipas, had been the federal Ministry of Economy delegate for
Nuevo León. So, he knew how the federal-subnational levels needed to coordinate. Flores
remained in his position despite government changes generating stability. In an evaluation
conducted during the early years of Prosoft it was deemed that “Flores had an adequate knowledge
of the main objectives of PROSOFT and developing the IT industry was a central objective for
Nuevo León” (UNAM 2006).
2.1.3. Synergistic Policy Processes and Outcomes
In Nuevo León, no policies towards the software sector were in place before 2004. But a
sectoral program was quickly set up between October 2003 and early 2004. The original Monterrey
City of Knowledge project did not envision a specific program to support the software industry
but some software sector leaders approached the governor to request support (Interview BSM1
2012; Interview BSM2 2012). Business cohesion facilitated this encounter and put the sector in a
strong position.
Governor González Parás proposed to set up the advisory council integrated by
representatives of the business, government and academia that would devise specific actions to
strengthen this industry. A prominent software industry leader acknowledged that in order to
achieve their goal of an IT sector policy: “Three elements coincided: the project of Monterrey
International City of Knowledge, the idea of creating civic councils that González Parás was
promoting, and our own demands for support towards sectoral development” (Interview BSM1
2012). So, executive leadership was important in the emergence of synergistic cooperation.
Overtime, having a group of bureaucrats with college and even post-graduate education in
the Ministry of Economic Development, several of whom had private sector experience and
holding similar training to their business counterparts eased the emergence of the programs as well
as the advisory council. The creation of the New Economy unit increased the quality of
policymakers related to the IT sector.
Amongst the initiatives that emerged from the synergistic cooperation between the public
and private sectors were: The Nearshore Application Outsourcing program; the creation of an
Advisory Council for Development of the Software Industry; the establishment of a formal
Software Cluster and the construction of a building for IT firms belonging to the Cluster in the
Technology and Innovation Park (PIIT).
104
• The Nearshore Application Outsourcing Program in 2003
A program to stimulate the sector was adopted in 2003-2004 (Ruiz Durán 2004, 32) with
input from the private sector and academia. Named “Nearshore Application Outsourcing”, the
program had very clear objectives: to increase the number of software companies by ten; raise the
sales from 120 to 2000 million dollars by 2010; train software development engineers; and
promote firms’ quality certification processes too. Finally, it envisioned the creation of an
organization to promote Nuevo León’s exports towards the U.S. This program eventually evolved
and was embraced by the C-Soft (Advisory Council for the Development of the Software Industry).
• The Advisory Council for the Development of the Software Industry in 2004
Besides the industry program, the regional government created an institution to channel the
consultation processes with business and academia. The Advisory Council for the Development
of the Software Industry of Nuevo León (Consejo Consultivo Ciudadano para el Desarrollo de la
Industria de Software) was founded in August 2004 and housed at the Ministry of Economic
Development (Gobierno del Estado de Nuevo León 2004). The council had 28 members
representing the regional and federal governments, business and the three largest universities of
the state. The three sectoral associations AETI, CANIETI-Noreste and ANADIC were also
incorporated and Blanca Treviño, Softtek’s CEO was appointed as the coordinator (Sánchez 2004).
The council’s main goal was to promote the development and consolidation of the software
industry in Nuevo León through the analysis of proposals stemming from the public, private and
academic sector.
• Prosoft’s Funding for Regional Projects in 2004
The regional efforts to develop the software sector were boosted when the federal program
Prosoft was allocated funds in 2004. Prosoft had been announced by President Vicente Fox in 2002
but it was granted monies for the first time by the federal congress in the 2004 fiscal budget. When
PROSOFT’s resources were available to states in 2004, Nuevo León had already developed close
business-government collaboration that facilitated the partnership between regional government
and private sector demanded by Prosoft’s rules. Nuevo León opted into Prosoft and became one
of the top three recipient states during the 2004-2015 period. Appendix 1 contains a table with the
total amounts of funds won by Nuevo León and a chart illustrating the composition by participating
sector for the two periods studied in this chapter.
105
• A Building at the Innovation and Research Technology Park (PIIT) in 2006
AETI and ANADIC cooperated and pushed for additional policies towards the sector.
Together they advocated for state-subsidized land for a building that would host a group of firms.
Knowing about the plans to create the Innovation and Technology Transfer Park (PIIT), a group
of businessmen through AETI and ANADIC requested a space there (BSM1 2012). According to
an interviewee (BSM2 2012), the original PIIT project did not consider software companies in it.
But sectoral representatives (including himself) approached the governor and showed their interest
in joining the park. The government agreed to grant them land in the park with two main
conditions: firms would need to register as a cluster and assume the financial costs of constructing
the building. Thus the “Monterrey IT Cluster” was formed with 25 members of AETI and 15
members from ANADIC, who accepted to share the construction costs (MCIC 2007). The federal
PROSOFT program funded the IT infrastructure required by the new facility.
2.1.4. Assessment
This first causal-process observation illustrates how high levels of business cohesion
combined with a medium level of bureaucratic quality, gave way to the emergence of a synergistic
process. The fact that firms were organized into associations and that all the associations
coordinated to lobby the governor made dialogue easier. Medium quality bureaucracy at the
subnational level was also important because the business sector met with a counterpart that was
capable of assessing the proposals and able to coordinate the dialogues that emerged through the
consultative council hosted by SEDEC.
The case also reveals a variable that merits further examination: the level of executive
leadership. A central characteristic of Natividad González Parás mandate starting in 2003 was the
creation of the Advisory Citizen Councils -33 in total-. The councils’ objective was to collect
private sector’ and universities’ input in addressing the state challenges in practically every policy
domain (economic, social, security, transportation, cultural) (González-Aréchiga, Guerrero, and
Alonso 2009, 35). Furthermore, González Parás participated in several public events inaugurating
the operations of the councils, including the Software Council, with a clear message of
collaboration. One possible explanation to this attitude, is that González Parás ruled as a PRI
governor but Nuevo León had already been ruled by the opposition party PAN. So, there was more
pressure to do a good job and attract support from new constituencies like the IT sector.
106
During his term (2003-20), González Paras also launched the project of Monterrey City of
Knowledge (MCK) through a decree that created a public decentralized organization with citizen
participation in 2003. The central goal of MCK was to plan, design and manage strategic programs
for Nuevo León. Since its origins the organization invited members of three sectors: public, private
and academia to work jointly. After some meetings, the group identified five strategic sectors:
Nanotechnology, Biotechnology, Mechatronics, Information Technologies and Communications,
and Health.
77
So perhaps at a deeper level, the degree of political competition between parties was
also a factor that pushed government into synergistic cooperation.
2.2. Second Period (2008-2015): Beyond Prosoft
2.2.1 Business cohesion: Highly Organized Firms
Federal and regional government initiatives towards the software sector increased business
cooperation and this resulted in additional synergistic processes. AETI and CANIETI-Noreste
agreed to expand coordination and spur small and medium enterprises growth through the “cluster”
figure that was being proposed by the regional government in 2004. The Ministry of Economic
Development supported the cluster formation. Jorge Garza Iglesias, CANIETI-Noreste’s president
of that time declared that meetings with Arturo Salcedo, director of Industrial Clusters at SEDEC,
had buttressed industry projects (J. M. Lopez 2004). In fact, one of the main challenges that
business organizations had traditionally faced was the general apathy of entrepreneurs. In 2005
Oscar Balderas then director of CANIETI-Noreste asserted that it was hard to convince firms to
become members of a chamber or association to gain strength and raise their voices: “there is a
lack of collaborative culture amongst them.”
In 2005 the members of the Advisory Council for the Development of the Software
Industry of Nuevo León requested governor González Parás to support their evolution into an
executive council (the C-Soft). Rather than just giving suggestions, they would have their own
staff and achieve tasks to fulfill recommendations. The governor liked the idea and requested
Antonio Zárate (then director of the Monterrey City of Knowledge) to host the initiative. The
council would have its own staff but the provincial government would facilitate them an office, a
77
Amongst the policies adopted in the context of the MCK were: the creation of a mixed fund for scientific research
and development of about US 6 million (70 million pesos); a Research and Technology Innovation Park (PIIT) to link
universities, firms and research centers in 2005; and the creation of industry “clusters” through the regional Ministry
of Economic Development (SEDEC).
107
secretary and for every peso that they spent, the regional government would spend another (BSM1
2012). To collect funds, the C-Soft set an advisory board membership fee. Since then, the C-Soft
was embraced as an important element of the MCK. Its three priorities would be: human capital
formation, stimulate innovation and financing for firms.
When González Parás left office, the C-Soft lost some of its funding so it developed
activities such as seminars and training in which it charged. To date, the C-Soft organizes big
industry events and has developed important collaborative projects for training with local
universities and firms. A government initiated project, had its own dynamic and organizational
structure. Many other informal linkages and exchanges emerged as a result. Universities and firms
have jointly devised training programs; and a group of firms finished the building at the PIIT and
conducted shared projects through the MTY-IT Cluster (BSM2 2012). So, there will probably be
no turning back.
Pre-existing associations, AETI, CANIETI-Noreste and ANADIC became a bridge
between the Software Council and small and medium enterprises. Shortly after the C-Soft was
created, some SMEs owners felt the government would neglect firms that did not belong to the
council (Interview BSM3 2012). But the business associations secured a seat and SMEs belonging
to CANIETI-Noreste are eligible for the benefits offered by C-Soft such as access to its training
programs and events.
78
AETI and CANIETI-Noreste consolidated in 2008. Both organizations signed a merger
agreement and AETI’s president was designated vice-president of the Information Technologies
branch inside CANIETI-Noreste. Former members of AETI automatically became members of
CANIETI-Noreste. The negotiation process took long but it succeeded. They set up a common
agenda and “the fusion increased the business sector leverage vis-à-vis the government.”
(Interview BSM3 2012). CANIETI-Noreste has strived to channel the benefits of the C-soft to
their members which are mostly small and medium sized (Interview BSM3 2012)
79
.
Business associations still play a critical role in gathering industry information and
representing SMEs. CANIETI-Noreste has focused on a more personalized contact with SMEs.
78
Initially, SMEs were confused about the roles that CANIETI-Noreste and the C-Soft had and whether they should
belong to one or the other, but CANIETI-Noreste insisted that they were part of the C-soft and the objective was to
consolidate the industry not to compete.
79
A small firm owner considered that they are not well represented in the C-soft, even if they have a seat through
CANIETI, the government only cares about the big firms. But more research would be needed to confirm if this view
is widespread (BSM5 2012).
108
This organization hired a person to offer member firms advice on PROSOFT’s funding application
process. The chamber closely follows governmental plans to ensure that the Information
Technology sector is kept as a priority. If there are plans to reduce funding they lobby, especially
at SEDEC and CONACYT. Another important role of CANIETI-Noreste is to generate a database
of firms in the sector and fostering monthly networking events. They also sponsored a survey of
wages to offer firms a salary-benchmark and prevent a spiral of competition for workers. Some of
its members have created “empresas integradoras” a figure where a few companies join together
to increase their capacity while serving a large client. An example is Origo (Interview Guzman
2012).
2.2.2. Medium-High Bureaucratic Quality
The profile of the regional ministers of economic development and the lower- ranks remain
the same as in the period 2000-2007. No Civil Service Career law has been adopted in Nuevo
León. However, the existence of the “Office for the Support to New Economy Firms” area has
allowed that newly appointed ministers can acquire a faster understanding of existing programs to
support the software sector. Some of the bureaucrats in lower- ranks have been promoted into
higher positions inside the ministry. For instance, Jorge Escamilla Ramos and Jesus Cantú Rueda,
became directors of the “Office for the Support to New Economy Firms” after working at SEDEC
for a few years.
This assessment of a medium-high quality bureaucracy resonates with a qualitative analysis
of the Mexican states’ implementation of Prosoft conducted in 2012. Industry observers considered
that the staff related to the IT programs had a good amount of experience in the sector and the
procedures that make relationships between actors work (Proyectos Estratégicos Gubernamentales
y de Negocios 2012b, 56). In that evaluation Nuevo León obtained 8 points which was the highest
number assigned to any state (Jalisco was the other state receiving that amount of points and 8
other states received 4 or 5 points). Meanwhile Puebla received 3 points which was the national
mode alongside 16 other states). The lower six states received between 1 and 3 points (Proyectos
Estratégicos Gubernamentales y de Negocios 2012b, 2012c).
2.2.3. Synergistic Policy Processes and Outcomes 2008-2015
Most of the policies developed during the second period of study are the consolidation or
ramification of those initiatives that started in the mid-2000s. Higher levels of business cohesion
and bureaucratic quality have ensured policy continuity. The stability of Prosoft’s funding has also
109
benefited this continuity at the regional level by providing much-needed financial resources for
the operation of the programs.
• Creation of a Training Institute: IDETI in 2008
To increase the human resources available in Nuevo León, the C-Soft created the Institute
for the Development of IT Talent (IDETI) in 2008. IDETI is supported by five universities-
UDEM, UANL, Tec de Monterrey, UR and Tec Milenio, that design and offer programs in
coordination with firms to better meet industry needs (González 2008). Students and professionals
are trained in business intelligence, specific computer programming languages and also in English.
Some of the resources to organize the courses come from Mexico First, a joint World Bank-Federal
Ministry of Economy initiative.
• Construction and Inauguration of the Building at PIIT in 2013
Another result of business-government collaboration was the construction of the IT Cluster
building at the Technology, Innovation and Technology Transfer Park. In it 35 businessmen would
share office space and infrastructure. The regional government donated the land, equipped it with
basic services and negotiated financial support from the federal government through Prosoft
(Ortega 2008). The 14,700 square meters building was inaugurated by governor Rodrigo Medina
and the 38 members of Monterrey IT Cluster in 2013.
• Prosoft Funding Continuity from 2008-2015
Rodrigo Medina (2009-2015) administration supported PROSOFT projects without
interruption. The Ministry of Economic Development in coordination with business associations
offered guidance to firms interested in submitting proposals. As a result, Nuevo León continued
to be a top recipient of federal resources (See Appendices 1 and 2).
2.2.4. Assessment
This second observation of Nuevo León’s case reveals that high business cohesion is an
important determinant of the adoption and continuity of industrial upgrading policies. Bureaucratic
quality increased in this period as a result of the implementation of several initiatives in 2003-2008
and the fact that some of the people at SEDEC were allowed to move up the ladder. At the
subnational level, the leadership of the executive was a critical element in the initial policy making
process. Overtime, the emergence of institutions helped preserve the synergistic business-state
collaboration.
110
Regarding executive leadership, in 2009 there were gubernatorial elections in Nuevo
León.
80
Voters favored continuity and PRI candidate, Rodrigo Medina, who had been endorsed by
González Parás won 49% of the total number of votes followed by the PAN candidate Fernando
Elizondo got 43.4%. Medina appointed Othón Ruiz, former director of the largest Mexican owned
bank in Mexico (Banorte) as head of SEDEC. The new administration faced a learning curve of
understanding of the strategic sectors (including software). But after that phase they had the
governors backing for the C-Soft (Interview BSM1 2012). Executive leadership decreased as
Rodrigo Medina did not actively participate in C-Soft’s activities. His support was rather indirect,
through the provision of funding and making sure that the SEDEC granted support for applying to
PROSOFT’s projects. Perhaps the real test was faced in 2016 because the PRI lost the
gubernatorial elections in 2015 to an independent candidate. But this case shows that the level of
institutionalization of business-government cooperation through the C-soft, has ensured the
survival of the programs despite government changes and political alternation.
81
3. Puebla Case: Lost Opportunities
The origins of the software sector in Puebla go back to more than two decades ago when
firms emerged or arrived to cater the manufacturing firms in the region. The state has an important
agglomeration of automotive and auto-parts firms attracted by the longstanding presence of
Volkswagen. One of the oldest and most important companies for the evolution of the software
sector is the multinational company T-Systems (formerly GEDAS) that set up its Mexican
headquarters in Puebla in 1995. Initially T-Systems provided IT services exclusively to
Volkswagen. But eventually the company developed other customers throughout Mexico and
Latin America (Ruiz-Duran 2002). The rest of the firms are domestically owned small and medium
companies. In 2004 there were approximately 191 firms in the sector (INEGI 2004).
80
The election was surrounded by corruption accusations against González Paras administration and denounces of
incumbent´s use of vote-buying strategies.
81
Since the new governor Jaime Rodriguez Calderon took power the , C-Soft has been actively involved in an initiative
called Industry 4.0 that aims at increasing innovation and the adoption of technology in the manufacturing sectors and
promoting high technology sectors consolidation (Nava 2017). In fact the new administration committed to continue
the programs towards the software industry in the State Development Plan (2016-2021) (Gobierno del Estado de
Nuevo León 2016, 233). The governor also visited India in June 2017 in an effort to attract additional foreign
investment from this country.
111
Puebla is a major university hub with nation-wide recognized public and private
universities such as Benemérita Universidad Autónoma de Puebla (BUAP), Universidad de las
Américas (UDLA), Tecnológico de Monterrey Campus Puebla, Universidad Iberoamericana,
Universidad Autónoma Popular del Estado de Puebla (UAPEP) and Universidad Tecnológica de
Puebla. A leading research center based in the region is the Institute for Astrophysics, Optics and
Electronics (INAOE). The state hosted around 5% of the total Information Technology college-
level students in Mexico, and 45 universities in the state offer IT degrees (Ruiz-Durán 2002).
3.1.First Period (2000-2007): Attempts to Spur the Software Sector and Prosoft’s Adoption
3.1.1. Medium-Low Business Cohesion
Software firms in Puebla displayed a medium-low level of business cohesion in the early
2000s. The only sectoral association was ANADIC-Puebla but its scope was limited because it
only represented the interests of computer equipment distributors. Businessmen recall that a few
software firms tried to organize a branch called ANADIC-Soft but it was a very small group (10
firms). The branch did not consolidate (Interview BSP2 2014). CANIETI could have helped to
coordinate software firms as it did in other regions through its regional offices (Northeast and
Northwest). Nevertheless, CANIETI did not set up one in Puebla.
Instead firms organized inside the encompassing association CANACINTRA-Puebla
82
.
There, software entrepreneurs created a branch called CANACINTRA-TI in the early 2000s to
coordinate their efforts especially in training and later towards the creation of a cluster. Its tasks
were mainly to organize monthly meetings and business trips, lobby the government and generate
an IT-services yearbook.
By 2008 CANACINTRA-TI had about 53 members and 13 of them founded a “cluster”
83
(Reyes 2008a). While this alternative granted firms the possibility of lobbying through a well-
known encompassing association, one of the disadvantages is that they were subject to the policies
and politics of CANACINTRA. Given that Information Technology was an emerging sector, its
voice was weaker than that of other well-established sectors such as textile and automotive. The
number of firms belonging to CANACITRA-TI remained stable and even declined within a decade
because by 2012, the branch had about 45 members (Interview BSP3 2014).
82
An encompassing association is one where members stem from a wide variety of economic activities thus it has a
diversified membership base.
83
Section 3.1.3. presents more information about the creation and evolution of the cluster.
112
3.1.2. Medium Bureaucratic Quality
Puebla’s ministers of Economic Development had private sector experience, and two out
of three of the ministers were lawyers. But none held graduate education or public administration
training. Antonio Zaraín Garcia was SECOTRADE minister from 1999-2005. He was a successful
entrepreneur from the furniture sector. But he had no prior experience in public administration. In
2005 Gerardo Fernández was appointed as minister. He owned a firm in the beverages sector, held
a law degree but no government experience. Fernández first blocked and then fired all the staff
that had worked with his predecessor accusing them of corruption acts which he could never prove
(García 2008).
84
In 2008 Governor Marín replaced Fernández with José Antonio López Malo, a
lawyer who had worked 12 years for SECOTRADE as minister from 2008-2011.
According to available information, no special unit inside the SECOTRADE was created
to stimulate high tech sectors. In fact, most of the programs related to the IT sector were hosted in
other ministries.
3.1.3. State-led Processes and Outcomes
During 2000-2007 a variety of programs to stimulate the software sector were created by
Puebla’s regional government. In some of these programs there were attempts to generate
business-government collaboration but these were rather shallow. When asked about FISEP and
CITIP not all interviewed businessmen had heard about them. Additionally, no special council or
body emerged to institutionalize these efforts to foster public-private cooperation.
• Program to Stimulate the Software Industry (FISEP) in 2002
One of the first initiatives to promote the development of the software industry was set up
unilaterally by the regional government. In 2002 Puebla’s Ministry of Education announced the
Program to Stimulate the Software Industry (Programa Fomento a la Industria del Software,
FISEP). FISEP’s main objective was to train technicians that met the market’ needs of skilled
labor. The program offered grants to university students that participated in a competitive process
every year (Gobierno de Puebla 2002). Microsoft donated MXN $2.5 million (around
US$250,000) for software (Juárez 2002).
84
Amongst the people he fired were the under secretary of Industrial Promotion, the director of Foreign investment
and the director of Industrial Parks.
113
• Center for Innovation and Information Technologies (CITIP) in 2006
Building on FISEP, the Government of Puebla launched the Center for Innovation and
Information Technologies (CITIP) in late 2006. But it is unclear who were the bureaucrats
involved in it. Apparently, it was housed in the Ministry of Education. CITIP was supported by
Microsoft Mexico and the Council for the Development of Industry, Trade and Services of Puebla.
Amongst its goals was to “articulate business, academia, and government to spur innovation and
technology development in the state” (CITIP-FISEP Presentation). CITIP encompassed several
initiatives including training through FISEP, attracting foreign projects through Development
Centers and supporting the professionalization of local IT firms. The center would be the first to
focus on developing software for mobile technologies and provide training, certification and IT
consulting to firms, students and government. There is not much information on how it evolved
since then or who directed it. Apparently the main activity was training (Reyes 2008b)
CITIP disappeared in 2011 with the Moreno Valle administration and some firms that had
collaborated with the project decided to create a “cluster” without regional government
participation or support (Damian 2011).
• Prosoft Funding Supported 2004-2009
Another strategy deployed by Puebla’s government was to co-finance the PROSOFT
federal initiative by contributing with 25% of the resources needed for the projects presented by
firms from Puebla in 2004. In 2006 there was a lack of participation from the regional government
which was quickly criticized by the president of Canacintra-Puebla, Charles Mtanous. Mtanous
publicly requested the state government to keep its funding towards the IT sector and Gerardo
Fernández, from the regional Ministry of Economic Development (SECOTRADE) promised to
pour $20 million pesos that year (Aroche 2007). Since then and until 2010 support was available
every year. Unfortunately, the following administration of Rafael Moreno Valle (2011-2018)
stopped collaboration with Prosoft from 2010-2015 which is the last year for which data is
available.
Despite the existence of Prosoft, no additional public-private collaboration projects were
sponsored by Puebla’s regional government. And business, rather than press for such
collaboration, decided to go alone. In 2010, Reyes Michel, the IT section President at
CANACINTRA, criticized that the lack of governmental support and did not met its promise of
114
building an industrial park dedicated to IT firms. Reyes said such unmet promises impeded growth
of Puebla’s software firms (Damian 2010).
3.1.4. Assessment
This first Puebla case reveals the outcomes of a medium-low cohesion business sector and
a medium-low bureaucratic quality. The process can be barely situated in the State-led policy
process given the various programs fostered by Puebla’s government.
Puebla was ruled by the PRI in the early 2000s when federal policies to stimulate the
software sector were announced.
85
Melquíades Morales (PRI) governed from 1999-2005. And
Mario Marín (PRI) ruled from 2005-2011. Neither Governor Morales nor Governor Marín (2005-
2011) assumed a major coordinating role nor sent a message of collaboration to stimulate the
software sector. Archival searches in Puebla’s newspapers do not reveal special declarations or
commitments towards generating institutions or programs alongside IT industry representatives
either. In reviewing Puebla’s Development Plans for the 2000-2008 period, the software sector
was not considered strategic.
Yet some policy initiatives emerged and both, Morales and Marín, administrations
committed to work with PROSOFT. From 2004-2011 Puebla funded 25% of the total amount of
resources every year except for 2005 (Appendices 1 and 2).
The business sector was incapable of overcoming differences and form a sectoral
organization that could directly voice their demands. Bureaucrats at the Ministry of Economic
Development had private sector experience but they did not display a coordinating capacity. And
the government did not encourage the institutionalization of public-private collaboration. As a
result, State-created programs did not survive the change of administration in 2011. Not even
PROSOFT with a federal component and the involvement of the CANACINTRA-TI.
3.2. Puebla 2 (2008-2015): Business-led Policy Processes
From 2008-2011 Puebla was still ruled by Mario Marín who continued to support
PROSOFT. Yet again, the there was no attempt to formalize business-government collaboration
in the industrial upgrading policies. Most public programs did not survive the change of
administration in 2011. Furthermore, from 2011-2015, when Moreno Valle was governor, Puebla
did not contribute to PROSOFT funds. Only firms and the federal government. Facing this
scenario, some firms have led some initiatives to increase coordination and get federal funding.
85
Despite the fact that elections were relatively close, Puebla had not yet experienced alternation.
115
3.2.1 Medium-high Business Cohesion
Firms in Puebla IT sector have tried to institutionalize their cooperation through the
“cluster” figure but not all of the attempts have been fruitful.
86
Through these clusters, firms
generated governance structures to develop joint projects. Some have included universities too.
Businessmen acknowledged that their main challenge is to increase collaboration levels through
sectoral associations: “In Puebla we need associations like CANIETI that can really help to
develop the sector” (BSP3 2014).
An early effort since 2005 was the cluster TI-Canacintra led by Raul Reyes Mitchel. It was
legally constituted by the 52 members of this section inside Canacintra (Reyes 2008a).
Nevertheless, it did not mature because of low trust levels and a decrease in membership levels
(Interview BSP2 2014, Interview BSP3 2014).
Another failed attempt was the foundation of Puebla en TI led by Daniel Santin. This
cluster grouped a set of small local companies and universities including: Altec Software,
DysoWeb, EduTechnology Consulting, Ersa Consultores, GobPortal.com, GRS Global, Imagen
Tecnológica, Impuestaria, InovaWeb Soluciones, Innova Visión, INNOVATI Consulting Group,
Instituto de Estudios Superiores en Gestión y Administración, Instituto Profesional en Terapias y
Humanidades (IPETH), Instituto Tecnológico Superior de Zacapoaxtla, Instituto Universitario de
Puebla, IPtel, MKT Tecnológico, NetworkCS, Novutek, ProcessNet, Red Line, Universidad
Autónoma de Puebla, VALIT Solutions y WideNetworks (N/A 2011). This cluster was not
recognized by the federal Ministry of Economy and eventually dissolved.
Puebla-TIC was created in 2007 by some members of CANACINTRA-TI and the CITIP
(which was the initiative led by Puebla’s regional government). Despite originally being a
government-business and university, Puebla-TIC has evolved into a firm-university cluster leaving
regional government aside. In 2011, it was legally registered as Cluster Puebla-TIC A.C. The
objective of this formalization was to generate a larger commitment level between members and
participate in federal-government funded initiatives like Prosoft. Nowadays, Puebla TIC has an
office space and a small permanent staff.
86
Theoretically clusters are geographically defined production systems, characterized by a large number of SME firms
involved at various phases in the production of a homogeneous product family. These firms are highly specialized in
a few phases of the production process, and integrated through a complex network of inter-organizational
relationships. But here it means a group of companies in the same sector that formalize their relationships through a
legal organization.
116
3.2.2. Medium-low Bureaucratic Quality
Bureaucratic quality in Puebla had a medium-low level from 2008-2015.
87
Public
employees at SECOTRADE held prior private sector experience and college degrees in economics
and international trade. Some of them even have graduate education in public administration or
had been working in the ministry in previous administrations. But despite having skilled human
resources, these bureaucrats have not developed know-how related to the innovation or high tech
areas. Along similar lines, in a study carried by Stezano and Padilla Pérez (2013, 33) the Ministry
of Competitiveness, Labor and Economic Development (SEDECO) received a medium evaluation.
From 2011 until 2013 SECOTRADE was led by Pablo Rodríguez but he resigned to run
for a congressional seat at the state legislature. Rodríguez came from the private sector and held a
Mechanical Engineering degree from UDLA. During his period, some analysts criticized the fact
that it was understaffed and several of the duties that were under SECOTRADE scope had been
transferred to other ministries. It was during this time where the Prosoft Funds were cancelled (S.
López 2012). According to some businessmen after the exit of Rodríguez “the private sector lost
contact with SECOTRADE” (Hernández 2014).
SECOTRADE lacked a director for the next ten months in 2013. Finally, Michel Chaín,
who was already working at SECOTRADE was named head of the ministry. Chain is an economist
from UNAM holding an International Business masters degree from the University of Groningen.
He had prior experience in government.
88
But again, ten months later Chaín resigned. Antonio Gali
who had been undersecretary of Economic Development and Business Promotion at
SECOTRADE was appointed minister in 2014. Gali López, held a bachelor’s degree in
International Business by UDLA and a masters in Public Policy by Universidad Iberoamericana
de Puebla.
By 2014 only one bureaucrat, Manuel Herrera, who graduated from business
administration and had a masters’ degree in Political Communication and Governance by George
Washington University, was assigned to provide information about Prosoft and refer potential
applicants to the delegate of the federal Ministry of Economy in Puebla. Since the state is not
87
The analysis in this section focuses on the Economic Development structures. In 2011 a Civil Service Career Law
was approved in Puebla. There is not a lot of information about its implementation but we can expect that this is a
long-term transformation rather than a short-term boost to bureaucratic quality.
88
Chain had been undersecretary of Planning at the Finance Ministry of Puebla from 2011 to 2014 and was strategic
manager at Financiera Rural at the federal government from 2008-2010
117
supplementing funds, firms present their projects through other promoting organizations, which
are the official channels through which firms can apply to Prosoft’s federal funds like CANIETI.
Representatives of the private sector in Puebla asserted that “bureaucrats do not have
experience neither with the IT sector nor with the procedures to make the relationship between
different actors work” and “there is a lack of personnel with vision and knowledge to promote the
development of the IT sector” (Proyectos Estratégicos Gubernamentales y de Negocios 2012c,
38).
Analysts also warned about a low capacity to coordinate the private sector and universities.
Stezano and Padilla Pérez noted that in Puebla: “(regional government organizations’) capabilities
are still very weak and respond to an exogenous factor (the increased flow of resources coming
from federal programs) and not from the state-level development of priorities that allow them to
generate their own policies or programs…In Puebla, we observe a low organizational capacity to
diagnose its industrial priorities through studies and diagnoses of innovation capacities and to
develop policy planning with the participation of multiple actors” (2013, 33).
3.2.3. Business-Led Policy Processes and Outcomes
Official SECOTRADE documents show the existence of an IT strategy on paper for the
2011-2017 term. The main objective was to “generate strategic alliances to create high value-added
projects through strengthening the economic infrastructure comprising the following pillars:
human capital, funding, research and development, intellectual property rights, technology
infrastructure and commercialization”. Among the actions planned were: maintaining a public-
private agenda to foster entrepreneurship in the IT sector; promoting associational projects of
innovation and signing agreements for research and development with universities”
(SECOTRADE 2011).
However, field research revealed that none of the actions has been executed. Interviewees
noted that the regional state decided not to invest in the IT sector. There are no newspaper
declarations or other reports where the minister of economic development or the governor backs
up the program. Hard evidence corroborates the neglect of the state level: the regional share of
Prosoft’s funds dropped to zero since the Moreno Valle administration took over in 2011 (see
Appendix 1).
A subnational assessment concluded that the Puebla state government had a “disarticulated
process of definition of sectoral plans between the government and the IT cluster with a limited
118
participation of academia and other private sector actors who develop their planning
independently. There are no shared strategies or goals, linked to the state development plans”
(Proyectos Estratégicos Gubernamentales y de Negocios 2012c, 51).
This view echoes another study that noted that “In Puebla, the state configuration around
innovation and competitiveness activities has been disarticulated…PROSOFT has had little
relevance. After difficulties in the operation of some projects, the state government stopped being
Promoting Organization, and currently (in 2012) CANIETI coordinates the funds of the program
operating in the state” (Stezano and Padilla Pérez 2013).
Similarly, in 2014 a journalist covering Puebla’s IT sector, described the situation as “the
IT sector in Puebla watches from afar government actions, without being participants of any
project that strengthens the firms of this industry and thus the local economy” (Reyes 2014).
In this scenario, the private sector has attempted to organize and lead some initiatives.
Especially through the cluster Puebla-TIC.
• Prosoft Funds Stall
Puebla’s regional contribution to Prosoft’s funding halted from 2011-2015 (See Appendix
1). The official reason is budget pressures. When interviewed, the bureaucrat in charge of funding
at Puebla’s Ministry of Economic Development (SECOTRADE) acknowledged that the state was
not participating in the PROSOFT federal initiative because even if the Information Technology
sector is strategic, in Puebla “it has been very problematic, very politicized” (Interview GVTP1
2014). Furthermore, he considered there is a lack of trust: “We believe that entrepreneurs are not
seeking to grow their business and development, but instead they are only seeking to secure funds
for their own benefits. There has been some support for firms through a CONACYT program
called Stimulus for Innovation” (Interview GVTP1 2014).
• Puebla TIC Office Space and Activities
Puebla-TIC consolidated without state support and achieved the recognition as official
cluster by the National Software Clusters. (V. López 2012) After having temporary offices at Tec
de Monterrey and UMAD, in 2013 Puebla-TIC inaugurated an office space in a modern building
for the organization’s staff. The place also has co-working space and three meeting rooms where
training sessions and other activities are held. A software factory for Puebla was launched as a
project of the Cluster funded by its associates and the federal Ministry of Economy (UMAD 2014).
In 2014 its members were 12 IT companies and two universities.
119
In 2016, Puebla-TIC leaders decried the fact that they were not receiving the promised state
support to build a Technology Park. Zúñiga noted that the construction of the technology complex
Tecnópolis could have contributed to the consolidation of the sector but the Marín administration
(2005-2011) and the Moreno administration (2011-2017) had only “failed attempts” in spite of the
interest of at least 40 firms affiliated at the CANACINTRA to participate in the project. The
Tecnópolis that was in charge of the BUAP University Project was cancelled in 2014 without any
explanation. Another attempt to support the sector was a park at the Tecnológico de Monterrey
campus that originally should have been coordinated by the Marín administration but ended under
control of the private university (Hernández 2016).
3.2.4. Assessment
The second period (2008-2011) reveals how Puebla’s IT sector is not taking advantage of
the federal availability of Prosoft funds because of a low level of executive leadership and a
medium-low bureaucratic quality. Business has attempted to increase its cohesion levels through
the creation of a cluster that is acknowledged as a partner of the federal government but it still not
strong enough to become a generator of more robust programs towards the sector.
In 2010 the president of CANACINTRA-TI, Raul Reyes, rebuked Mario Marín
government for not giving enough support to the IT sector and not meeting its commitment to
create an industrial park dedicated to software. Reyes urged the then elected governor Rafael
Moreno to back the sector (Damian 2010). Theoretically a PAN-governor would have more
affinity with the business sector. But the relationship between business sector organizations and
the regional government actually worsened. Puebla experienced an alternation of power in 2011.
Rafael Moreno (PAN) was elected governor in the 2010 elections with 50% of the votes followed
by the PRI candidate who received 40%.
During Rafael Moreno’s (2011-2017) mandate, the already low levels of formalization of
industrial upgrading policies towards the IT sector have dwindled and so have the possibilities of
moving into a synergistic relationship between state and business. Software sector leaders
interviewed claim that the governor has never received them. Others said there is “moral” support
but no resources because in prior administrations firms did not adequately use public funds.
In sum, despite Puebla’s potential for the software sector it is a case of lost opportunities.
A combination of factors such as medium business cohesion and medium-low bureaucratic quality
have generated no public-private collaboration and at most, a business-led scenario where firms
120
are trying to set up programs for the sector. The situation has been aggravated by the neglect of
the executive, governor Moreno Valle, who has not shown any interest towards the software
industry. A report commissioned by the federal Ministry of Economy in 2012 concluded that “in
Puebla there was an absence of commitment of decision makers at the highest level and limited
communication amongst different actors to achieve goals for the sector” (Proyectos Estratégicos
Gubernamentales y de Negocios 2012c, 51).
89
4. Conclusions
Qualitative analysis of the processes of industrial upgrading policy making towards the
software sector in Mexico reveals that business cohesion does shape the prospects of industrial
upgrading program adoption and the type of policy-making processes. At this level, governor’s
(executive) leadership is relevant, especially because high levels of bureaucratic quality are hard
to achieve.
Bureaucratic quality at the subnational level is harder to achieve than at the federal
level given that in most of the states there is no civil service path. Regional ministers are appointed
by the governor and lower-rank bureaucrats at the subnational level have less stability and long-
term prospects than their federal level counterparts. They often change when a new governor
assumes office.
Overall, the results are in line with the hypotheses derived from the analytic framework.
Despite having similar innovation potential and the opportunity to access federal funding through
Prosoft, Nuevo León and Puebla displayed different trajectories in the policy making processes
towards the software sector from 2000-2007 and 2008-2015 (See Table 3). Nuevo León, started
with a medium-high level of business cohesiveness, medium bureaucratic quality and an executive
that displayed leadership in encouraging innovation intensive activities. In this scenario, business
organizations were able to lobby, be heard and work with a government that created formal spaces
to discuss proposals at the Advisory Council for the Software Industry. The outcome was a
synergistic process that eventually institutionalized, has generated several joint programs and has
survived changes in administration.
89
More generally, private sector leaders from business organizations like Fernando Treviño, the president of the
Consejo Coordinador Empresarial of Puebla, have noted that it “took hard work to establish a relation with governor
Rafael Moreno”. With Antonio Gali, governor since 2016” there is more interaction, he is more accessible” (S. López
2016).
121
On the other hand, Puebla, displayed medium levels of business cohesion, and despite
initially having medium bureaucratic quality, no synergistic processes appeared. The state led
some initiatives hosted in the Ministry of Education but the programs instituted were short-lived.
In the first period 2000-2007 and until 2011, Puebla opted into Prosoft and this federal program
required public-private collaboration. However, neither a wider process nor institutions have been
created to institutionalize public-private dialogue. Governors have not led any important initiatives
for this purpose and bureaucrats have insufficient quality or capacities to do so.
Firms have tried to increase the levels of business cohesion but they have not been very
successful. The creation of the cluster Puebla-TIC moves Puebla into a medium-high level of
business cohesion but it is not likely that the state bureaucratic quality will move into high levels
given the current scenario. As a result, software firms have faced a hard time in convincing an
executive government that does not prioritize the IT sector and a weak bureaucratic body that
doesn’t have enough knowledge, skills nor autonomy to carry forward programs and
institutionalize them.
Table 3. Summary of Case Studies
Cases Nuevo León Puebla
Period 1
2000-2007
Period 2
2008-2015
Period 1
2000-2007
Period 2
2008-2015
Independent
Variables
Level of
business
organization
Medium-
High
High
Medium-
Low
Medium
Level of
bureaucratic
quality
Medium Medium-High Medium
Medium-
Low
Dependent
Variable
Type of
industrial
policy
process
Synergistic
Synergistic
State-Led
Business-
Led
It is important to note that the qualitative measurements of business cohesiveness and
bureaucratic quality based on fieldwork and secondary sources of Nuevo León and Puebla
presented in this chapter, are in line with another study of which I was not aware until recently
122
(Proyectos Estratégicos Gubernamentales y de Negocios 2012a). This finding serves as a
measurement validity test.
By 2008, Nuevo León software cluster ranked 2
nd
and Puebla 15
th
out of 23 in a
competitiveness assessment (Secretaría de Economía 2008). Meanwhile, in terms of the Total
Gross Production of the IT sector, Nuevo León has maintained its 2
nd
place nationwide since 2003,
whereas Puebla has fallen from being the 5
th
largest producer of IT services in 2003 to the 12
th
place in 2013 (See Appendices 3 and 4).
Cross-case comparison between Nuevo León and Puebla reveals that bureaucratic quality
and business cohesion are critical for the type of policy-processes and eventually the policies
adopted. And there are actions that can be taken to improve both of them such as: recruiting people
with background in IT related programs, encouraging business to work together and setting up a
formal council.
Over-time comparison of each case reveals the different dynamics that are possible with
varying bureaucratic quality and business cohesion levels. The cases illustrate – synergistic, state-
led and business-led processes. Additional case studies could be undertaken to see when and how
the quiescent case occurs – perhaps a state like Oaxaca which is often compared with Puebla would
be the most appropriate.
A phenomenon that deserves further exploration is the effect of the presence of large firms,
capital ownership and their role in the policy making processes. In Puebla, the largest company T-
Systems, which is foreign-owned, did not assume a proactive role, whereas in Nuevo León,
Softtek, one of the largest Mexican-owned companies, was a key advocate of government
programs and supported business associations’ petitions.
Another issue that merits careful analysis is how specific attributes of democracy,
especially political competition, alternation and accountability, affect the prospects of industrial
upgrading policies through their impact on the executive and legislative willingness to work with
the private sector and the type of programs adopted.
123
Appendix 1. Prosoft’s Funds in Puebla and Nuevo León (2004-2015). Million Pesos.
Contributor 2004 2005 2006 2007 2008 2009
PUE NL PUE NL PUE NL PUE NL PUE NL PUE NL
PROSOFT 0.53 18.38 3.93 27.2 1.88 58.9 15.79 52.6 15.51 81.52 15.58 53.8
REGION 0.5 13.5 3.37 20 0 34.9 14.45 37.14 11.3 51.6 15.58 30.65
UNIVERSITY 0.3 0.99 0 0 0.39 3.49 0.41 3.21 2.28 4.6 0.42 0
BUSINESS 0.03 19.14 3.09 52.3 6.22 116.9 29.9 92.9 33.73 143.3 29.08 141.85
OTHER 0 0 11.35 2.8 0 8.4 0 0 0 0 0 0
TOTAL 1.36 52.01 21.74 102.3 8.49 222.59 60.55 185.85 62.82 281.02 60.66 226.3
Contributor 2010 2011 2012 2013 2014 2015
PUE NL PUE NL PUE NL PUE NL PUE NL PUE NL
PROSOFT 20 45.39 11.04 92.6 3.63 82.6 7.03 45 7.7 117.27 5.41 59.04
REGION 20 33.6 0 38.7 0 24.1 0 9.11 0 44.5 0 0
UNIVERSITY 3.11 1.43 0 0 0 0 0 0 0 0.94 0 0
BUSINESS 39.59 111.52 30.02 173.22 8.54 145.1 18.68 102.2 17.97 283.96 16.24 177.2
OTHER 0 0 0 0 0 0 0 0 0 0 0 0
TOTAL 82.7 191.94 41.06 304.52 12.17 251.8 25.71 156.31 25.67 446.67 21.65 236.24
Source Prosoft (2017).
*Prosoft’s program was designed so that in every project business contributes with at least
50% of the total budget requested, the federal government with 25% and the state government
with 25%. When the state government does not participate, business has to cover for at least
75% of the project.
124
Appendix 2. Evolution of Prosoft’s Contributions by Actor.
0
50
100
150
200
250
300
PUEBLA NL PUEBLA NL PUEBLA NL PUEBLA NL PUEBLA NL PUEBLA NL
2004-2009
BUSINESS REGION PROSOFT UNIVERSITY
0
50
100
150
200
250
300
350
400
450
PUEBLA NL PUEBLA NL PUEBLA NL PUEBLA NL PUEBLA NL PUEBLA NL
2010-2015
BUSINESS REGION PROSOFT UNIVERSITY
125
Appendix 3. The Software Industry in Mexican States. Key Indicators. 2013
90
90
Similar tables with facts from 2003 and 2008 available upon request.
State Economic
Units
Total
Employees
Total Gross
Production*
Gross Value
Added*
Total
Investment*
Gross Fixed
Capital*
Total National 6,608 228,727 85,659 49,140 2,253 2,242
Mexico City 1,805 98,156 43,790 23,807 942 924
Nuevo León 548 24,644 12,262 7,697 759 762
Jalisco 650 17,743 5,529 3,216 237 242
Estado de Mexico 281 11,402 2,522 1,296 45 45
Querétaro 217 7,757 2,405 1,521 18 18
Coahuila 155 7,020 2,218 1,344 24 24
Sonora 222 9,986 2,155 1,264 8 8
Veracruz 191 3,100 2,023 1,279 12 12
Baja California 255 8,258 1,803 1,337 32 33
Tamaulipas 101 1,854 1,385 627 19 19
Chihuahua 153 6,626 1,231 660 (5) (5)
Puebla 234 4,687 1,115 628 53 54
Guanajuato 231 3,457 829 443 11 11
Campeche 39 1,121 797 642 1 1
Tabasco 59 1,250 629 493 1 2
Morelos 70 1,749 618 458 1 2
San Luis Potosí 103 1,427 382 211 8 5
Quintana Roo 73 797 332 136 2 2
Hidalgo 43 3,192 304 245 0 0
Sinaloa 174 1,490 295 207 7 8
Yucatan 121 1,651 277 92 6 6
Aguascalientes 102 1,655 246 162 0 0
Baja California Sur 66 736 156 126 3 3
Chiapas 93 434 151 107 22 22
Guerrero 47 506 149 39 (0) 0
Durango 52 1,050 143 107 0 0
Zacatecas 39 327 136 72 18 17
Oaxaca 84 631 113 71 2 2
Michoacán 113 617 93 45 1 1
Colima 43 332 46 31 2 2
Tlaxcala 14 90 17 12 3 3
Nayarit 0 0 0 0 0 0
126
Appendix 4. Software Industry in Mexican States Ranking
(By Total Gross Production)
State Ranking
2003
Ranking
2008
Ranking
2013
Mexico City 1 1 1
Nuevo León 2 2 = 2 =
Jalisco 7 3 3
Estado de Mexico 3 4 4
Querétaro 15 6 5
Coahuila 8 12 6
Sonora 9 11 7
Veracruz 10 10 8
Baja California 11 9 9
Tamaulipas 6 7 10
Chihuahua 4 5 11
Puebla 5 8 ⇓ 12 ⇓
Guanajuato 14 14 13
Campeche 16 20 14
Tabasco 19 23 15
Morelos 13 15 16
San Luis Potosí 18 27 17
Quintana Roo 12 16 18
Hidalgo 24 26 19
Sinaloa 20 18 20
Yucatán 17 13 21
Aguascalientes 23 19 22
Baja California Sur 25 17 23
Chiapas 29 22 24
Guerrero 28 24 25
Durango 31 21 26
Zacatecas 21 30 27
Oaxaca 27 28 28
Michoacán 22 25 29
Colima 26 31 30
Tlaxcala 32 29 31
Nayarit 30 32 32
Source: Elaborated by the author with data from INEGI Census.
127
Appendix 5.
List of Interviews Monterrey.
July 2012
Interviewee Position Organization Type of
actor
BSM1 President Nuevo León Software
Advisory Council
Private
Sector
Also Vice-
president of
National Clusters
Council
BSM2 Executive
Director
Monterrey IT Cluster Private
Sector
Also former
President of
CANIETI-Noreste
(in 2005)
BSM3 Executive
Director
Canieti- Noreste Private
Sector
BSM4 President AETI Private
Sector
BSM5 President Origo (Software
Integrator of SMEs)
Private
Sector
GVTM1 Director of
innovation
Innovation and
Technology Transfer
Institute, Nuevo León
Government
Regional
government
UNI1 Vice director of
Talent
Development
Center for the
Development of the
Software Industry,
Universidad Autónoma
de Nuevo León
University
128
Appendix 6.
List of Interviews Puebla
May 2014
GVTP1 Coordinator
of Funds
Management
and
Allocation
Ministry of
Competitiveness,
Labor and Economic
Development
(SECOTRADE)
Regional
government
Current Minister of
SECOTRADE (2017)
BSP1 President Puebla en Ti Private Sector
BSP2 Executive
Director and
Board Vice
President
Cluster Puebla TIC Private Sector Former President of
IT Sector in
CANACINTRA
(2011-2013)
BSP3 Vice-
president of
IT Sector
CANACINTRA Private Sector
UNIP1 Director Puebla Technology
Park. Tec de
Monterrey
University
UNIP2 Director Design and
Technology
Innovation Institute
University /
Federal
government
Former federal
Ministry of Economy
delegate for Puebla
and Tlaxcala (2010-
2013)
GVTF1 Executive
Director
Infotec Federal
government
Former director of
PROSOFT.
GVTF2 Subdirector.
IT Domestic
Market.
DGISCI
Ministry of
Economy
Federal
government
129
Chapter 6.
Conclusions and Research Agenda
Developing countries urgently need to transition into higher innovation activities, amongst
many other challenges they face. Neither state-planned economies, nor unfettered markets, have
proven to be the solution in the past, as innovation reports reveal (Lederman et al. 2014; WIPO,
Cornell University, and INSEAD 2017). Hence, alternatives must be weighed theoretically and
empirically. This dissertation has studied business-government interactions for industrial
upgrading. The main conclusion is that despite prevailing theories and real-life examples of state-
business collaboration ending in rent-seeking (Bates 1981; Irujo and Gil 2017; Krueger 1974,
1990; The Economist 2016), under certain conditions, public-private engagement in the adoption
and design of policies towards high-tech sectors might actually generate positive outcomes.
Research on the developmental state largely focused on economic transformations prior to
the digital era and mostly under authoritarian regimes. Thus, they are insufficient to explain the
efforts to increase knowledge intensive activities in a global era (Amsden 2001; Evans 1995;
Schneider and Maxfield 1997; Wade 1990). More recent studies have focused in a few small, yet
strategic countries like Israel, Taiwan and Ireland (Breznitz 2007; O’Riain 2004). Meanwhile,
business politics analyses (Castaneda 2016; Etchemendy 2011; Fairfield 2015; Özel 2015), have
highlighted how firms actually shape the outcomes of economic policy-making processes such as
structural adjustment, trade liberalization and taxation. But they have paid scant attention to
industrial policy (Schneider 2015).
Seeking to fill these gaps in the literature, this dissertation advances an analytical
framework building on the two bodies of literature: on the developmental state and on business
politics. The framework posits that bureaucratic quality and business cohesion generate diverse
industrial upgrading policymaking processes and outcomes. This dissertation also presents
empirical analyses to test and refine the propositions. The study relies on a mixed-methods
research strategy that conveys the strengths of quantitative and qualitative analyses, and improves
the prospects of making valid causal inference (Collier, Brady, and Seawright 2004; Lieberman
130
2005). On the one hand, statistical analysis is useful to test general hypotheses and relationships
between variables. Meanwhile, qualitative small-N analyses can develop better variable
measurement and generate theoretical insights. More specifically, this dissertation combines: 1) a
cross-national event history analysis of software policy adoption in Latin America (1995-2012);
2) an in depth case study of Mexico, to process trace and find the causal mechanisms underlying
the adoption and design of the Program to Support the Software Sector (1990-2005); and 3) two
Mexican subnational cases, Nuevo León and Puebla during two periods (2000-2007/ 2008-2015).
This chapter concludes this dissertation. Section 1 presents the most important findings of
this dissertation. Section 2 discusses the pending research agenda. Section 3 wraps up with some
concluding remarks.
1. Key Findings
Chapter 2 set up an analytical framework to study industrial upgrading policies in
democratic developing countries. The framework bridges the most valuable ideas and concepts
from the developmental state and business politics literatures. In doing so, it expands available
explanations building a dynamic framework to consider state-business interactions in industrial
upgrading policy making processes. The model posits that the level of bureaucratic quality and the
degree of business cohesion generate four types of policy making processes: synergistic, state-
led, business-led and quiescent. These processes result in various types of policies adopted:
inclusive, competitive, narrow and particularistic.
Chapter 3 quantitatively tests the effects of bureaucratic quality and business cohesion on
the adoption of industrial upgrading policies. Its main contribution is to go beyond existing works
of economic transformation largely based in case studies (Amsden 2001; Breznitz 2007; Evans
1995; O’Riain 2004; Schneider and Maxfield 1997; Wade 1990), and conduct a first cross-country
quantitative study of the phenomenon in the digital era. The event history analysis of the policies
to stimulate the software sector relies on an original dataset that coded the adoption of programs
to support the software sector from 1995-2012 in 17 democratic Latin American countries. The
main finding is that the quality of state bureaucracy and the level of business cohesion do matter
for industrial upgrading policy adoption after controlling for other explanations (human capita,
resource dependence, and globalization). However, this is a very limited sample so testing these
131
hypotheses using a larger set of developing countries is desirable. Another alternative to refine and
go beyond this analysis is conducting a qualitative study.
Thus, chapter 4 unpacks the patterns of business-state engagement through an in-depth case
study of Mexico’s Program to Support the Software Industry (Prosoft). Based on field research the
analysis traces the processes and the causal mechanisms underlying policy adoption and design of
industrial upgrading policies. The case shows how a tight group of high quality bureaucrats and a
relatively cohesive business sector, generated a synergistic process of policy design and adoption.
The outcome of such pattern of collaboration was a program in which funds are allocated through
well-specified criteria and a competitive process.
Another takeaway from Chapter 4 is that in increasingly democratic contexts two important
actors are: the executive power and the legislators. The executive acted as a veto player- one that
either grants support to industrial upgrading or not- and then lets the bureaucrats do their job.
Congressional approval was critical to securing the funding needed to implement the program and
ensure its survival after political alternation.
Finally, the Mexican case also scrutinized the effects of the presence of foreign-owned
multinational companies and large domestic firms in the software sector. These two actors are a
potential obstacle for industrial upgrading (Gallagher and Zarsky 2007; Schneider 2013a).
However, this case study revealed that incorporating MNEs and large domestic firms into the
sectoral associations allowed the software sector to have more resources to advocate for
government support. High bureaucratic quality ensured that support was available for all types of
firms, including SMEs. In Puebla, a case examined in Chapter 5, MNEs were not incorporated into
associations and the results have been disappointing.
Chapter 5 presents a set of comparative cases of Nuevo León and Puebla from 2000-2015
overcoming the traditional approach of studying industrial upgrading at national level. Mexico
represents an interesting laboratory because 23 out of 32 states have devised policies to foster the
software sector in the past fifteen years. Overall, the results are in line with the hypotheses derived
from the analytic framework. Despite having similar innovation potential and the opportunity to
access federal funding through Prosoft, Nuevo León and Puebla displayed different trajectories in
the policymaking processes.
Cross-case comparison and over-time comparison of the cases, highlight the different
dynamics that emerge with varying degrees of bureaucratic quality and business cohesion levels.
132
The cases illustrate the synergistic, state-led and business-led processes with varying policy
outcomes. They also reveal that bureaucratic quality at the subnational level is harder to achieve
than at the federal level. Most Mexican states have no civil service law, or if there is one it is not
enforced. In this scenario, the governor’s (executive) leadership is a factor that becomes more
prominent.
High business cohesion mitigates the lack of bureaucratic quality temporarily. In the
medium run regions need to find ways to increase their institutional capacity. Additional case
studies could be undertaken to see when and how the quiescent scenario occurs – perhaps a state
like Oaxaca which is neighbors with Puebla would be the most appropriate.
Overall, more research is required on the political economy of industrial policy and on
whether developing countries, including Latin American ones, can overcome structural
limitations; a lack of fully meritocratic bureaucracies; and actor’s short-term objectives (like
raising profits and winning elections); to encourage state-business collaboration for economic
transformation.
2. Research Agenda
Quantitative analysis and process-tracing revealed that bureaucratic quality and business
cohesion are proximate causes of the adoption of industrial upgrading policies. An issue that
deserves further inquiry is how the levels of democracy, especially political competition and
alternation, affect the prospects of industrial upgrading policies through their effects on the
executive’ willingness to work with the private sector, and the professionalization of bureaucrats.
To further refine and test the analytical framework it would be good to undertake more
national level case studies of the efforts to foster the software sector in other Latin American and/or
middle-income countries. Argentina and Brazil, have implemented programs to support the
software sector with varied levels of business cohesion. The types of policies adopted differ than
those in Mexico, so these are two other interesting pair of cases that merit exploration. Costa Rica
and Uruguay, are another relevant set. They are both small economies lacking a large
manufacturing base. Costa Rica displays low levels of business articulation in the software sector.
Meanwhile Uruguay has a well-established software sectoral association.
A promising research avenue lies in the study of industrial upgrading policies at the
subnational level. Little is known about the topic in democratic developing countries, especially
133
in Latin America. Therein lies the level in which policy makers and business have a closer
exposure to the realities and challenges of economic transformation.
A phenomenon that deserves additional inquiry is the effect of large firms, their capital
ownership and their role in the policy making processes. This dissertation showed that at the
federal level, large MNCs generated a positive leadership through their participation in sectoral
associations. But at the subnational level in Puebla, the largest company T-Systems, which is
foreign-owned, did not assume a proactive role. Meanwhile in Nuevo León, Softtek, one of the
largest Mexican-owned companies, was a fundamental advocate of government programs and
supported business associations’ petitions. What the Mexico’s federal case and Nuevo León’s
analysis suggest is that sectoral organizations can incorporate large MNCs and push them to start
a dialogue and develop stronger linkages with local SMEs and governments. Something that has
been missing out of the promised benefits of neoliberal policies, and specifically of the inflows of
foreign direct investment (Gallagher and Zarsky 2007; Paus and Gallagher 2008).
Finally, time and resources constraints did not allow me to study the implementation and
monitoring stages of industrial upgrading policies which is an equally pressing issue. In the near
future I would like to answer the following questions: What accounts for the success or failure of
policies and how do actors involved in the adoption and design stages shape the actual outcomes
of the programs once they have been implemented?
3. Concluding Remarks
After so many disappointments with easy recipes for success, like those offered by
neoliberals, there is skepticism at finding one-size fits all recommendations (Rodrik 2007). But a
few policy lessons can be drawn from this dissertation. First and foremost, the analysis underscores
how business cohesion has been a critical factor in the adoption of industrial upgrading policies.
Sectoral organizations even when young or small, can help firms generate proposals and approach
the government in a stronger position. Relatedly, business cohesion can be directly or indirectly
encouraged by governments. In fact, there is a positive reinforcing dynamic in those settings where
the state sets up an official council and invites business associations to collaborate in the policy
decision processes. In the medium run, these kinds of institutional structures survive political
alternation and allow for longer-term policies. After all, innovation is something that takes time.
134
Secondly, rising bureaucratic quality, to increase collaboration with the private sector and
place the state in a better position to avoid capture, is an urgent challenge not only at the federal
level but also, and perhaps more importantly, at the provincial and local levels. Recent approval
of civil service career laws at the subnational level are obviously not enough, the legislation need
to be implemented and enforced.
Business organizations are legacies of the corporatist period so it might appear
counterintuitive to encourage them in a time requiring flexibility and democracy. Yet, when
targeted towards productive goals such as facilitating firm participation in policy making through
a transparent and well-structured mechanism, their effects can be quite positive. Additionally,
business cohesion through associations can also help overcome the lack of trust amongst firms of
different sizes and seems to be a good option to channel MNEs power into formal and more
accountable collaboration mechanisms.
In a digital era where fast technological change is leaving Latin America and many other
developing countries in a middle-income trap; the rules and mechanisms for public-private
engagement need to be reconsidered in order to minimize corruption and maximize the efforts to
spur knowledge intensive activities. State-business collaboration for industrial upgrading requires
well-organized firms and skilled bureaucrats that push for the adoption of programs to support
emerging sectors, like software. These two critical actors can work together to convince other state
actors to support their initiatives and generate institutions that endure political cycles. Otherwise,
the gap between countries producing knowledge-intensive outputs will continue to grow, and the
possibilities for increasing social and economic well-being through innovation will fade away.
135
References
Acemoglu, Daron, and James A Robinson. 2012. Why Nations Fail: The Origins of Power,
Prosperity, and Poverty. New York: Crown Business.
Aguilar, Alberto. 1999. “Nombres, Nombres y...Nombres.” Reforma: Negocios 3A.
Alesina, Alberto, Vittorio Grilli, and Gian Maria Milesi-Ferrett. 1993. The Political Economy of
Capital Controls. National Bureau of Economic Research.
Allison, Graham, and Philip Zelikow. 1971. Essence of Decision. Boston, MA: Little Brown.
AMITI. 2001. Esquema de Apoyo Gubernamental Al Software. Mexico City.
Amsden, Alice H. 1989. Asia’s Next Giant. New York: Oxford University Press.
———. 2001. The Rise of “The Rest”: Challenges to the West from Late-Industrializing
Economies. New York: Oxford University Press.
Aroche, Ernesto. 2007. “Pide Mtanous Tenera La Bolsa Abierta Para Financiar Clúster de
Empresas Informáticas.” La Jornada de Oriente.
Baker, Andy, and Kenneth F. Greene. 2016. “Latin American Election Results with Party
Ideology Scores. 2.0.” http://spot.colorado.edu/~bakerab/VRL Website Data.zip.
Bates, Robert H. 1981. Markets and States in Tropical Africa: The Political Basis of Agricultural
Policies. University of California Press.
Beach, D, and R. B. Pedersen. 2016. “Selecting Appropriate Cases When Tracing Causal
Mechanisms.” Sociological Methods & Research: 1–35.
Beck, Nathaniel, and Jonathan N. Katz. 2011. “Modeling Dynamics in Time-Series–Cross-
Section Political Economy Data.” Annual Review of Political Science 14(1): 331–52.
Beck, Nathaniel, Jonathan N Katz, and Richard Tucker. 1998. “Taking Time Seriously: Time-
Series-Cross-Section Analysis with a Binary Dependent Variable.” American Journal of
Political Science: 1260–88.
Boas, Taylor C., F. Daniel Hidalgo, and Neal P. Richardson. 2014. “The Spoils of Victory:
Campaign Donations and Government Contracts in Brazil.” The Journal of Politics 76(2):
415–29.
Box-Steffensmeier, Janet M, and Bradford S Jones. 2004. Event History Modeling: A Guide for
Social Scientists. Cambridge University Press.
Breznitz, Dan. 2007. Innovation and the State. New Haven: Yale University Press.
Brooks, Sarah M., and Marcus J. Kurtz. 2007. “Capital, Trade, and the Political Economies of
136
Reform.” American Journal of Political Science 51(4): 703–20.
Brown Grossman, Flor, and Lilia Domínguez Villalobos. 2015. “Cadenas de Valor Globales En
Servicios: El Caso de La Industria de TI En México.” Economía: Teoría y Práctica (43):
37–71.
Buchanan, James. 1980. “Rent Seeking and Profit Seeking.” In Toward a Theory of the Rent-
Seeking Society, eds. James M Buchanan, Robert D Tollison, and Gordon Tullock. College
Station: Texas A & M University.
Cali, Massimiliano, and Kunal Sen. 2011. “Do Effective State Business Relations Matter for
Economic Growth? Evidence from Indian States.” World Development 39(9): 1542–57.
Cámara de Diputados de México. 2003. “Punto de Acuerdo, A Fin de Crear Un Fondo Para El
Desarrollo de La Industria de Software.” Sistema de Información Legislativa.
Cammett, Melani. 2007. Globalization and Business Politics in Arab North Africa: A
Comparative Perspective. Cambridge: Cambridge University Press.
Carrera, Sergio. 2015. “La Necesidad de Políticas Industriales En La Economía Moderna. El
Caso Del Fomento a La Industria de Tecnologías de Información En México.” UNAM.
Carter, David B, and Curtis S Signorino. 2010. “Back to the Future: Modeling Time Dependence
in Binary Data.” Political Analysis 18(3): 271–92.
Castaneda, Néstor. 2016. “Business Coordination and Tax Politics.” Political Studies: 1–22.
Castells, Manuel. 2000. The Rise of The Network Society: The Information Age: Economy,
Society and Culture. Oxford: Wiley-Blackwell.
Castro, Claudio. 1996. “Aspectos Del Programa de Desarrollo Informático.” In Foro de Consulta
Sobre Derecho E Informática, Cámara de Diputados.
Chang, Ha-Joon. 2002. Kicking Away the Ladder. London: Anthem.
Cimoli, Mario, Giovanni Dosi, and Joseph E Stiglitz. 2009. Industrial Policy and Development:
The Political Economy of Capabilities Accumulation. Oxford: Oxford University Press.
Collier, David, Henry E Brady, and Jason Seawright. 2004. “Sources of Leverage in Causal
Inference: Toward an Alternative View of Methodology.” In Rethinking Social Inquiry:
Diverse Tools, Shared Standards, Lanham, MD: Rowman and Littlefield.
Corden, W. 1984. “Booming Sector and Dutch Disease Economics: Survey and Consolidation.”
Oxford Economic Papers.
Cox, Gary W., and Mathew D. McCubbins. 2000. “Political Structure and Economic Policy: The
137
Institutional Determinants of Policy Outcomes.” In Presidents, Parliaments and Policy.,
eds. Stephan Haggard and Mathew D. McCubbins. , 21–63.
Culpepper, Pepper. 2001. “Employers’ Associations, Public Policy, and the Politics of
Decentralized Cooperation in Germany and France .” In Varieties of Capitalism: The
Institutional Foundations of Comparative Advantage, eds. Peter A. Hall and David Soskice.
Oxford: Oxford University Press, 275–306.
———. 2011. Quiet Politics and Business Power: Corporate Control in Europe and Japan.
New York: Cambridge University Press.
Dahlström, Carl et al. 2015. “The QoG Expert Survey Dataset II.”
Dahlström, Carl, Victor Lapuente, and Jan Teorell. 2012. “The Merit of Meritocratization.”
Political Research Quarterly 65(3): 656–68.
Damian, Tania. 2010. “El Gobierno No Apoya Tecnologias de La Informacion.” La Jornada de
Oriente.
———. 2011. “Busca TI Crear Parque Tecnológico En Puebla.” El Economista.
Dargent, Eduardo. 2014. Technocracy and Democracy in Latin America: The Experts Running
Government. Cambridge Studies in Comparative Politics.
Diario Oficial de la Federación. 1996. “Programa de Desarrollo Informatico.” Diario Oficial de
la Federacion.
Doner, Richard. 1992. “Toward an Institutionalist View of Development.” World Development
44(3): 398–431.
Doner, Richard, and Ben Ross Schneider. 2000. “Business Associations and Economic
Development : Why Some Associations Contribute More Than Others.” Business and
Politics 2(3): 261–88.
Dutrénit, Gabriela et al. 2013. Ranking Nacional de Ciencia, Tecnología E Innovación.
Capacidades Y Oportunidades de Los Sistemas Estatales de CTI. Mexico City.
Etchemendy, Sebastián. 2011. Models of Economic Liberalization: Business, Workers, and
Compensation in Latin America, Spain, and Portugal. New York: Cambridge University
Press.
Evans, Peter. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton, N.J.:
Princeton University Press.
Fairfield, Tasha. 2010. “The Politics of Revenue-Raising Tax Reform in Latin America.”
138
ProQuest Dissertations and Theses.
———. 2015. Private Wealth and Public Revenue. New York: Cambridge University Press.
Falleti, Tulia G. 2006. “Theory-Guided Process-Tracing in Comparative Politics: Something
Old, Something New.” Newsletter of the Organized Section in Comparative Politics of the
American Political Science Association 17(1): 9–14.
Fernández-Arias, Eduardo, Charles Sabel, Ernesto H Stein, and Alberto Trejos. 2016a. Two to
Tango: Public-Private Collaboration for Productive Development Policies.
Fernández-Arias, Eduardo, Charles Sabel, Ernesto Stein, and Alberto Trejos, eds. 2016b. Public-
Private Collaboration for Productive Development Policies. Washington: Inter-American
Development Bank.
Foxley, Alejandro. 2012. La Trampa Del Ingreso Medio. El Desafío de Esta Década Para
América Latina. Santiago, Chile.
Frenkel, Roberto, and Martin Rapetti. 2011. Fragilidad Externa O Desindustrialización: ¿Cuál
Es La Principal Amenaza Para América Latina En La Próxima Década?
Frieden, Jeffry, and Ernesto Stein. 2000. The Political Economy of Exchange Rate Policy in
Latin America: An Analytical Overview.
Gallagher, Kevin, and L Zarsky. 2007. The Enclave Economy: Foreign Investment and
Sustainable Development in Mexico’s Silicon Valley.
García, Fermín Alejandro. 2008. “Llegó La Hora de La Revancha de José Antonio López Malo.”
La Jornada de Oriente.
Garcia, Georgiana, and Jorge Taboada. 2000. “Fomenta Legislacion Confianza En E-
Commerce.” Reforma.
George, Alexander L, and Andrew Bennett. 2005. Case Studies and Theory Development in the
Social Sciences. MIT Press.
Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge,
MA: Harvard University Press.
Gibson, Edward L. 2005. “Boundary Control: Subnational Authoritarianism in Democratic
Countries.” World Politics 58(1): 101–32.
Giraudy, Agustina. 2015. Democrats and Autocrats: Pathways of Subnational Undemocratic
Regime Continuity within Democratic Countries. Oxford University Press.
Gobierno del Estado de Nuevo León. 2004. Acuerdo Por El Que Se Crea El Consejo Ciudadano
139
de Desarrollo de La Industria de Software. Monterrey.
———. 2016. Plan Estatal de Desarrollo 2016-2021. Monterrey.
González-Aréchiga, Bernardo, Néstor Guerrero, and Sonia Alonso. 2009. “Consejos Ciudadanos
En La Administración Estatal 2003-2009 Para El Estado de Nuevo León.” In Compromiso
Ciudadano: Participación Y Gestión Pública En Nuevo León, ed. Freddy Mariñez.
Monterrey: Fondo Editorial de NL, 35–50.
González, Palmira. 2008. “Nace Centro Para Ingenieros En Software.”
Grossman, Gene M, and Elhanan Helpman. 2001. Special Interest Politics. Boston, MA: MIT
Press Books.
Hacker, Jacob, and Paul Pierson. 2002. “Business Power and Social Policy: Employers and the
Formation of the American Welfare State.” Politics & Society 30(2): 277–325.
Haggard, Stephan. 1990. Pathways from the Periphery: The Politics of Growth in the Newly
Industrializing Countries. Ithaca: Cornell University Press.
Haggard, Stephan, Sylvia Maxfield, and Ben Ross Schneider. 1997. “Theories of Business and
Business-State Relations.” In Business and the State in Developing Countries, eds. Sylvia
Maxfield and Ben Ross Schneider. Ithaca: Cornell University Press, 36–60.
Hall, Peter A., and David W. Soskice. 2001. Varieties of Capitalism. Oxford University Press.
Hancké, Bob. 2009. Debating Varieties of Capitalism: A Reader. Oxford University Press.
———. 2011. no. LSQS Endogenous Coordination: Multinational Companies and the
Production of Collective Goods in Central and Eastern Europe.
Hanushek, Eric A, and Ludger Woessmann. 2012. “Do Better Schools Lead to More Growth?
Cognitive Skills, Economic Outcomes, and Causation.” Journal of economic growth 17(4):
267–321.
Hernández, Miguel. 2014. “Secotrade Tiene Nuevo Titular Después de 10 Meses.” El
Economista.
———. 2016. “Sector TI, Sin Apoyo Gubernamental Para Crecer En Puebla.” El Economista.
Hirschman, Albert O. 1970. 25 Exit, Voice, and Loyalty: Responses to Decline in Firms,
Organizations, and States. Cambridge, MA: Harvard University Press.
Huber, John D, and Charles R Shipan. 2002. Deliberate Discretion?: The Institutional
Foundations of Bureaucratic Autonomy. Cambridge University Press.
INEGI. 1993. La Situación de La Informática En México. Mexico City.
140
International Country Risk Guide (ICRG). 2017. “International Country Risk Guide (ICRG)
Researchers Dataset.”
Irujo, José María, and Joaquín Gil. 2017. “Lawyer at Center of Odebrecht Scandal: ‘The
Company Bribed More than 1,000 People.’” El País.
Jaccard, J, and J Jacoby. 2010. Theory Construction and Model-Building Skills: A Practical
Guide for Social Scientists. New York: Guilford Publications.
Johnson, Chalmers. 1982. MITI and the Japanese Miracle: The Growth of Industrial Policy,
1925-1975. Standford: Standford University Press.
———. 1998. The Idea of Industrial Policy.
Juárez, Ignacio. 2002. “Se Presentó El Programa Fomento a La Industria Del Software.” La
Jornada de Oriente.
Kang, David C. 2002. Crony Capitalism: Corruption and Development in South Korea and the
Philippines. Cambridge University Press.
Karl, Terry Lynn. 1997. The Paradox of Plenty. Berkeley: University of California Press.
Kastner, Scott L, and Chad Rector. 2005. “Partisanship and the Path to Financial Openness.”
Comparative Political Studies 38(5): 484–506.
Khan, Mushtaq, and Stephanie Blankenburg. 2009. “The Political Economy of Industrial Policy
in Asia and Latin America.” In Industrial Policy and Development., eds. Mario Cimoli and
Giovanni Dosi. Oxford: Oxford University Press.
Kingstone, Peter R. 1999. Crafting Coalitions for Reform: Business Preferences, Political
Institutions, and Neoliberal Reform in Brazil. University Park, PA: Penn State Press.
Kohli, Atul. 2004. State-Directed Development: Political Power and Industrialization in the
Global Periphery. New York: Cambridge University Press.
Krueger, Anne O. 1974. “The Political Economy of the Rent-Seeking Society.” The American
Economic Review 64(3): 291–303.
———. 1990. “Government Failures in Development.” The Journal of Economic Perspectives
4(3): 9–23.
Lederman, Daniel, Julián Messina, Samuel Pienknagura, and Jamele Rigolini. 2014. Latin
American Entrepreneurs: Many Firms but Little Innovation. Washington, D.C.
Lieberman, Evan S. 2005. “Nested Analysis as a Mixed-Method Strategy for Comparative
Research.” American Political Science Review 99(3): 435–52.
141
Lopez, Ernesto. 1996. “Impulsaran a La Industria Informatica Nacional.” Reforma.
———. 1997. “Genera Maquila Electrónica Dls. 10 Mil Millones.” Reforma.
———. 1998a. “Apoyará AMITI a Toda Empresa.” Reforma.
———. 1998b. “Entrevista/ Jesus de La Rosa/ Buscan Fincar Desarrollo de Industria
Mexicana.” Reforma.
Lopez, Jose Manuel. 2001. “Refuerzan Combate a Impuesto.” El Norte.
———. 2004. “Acuerdan Acelerar Cluster Tecnológico.” El Norte.
López, Socorro. 2012. “Artículo de Decoración, Pablo Rodríguez Regordosa En La Secotrade.”
Balance. Puebla Online.
———. 2016. “CCE Evitará Empresarios ‘chapulines.’” Balance. Puebla Online.
López, Verónica. 2012. “Logra Puebla Clúster de Tecnologías de La Información Con Aval de
Gobierno Federal.” Milenio.
Luna, Matilde; Salas-Porras, Alejandra. 2012. ¿ Quién Gobierna América Del Norte? Élites,
Redes Y Organizaciones. ed. Alejandra Luna, Matilde; Salas-Porras. Mexico City: UNAM.
Mainwaring, Scott, and Matthew Soberg Shugart. 1997. Presidentialism and Democracy in Latin
America. Cambridge: Cambridge University Press.
Mares, Isabela. 2003. The Politics of Social Risk : Business and Welfare State Development.
Cambridge University Press.
Martin, Cathie Jo, and Duane Swank. 2012. The Political Construction of Business Interests:
Coordination, Growth, and Equality. New York: Cambridge University Press.
Mazzuca, Sebastian. 2015. Economía Política Del Crecimiento. Cadenas Causales Y
Mecanismos Institucionales. Corporación Andina de Fomento.
McCubbins, Mathew D, Roger G Noll, and Barry R Weingast. 1987. “Administrative Procedures
as Instruments of Political Control.” Journal of Law, Economics, & Organization 3(2):
243–77.
Medina, Francisco. 2004. Programa Sectorial de Ciencia Y Tecnología Para El Desarrollo
2004-2009.
Méndez, Verónica. 2003. “Busca México Impulsar Software.” Reforma.
Mochi, Prudencio. 2006. La Industria Del Software En México En El Contexto Latinoamericano
Y Mundial. Mexico City: UNAM.
Monaldi, Francisco. 2013. “Decentralizing Power in Latin America: The Role of Governors in
142
National Policymaking.” In How Democracy Works: Political Institutions, Actors and
Arenas in Latin American Policymaking, eds. Carlos Scartascini, Ernesto Stein, and
Mariano Tommasi. Cambridge, MA: Inter-American Development Bank, 177–216.
Moreno-Brid, Juan Carlos. 2013. “Industrial Policy: A Missing Link in Mexico’s Quest for
Export-led Growth.” Latin American Policy 4(2): 216–37.
Moscovich, Lorena. 2016. “El Doble Estatus de Los Cuerpos Burocráticos Subnacionales: El
Balance Entre Gestión Pública Y Acumulación Política. Propuesta de Análisis Con
Evidencia de Un País Federal.” Revista Sociedad Argentina de Análisis Político. 10(1): 97–
122.
Munck, Gerardo L. 2004. “Tools for Qualitative Research.” In Rethinking Social Inquiry:
Diverse Tools, Shared Standards, eds. David Collier, Henry E Brady, and Jason Seawright.
Rowman & Littlefield, 105–21.
Murillo, Maria Victoria, and Cecilia Martínez-Gallardo. 2007. “Political Competition and Policy
Adoption: Market Reforms in Latin American Public Utilities.” American Journal of
Political Science 51(1): 120–39.
N/A. 2011. “Nace Una Asociación de Empresas Con El Corazón En Puebla.” Pyme Empresario.
http://www.pymempresario.com/empresarios/nace-una-asociacion-de-empresas-con-el-
corazon-en-puebla/.
NA. 2003. “De Regios.” El Norte.
Nava, Rubén. 2017. “Alista NL Proyectos de Industria 4.0.” El Norte.
Nolke, Andreas, and Arjan Vliegenthart. 2009. “Enlarging the Varieties of Capitalism: The
Emergence of Dependent Market Economies in East Central Europe.” World Politics 61(4):
670–702.
Noriega, Pablo. 1992. La Política Informática Nacional.
———. 1993a. On Software Export Opportunities for Development. A Few Comments from a
Mexican Perspective. Mexico City.
———. 1993b. “The Mexican IT Market: Regulatory Environment and Business Opportunities.”
In Informatics in the Americas, La Jolla.
Nugent, Jeffrey B., and Grigor. Sukiassyan. 2009. “Alternative Strategies for Firms in
Oppressive and Corrupt States: Informality or Formality Via Business Associations?”
Contemporary Economic Policy 27(4): 423–39.
143
O’Riain, Sean. 2004. The Politics of High Tech Growth : Developmental Network States in the
Global Economy. Cambridge: Cambridge University Press.
Olson, Mancur. 1965. The Logic of Collective Action. Cambridge: Harvard University Press.
———. 1982. The Rise and Decline of Nations. New Haven: Yale University Press.
Olvera, Silvia. 2006. “Oficializan Cambios En La Sedec.” El Norte.
Öniş, Z. 1991. “The Logic of the Developmental State.” Comparative Politics 24(1): 109–26.
Ornston, Darius. 2012. “Creative Corporatism: The Politics of High-Technology Competition in
Nordic Europe.” Comparative Political Studies 46(6): 702–29.
Ortega, Dolores. 2008. “Avanza El Monterrey IT Cluster.” El Norte.
Osterroth, Maria. 2002. “Defiende La Canieti Su Representatividad.” Reforma.
Özel, Isik. 2015. State–Business Alliances and Economic Development: Turkey, Mexico and
North Africa. London: Routledge.
Palacios, Juan J. 2001. Globalization and E-Commerce: Growth and Impacts in Mexico. Irvine.
Paramo, Leonor. 2007. “Evidencias Del Valor de Las TI En El Desarrollo de La Competitividad
de Mexico.” In Semana de Ingenieria En Computacion, Mexico City: UAM Azcapotzalco.
Pastor, Manuel, and Carol Wise. 1994. “The Origins and Sustainability of Mexico’s Free Trade
Policy.” International Organization 48(3): 459–89.
Paus, Eva. 2014. “Latin America and the Middle Income Trap.” Financing for Development
Series.
Paus, Eva, and Kevin Gallagher. 2008. “Missing Links: Foreign Investment and Industrial
Development in Costa Rica and Mexico.” Studies in Comparative International
Development.
Peres, Wilson. 2009. “The (Slow) Return of Industrial Policies in Latin America and the
Caribbean.” In Industrial Policy and Development: The Political Economy of Capabilities
Accumulation, eds. Mario Cimoli, Giovanni Dosi, and Joseph E Stiglitz. Oxford: Oxford
University Press, 175–200.
Persson, Torsten, and Guido Tabellini. 2000. Political Economics: Explaining Economic Policy.
Cambridge: MIT Press.
Posner, Richard A. 2010. “From the New Institutional Economics to Organization Economics:
With Applications to Corporate Governance, Government Agencies, and Legal
Institutions.” Journal of Institutional Economics 6(1): 1–37.
144
Prebisch, Raúl. 1963. Hacia Una Dinámica Del Desarrollo Latinoamericano. Mexico City:
Fondo de Cultura Económica.
Proyectos Estratégicos Gubernamentales y de Negocios. 2012a. Diálogos Con El Sector de TI
Para Identificación de Estrategias Locales. Nuevo León. Mexico City.
———. 2012b. Diálogos Con El Sector de TI Para Identificación de Estrategias Locales. Nuevo
León. Mexico City.
———. 2012c. Diálogos Con El Sector de TI Para Identificación de Estrategias Locales.
Puebla. Mexico City.
Quinn, Dennis P., and a. Maria Toyoda. 2007. “Ideology and Voter Preferences as Determinants
of Financial Globalization.” American Journal of Political Science 51(2): 344–63.
Rauch, James, and Peter Evans. 1999. “Bureaucracy and Growth: A Cross-National Analysis of
the Effects of‘ Weberian’ State Structures on Economic Growth.” American Sociological
Review 64(5): 748–65.
Reyes, Martha. 2008a. “El Cluster de TI Se Fortalece.” El Sol de Puebla.
———. 2008b. “Software ‘Made in’ Puebla.” El Sol de Puebla.
Rodríguez-Clare, Andrés. 2005. Innovation and Technology Adoption in Central America.
Washington, D.C.
Rodrik, Dani. 2007. One Economics, Many Recipes: Globalization, Institutions, and Economic
Growth. New Jersey: Princeton University Press.
———. 2008. Normalizing Industrial Policy. Washington, D.C.
Roemer, Michael. 1983. “Dutch Disease in Developing Countries: Swallowing Bitter Medicine.”
In The Primary Sector in Economic Development, Routledge.
Rovira, Sebastian, and Giovanni Stumpo, eds. 2013. Entre Mitos Y Realidades: TIC, Políticas
Públicas Y Desarrollo Productivo En América Latina. Santiago: CEPAL.
Ruiz Durán, Clemente. 2004. Potencialidades de Las Entidades Federativas Para Desarrollar
Núcleos de Economía Digital. Mexico City: UNAM.
Sachs, Jeffrey D., and Andrew M. Warner. 2001. “The Curse of Natural Resources.” European
Economic Review 45(4–6): 827–38.
Saeigh, Sebastián. 2010. “Active Players or Rubber Stamps? An Evaluation of the Policymaking
Role of Latin American Legislatures.” In How Democracy Works: Political Institutions,
Actors and Arenas in Latin American Policymaking, eds. Mariano Tommasi, Ernesto Stein,
145
and Carlos Scartascini. IDB, 47–76.
Sánchez-Ancochea, Diego. 2009. “State, Firms and the Process of Industrial Upgrading: Latin
America’s Variety of Capitalism and the Costa Rican Experience.” Economy and Society
38(1): 62–86.
Sánchez, Verónica. 2004. “Impulsarán Industria de Software En NL.” El Norte.
Scartascini, Carlos, Ernesto Stein, and Mariano Tommasi, eds. 2010. How Democracy Works:
Political Institutions, Actors and Arenas in Latin American Policymaking. IDB
Publications.
Schmidt, Vivian. 2009. “Putting the Political Back into Political Economy by Bringing the State
Back in yet Again.” World Politics 61(3): 516–46.
Schneider, Ben Ross. 2002. “Why Is Mexican Business so Organized?” Latin American
Research Review 37(1): 77–118.
———. 2004. Business Politics and the State in Twentieth-Century Latin America. New York:
Cambridge University Press.
———. 2008. Comparing Capitalisms: Liberal, Coordinated, Network, and Hierarchical
Varieties.
———. 2009. “Hierarchical Market Economies and Varieties of Capitalism in Latin America.”
Journal of Latin American Studies 41(3): 553.
———. 2010. “Business Politics and Policymaking in Contemporary Latin America.” In How
Democracy Works: Political Institutions, Actors and Arenas in Latin American
Policymaking, eds. Ernesto Stein, Mariano Tommasi, and Carlos Scartascini. IDB, 217–46.
———. 2013a. Hierarchical Capitalism in Latin America: Business, Labor and the Challenges
of Equitable Development. New York: Cambridge University Press.
———. 2013b. Institutions for Effective of Business-Government Collaboration: Micro
Mechanisms and Macro Politics in Latin America.
———. 2015. Designing Industrial Policy in Latin America: Business-State Relations and the
New Developmentalism. New York: Palgrave Macmillan.
Schneider, Ben Ross, and Sylvia Maxfield. 1997. Business and the State in Developing
Countries. eds. Sylvia Maxfield and Ben Ross Schneider. Ithaca: Cornell University Press.
Schrank, Andrew, and Marcus Kurtz. 2005. “Credit Where Credit Is Due: Open Economy
Industrial Policy and Export Diversification in Latin America and the Caribbean.” Politics
146
& Society 33(4): 671–702.
Seawright, Jason. 2016. Multi-Method Social Science: Combining Qualitative and Quantitative
Tools. Cambridge: Cambridge University Press.
Shadlen, Kenneth C. 2004. Democratization Without Representation. University Park, PA:
Pennsylvania State University Press.
Shafer, D. Michael. 1994. Winners and Losers: How Sectors Shape the Developmental Prospects
of States. Ithaca: Cornell University Press.
Silva, Eduardo. 1998. The State and Capital in Chile: Business Elites, Technocrats, and Market
Economics. Boulder: Westview Press.
Smith, William C. et al. 2014. “Special Section: Political Economy and the Future of Latin
American Politics.” Latin American Politics and Society 56(1): 1–33.
Snyder, Richard. 2001. Politics after Neoliberalism: Reregulation in Mexico. Cambridge
University Press.
Somuano, Fernanda, and Fernando Nieto. 2017. Subnational State Capacity and Civic and
Political Participation in Mexico.
Stezano, Federico, and Ramón Padilla Pérez. 2013. Gobernanza Y Coordinación Entre El
Ámbito Federal Y Estatal En Las Instituciones Y Programas de Innovación Y
Competitividad En México. Banco Interamericano de Desarrollo.
Thacker, Strom C. 2000. Big Business, the State, and Free Trade: Constructing Coalitions in
Mexico. New York: Cambridge University Press.
The Economist. 2016. “Dealing with Murky Moguls.” The Economist 419(8988): 14.
Tigre, Paulo Bastos, and Felipe Silveira Marques, eds. 2009. Desafíos Y Oportunidades de La
Industria Del Software En América Latina. Santiago: CEPAL.
Tsebelis, George. 2002. Veto Players: How Political Institutions Work. Princeton University
Press.
UNAM. 2006. Evaluación Externa Del Programa Para El Desarrollo de La Industria Del
Software. Ejercicio 2006. Mexico City.
Valencia, Verónica. 2002. “Buscan Censar a Empresas de TI.” El Norte.
Vizcaino, Adriana. 2001. “Ven a Mexico Capaz de Exportar Software.” Reforma.
Wade, Robert. 1990. Governing the Market: Economic Theory and the Role of Government in
East Asian Industrialization. New Jersey: Princeton University Press.
147
Weiss, Linda. 2010. “The State in the Economy : Neoliberal or Neoactivist ? Citation : The State
in the Economy : Neoliberal or Neoactivist ?” In Oxford Handbook of Comparative
Institutional Analysis, eds. Colin Crouch, Peer Hull Kristensen, Glen Morgan, and Ove Kai
Pedersen. Oxford: Oxford University Press, 1–26.
Weyland, Kurt. 1997. “‘Growth with Equity’ in Chile’s New Democracy?” Latin American
Research Review 32(1): 37–67.
Whitfield, L, O Therkildsen, L Buur, and AM Kjær. 2015. The Politics of African Industrial
Policy: A Comparative Perspective. Cambridge University Press.
WIPO, Cornell University, and INSEAD. 2017. Global Innovation Index 2017. Ithaca;
Fontainebleau; Geneva.
World Bank. 2017. “World Development Indicators.”
http://databank.worldbank.org/data/home.aspx.
Yeung, Henry Wai-chung. 2014. “Governing the Market in a Globalizing Era: Developmental
States, Global Production Networks and Inter-Firm Dynamics in East Asia.” Review of
International Political Economy 21(1): 70–101.
Abstract (if available)
Abstract
This dissertation studies the adoption and design of industrial upgrading policies in democratic developing countries. In doing so it highlights business-government interactions and their outcomes. The research on the adoption of industrial upgrading policies tackles the following questions: Who are the actors, interests and coalitions that drive the adoption of industrial upgrading policies? Why do some countries (and provinces) adopt industrial upgrading policies and others do not? Concerning the design, I address the questions: What accounts for differences in the characteristics of industrial upgrading programs? How might variation in the quality of government and business cohesion shape the measures adopted? Overall, we need a more comprehensive understanding of the process of industrial upgrading policy underscoring state- business interactions. This is the main contribution of my doctoral dissertation. ❧ Developmental state literature revealed that meritocratic bureaucrats in East Asia were capable of stimulating the transition into higher value-added activities. Yet these works largely focused on economic transformations prior to the digital era and mostly under authoritarian regimes. Thus, they are insufficient to explain the efforts to increase knowledge intensive activities in a global and more open scenario. Some recent studies have focused in a few small, yet strategic countries like Israel, Taiwan and Ireland. Meanwhile, business politics analyses, have highlighted how firms actually shape the outcomes of economic policy-making processes such as structural adjustment, trade liberalization and taxation. But they have paid scant attention to industrial policy in developing countries. ❧ Seeking to fill these gaps in the literature, this dissertation advances an analytical framework that builds on the developmental state and on business politics theories. The framework posits that bureaucratic quality and business cohesion generate diverse industrial upgrading policymaking processes and outcomes. To test and refine the propositions, I conduct a set of empirical studies. The analysis relies on a mixed-methods research strategy that conveys the strengths of quantitative and qualitative techniques. More specifically, this dissertation combines: 1) a cross-national event history analysis of software policy adoption in Latin America (1995- 2012)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The role of local actors in international development: how incentive types affect foreign aid effectiveness
PDF
Politics behind appointments: policy, patronage, and party systems in the Andes
PDF
Economic policy-making in a changing environment: the politics of success in Peru vis-à-vis China
PDF
Social policy and presidential ideology in Latin America: the political economy of social spending and anti-poverty programs
PDF
The murky risk of trade protectionism in an interconnected and uncertain global economy: a state, industrial, and regional analysis
PDF
Power to the masses? The rise of direct democracy in Latin America
PDF
Habits and policy: the socio-cognitive foundations of foreign policy stability
PDF
Budget institutions and the positive theory of fiscal policy
PDF
Financial crises and trade policy in developing countries
PDF
Policies for genetically modified crops in developing countries: the role of non-state actors
PDF
Circuit breakers: how policy entrepreneurs interrupted the electric flow with Peru’s first renewable energy legislation for the grid
PDF
Creating destruction: the political economy of zombie firms
PDF
Essays on political economy and corruption
PDF
Identity salience and political violence: centering social identity in the study of intrastate conflict
PDF
The strategy of firms towards political elites and consumers in polarized western democracies
PDF
Market failure mentality in Japanese industrial policy: case studies of robotics and aircraft industries
PDF
Volatile nationalism: nationalism and its influence on maritime disputes
PDF
The impact of trade liberlaization on firm performance in developing countries -- new evidence from Pakistani manufacturing sector
PDF
The erosion of democracy in the post–Cold War world: understanding the phenomenon and its causes
PDF
International politics and domestic institutional change: the rise of executive war-making autonomy in the United States
Asset Metadata
Creator
Rangel Padilla, Mariana
(author)
Core Title
Developing countries in the digital era: state and business interactions for industrial upgrading
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Political Science and International Relations
Publication Date
11/14/2019
Defense Date
10/06/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
developing countries,Development,industrial policy,Latin America,OAI-PMH Harvest,software,state-business relations,upgrading
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Munck, Gerardo Luis (
committee chair
), Nugent, Jeff (
committee member
), Wise, Carol (
committee member
)
Creator Email
marianarangel@gmail.com,rangelpa@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-454622
Unique identifier
UC11265402
Identifier
etd-RangelPadi-5910.pdf (filename),usctheses-c40-454622 (legacy record id)
Legacy Identifier
etd-RangelPadi-5910.pdf
Dmrecord
454622
Document Type
Dissertation
Rights
Rangel Padilla, Mariana
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 a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
developing countries
industrial policy
state-business relations
upgrading