University of Southern California Dissertations and Theses

The Economic Impact Of Savings And Loan Associations On Residential Construction

(USC Thesis Other)

The Economic Impact Of Savings And Loan Associations On Residential Construction

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Copyright by BASILIOS ESTATHIOS TSAGRIS 1964 THE ECONOMIC IMPACT OF SAVINGS AND LOAN ASSOCIATIONS ON RESIDENTIAL CONSTRUCTION by Basilios Estathios Tsagris A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Economics) June 1964 UNIVERSITY O F S O U TH ER N CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFO RNIA 0 0 0 0 7 This dissertation, written by Bas i lio8 Est a t hi o s Tsagr i s under the direction of hXU...Dissertation Com mittee, and approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillm ent of requirements for the degree of D O C T O R OF P H IL O S O P H Y Date... DISSERTATION COMMITTEE .. 0 Cnairman TABLE OF CONTENTS CHAPTER PAGE PART A. INTRODUCTION I. PURPOSE, SCOPE, AND SIGNIFICANCE OF THE STUDY.......................... 1 Introduction ...................... 1 The Statement of the Problem .... 5 Significance and Scope of the Study . 6 Review of the Literature ......... 8 The Method of Investigation........ 36 Organization of Remainder of the Study.............. 37 PART B. THE DEMAND FOR NONFARM RESIDENTIAL CONSTRUCTION MORTGAGE CREDIT II. THE INFLUENCE OF ECONOMIC BASE CHANGES ON NEW NONFARM RESIDENTIAL CONSTRUCTION ...................... 42 Introduction ...................... 42 Economic Base Theory............ 43 Regional, Selected States, and Selected Metropolitan Area Employment Level Changes Compared with Nonfarm Residential Housing Starts.......................... 47 ill CHAPTER PAGE Regional Responses to Changes in the Econonic Base.............. 48 Housing Starts Response to Economic Base Changes in Selected States . 51 Housing Starts Response to Economic Base Changes in Selected Metropolitan Areas ............ 61 The Relationship of Nonfarn Housing Starts and Personal Incone .... 72 Changes in the Econonic Base of the Sacranento Metropolitan Area Related to Changes in Local Residential Construction Starts . . 77 Enploynent Growth in the Sacranento Econonic Base.................. 78 Sacranento Econonic Base Incone Level Changes.................. 87 Per Capita Incone................ 99 The Inpact of Aerojet-General Corporation on the Sacranento Metropolitan Area Econonic Base . 102 Sunnary............................ 110 iv CHAPTER PAGE III. THE INFLUENCE OF SELECTED DEMOGRAPHIC FACTORS ON NEW NONFARM RESIDENTIAL CONSTRUCTION ...................... Ill Regional Distribution of Population and Now Nonfarn Roaidontial Construction .................. 112 Couparison botwoon the 1950 to 1960 Growth and Distribution of Population and Housing Units in Selected States ................ 116 The Influence of Migration on Housing Unit Growth............ 123 Migration, Enployaent, and Housing Deaand........................ 126 PART C. THE SUPPLY OF REAL ESTATE CREDIT IV. THE INFLUENCE OF THE MONEY RATE AND MONEY AVAILABILITY ON HOUSING STARTS 131 The Cost of Money Influence on New Nonfarn Residential Construction . 133 The Relationship between Money Availability and New Nonfarn Residential Construction ........ 144 V CHAPTER PAGE Comparison of Bond Price Changes and Changes in Residential Construction in Four Geographic Regions .... 148 Comparison of Bond Price Changes and Changes in Residential Construction in Thirteen Metropolitan Areas . . 152 Appendix Note.................. 159 V. MAJOR SOURCES OF RESIDENTIAL REAL ESTATE CREDIT.................. 164 Introduction ...................... 164 The Savings and Loan Industry in the United States .................... 165 Structure and Growth........ 165 Major Assets and Liabilities . . . 169 The California Savings and Loan Industry.................... 176 Structure and Growth........ 176 Major Assets and Liabilities of California Associations ........ 183 Institutional Lenders' Role in Providing Real Estate Credit in California and the Nation.... 188 CHAPTER vi PAGE Institutional Mortgage Lending Activities in the United States . 188 Institutional Mortgage Lending Activities in California .... 195 Trends in Residential Construction Funds Provided by Savings and Loan Associations .................... 199 Savings Associations' Role in Financing Residential Construction in the United States 199 Residential Construction Funds Provided by California Associations .................. 201 Mortgage Credit Availability Iupact on the Number of Housing Starts Financed by Associations .... 204 VI. THE ECONOMIC IMPACT OF SAVINGS AND LOAN ASSOCIATIONS ON RESIDENTIAL CONSTRUCTION IN THE SACRAMENTO AREA . 210 Introduction ...................... 210 The Sacramento Savings and Loan Industry........................ 211 Composition and Growth.......... 211 f CHAPTER vii PAGE Assets, Liabilities, and Residential Construction Starts . 215 The Structure of the Sacranento Mortgage Market .................. 222 Institutional Lenders' Real Estate Credit Activities .............. 225 Government Underwriting Activity . 238 Effect of Federal Reserve Board Monetary Policy Actions on Local Lenders....................... 240 Lender Survey: Purpose and Methodology................. 240 Anticipated Inpact of Monetary Policy Action: A General Summary 248 Summary of Responses......... 250 Questions Asked Only of Savings and Loan and Bank Respondents .... 267 Savings and Loan Respondents' Replies ................ 268 Commercial Bank Respondents' Replies..................... 288 Summary....................... 305 viii CHAPTER PAGE PART D. GENERAL SUMMARY AND CONCLUSIONS VII. SUMMARY AND CONCLUSIONS.............. 308 Summary........................... 309 Economic Base Theory............ 310 Employment Level Changes Compared with Changes in Nonfarm Residential Housing Starts . . . 311 Income and Housing Starts ........ 313 Sacramento Economic Base Changes Related to Local Housing Starts: An Empirical Analysis .......... 314 The Influence of Selected Demographic Factors on Nonfarm Residential Construction Starts . 316 The Influence of Credit Availability and Cost of Money on Housing St arts............. 319 Savings and Loan Associations: A Major Source of Residential Real Estate Credit........... 321 Impact of Credit Availability on Residential Construction Financing by Savings and Loans . 323 ix CHAPTER PAGE Inpact of Savings and Loan Associations on Residential Construction in the Sacranento A r e a ............................ 323 Lender Survey Findings ............. 327 Conclusions.......................... 330 Suggested Areas of Needed Research . 332 BIBLIOGRAPHY.................................... 335 APPENDIX........................................ 348 / LIST OF TABLES TABLE PAGE 1* Correlation Coefficients of Per Fanily Incone and Nonfarn Housing Units Authorized, Selected Areas, 1954-1960 . 75 2. California and Sacranento Metropolitan Area Labor Force as a Percentage of Population, 1950, 1953, 1956, 1959, and 1960 79 3. Sacranento Metropolitan Area Basic and Nonbasic Industry Enploynent, 1940, 1950, and 1960 81 4. Sacranento Metropolitan Area Basic and Nonbasic Industry Enploynent Ratios, 1940, 1950, and 1960 83 5. Sacranento Metropolitan Area Basie and Nonbasic Industry Enploynent Ratios, 1940-1950 and 1950-1960 .............. 85 6. Wage and Salary Workers in Governnent and Manufacturing in the Sacranento Metropolitan Area, 1950, 1959, and 1960 86 7. Changes in Total Personal Incone, United States, California, and the Sacranento Metropolitan Area, 1950 and 1955-1959 . 88 xi TABLE PAGE 8. Wages and Salaries as a Per Cent of Total Personal Income, California and Sacramento Metropolitan Area, 1950 and 1955-1959 ............................ 91 9. Average Wage Per Worker, United States, California, and the Sacramento Metropolitan Area, 1953, 1956, and 1959 93 10. Estimated Average Annual Wages Paid in Selected Industries, United States, California, and the Sacramento Metropolitan Area, 1953 and 1959 . . . 95 11. Estimated Annual Average Wages Paid in Selected Manufacturing Industry Classifications, California and the Sacramento Metropolitan Area, 1959 • • 96 12. Average Annual Wages Paid to Government Employees in the Sacramento Metropolitan Area, 1959 .............. 98 13. Per Capita Income, United States, California, and the Sacramento Metropolitan Area, 1950 and 1955-1959 . 101 14. Multiplier Effect of Additional Aerojet Employment, 1954-1961 ................ 106 xii TABLE PAGE 15. Average Enploynent and Average Annual Wage Paid by Aerojet-General Corporation, Sacranento, 1957-1962 . . 109 16. Percentage Estinatee of Civilian Population and Nonfarn Housing Starts by Regions of the United States, 1950-1960 ............................ 113 17. Population and Housing Units Growth, the United States and Selected States, 1950 to 1960 117 18. Population Estinates of Selected States: 1960 and 1950 ........................ 120 19. Housing Units, United States and Selected States: 1950 and 1960 ................ 121 20. Net Migration, United States and Selected States: 1950 to 1960 125 21. Share of Population Growth Attributable to Net Migration and Nonagricultural Enploynent Increases, United States and Selected States, 1950-1960 127 22. Percentage Increases in Bond Prices and Housing Starts, Selected Dates .... 141 23. Correlation of Housing Starts and Bond Price Data, Selected Dates .......... 160 xiii TABLE PAGE 24. Correlation Coefficients of Bond Prices and Housing Starts: United States, Regions of the United States, Selected States, Metropolitan Areas, and Dates • 162 25. The Structure and Growth of the United States Savings and Loan Industry: Nuaber of Associations by Type of Charter, Selected Dates .............. 166 26. The Structure and Growth of the United States Savings and Loan Industry: Assets of Associations by Type of Charter, Selected Dates .............. 168 27. Average Assets of United States Savings Associations by Type of Charter, Deceaber 31, 1960 .................... 170 28. Total Assets of All Savings Associations 171 29. Total Liabilities of All Savings Associations ........................ 174 30. The Structure and Growth of the California Savings and Lean Industry: Nuaber of Associations by Type of Charter, 1950-1960 177 I" 1 xiv TABLE PAGE 31. Tt-i Structure and Growth of the California Savings and Loan Industry: Assets of Associations by Type of Charter, 1950-1960 179 32• Average Assets of California Savings Associations by Type of Charter, Deceaber 31, 1960 .................... 181 33. Change in Share of Total Assets of the California Savings and Loan Industry by Type of Association Charter, 1950 and 1960 182 34. Major Assets and Liabilities of Insured Savings Associations Operating in California, 1950-1960................. 185 35. Nonfarn Mortgage Recordings of $20,000 or Less, in the United States, by Type of Lender, 1950-1960 .......... 189 36. Nonfarn Mortgage Recordings of $20,000 or Less, in the United States, by Type of Lender, 1950-1960 . . . ; ............. 192 37. Nonfarn Mortgage Recordings of $20,000 or Less, in the United States, by Type of Lender, 1950-1960 ..... .......... 194 XV TABLE PAGE 38. Nonfarn Mortgage Recordings of $20,000 or Less, in California, by Type of Lender, 1950-1960 .............................. 196 39. Proportion of Hone Mortgages Recorded, in California, by type of Lender, 1950-1960 .............................. 197 40. Mortgage Loans Made in the United States by All Savings Associations, by Purpose, 1950-1960 200 41. Mortgage Loans Made by Insured California Associations, by Purpose, 1950-1960 . . 203 42. Savings and Loan Association Mortgage Credit Availability and Nuaber of Construction Loans Made by the Associations in the United States, 1950-1960 .............................. 205 43. The Composition and Growth of the Sacranento Savings and Loan Industry: Nuaber of Associations by Type of Charter, 1950-1960 212 44. The Composition and Growth of the Sacramento Savings and Loan Industry: Assets of Associations tyy Type of Charter, 1950-1960 214 table page 45. Trends in Intra-Industry Competition as Shown by Share of Total Assets and Average Assets of Savings and Loan Associations, United States, California, and Sacramento, 1950 and 1960 216 46. Major Assets and Liabilities of Insured Sacramento Savings Associations, 1950- 1960 218 47. Sacranento Savings and Loan Associations' Role in Financing Local Residential Construction: 1951-1960 .............. 221 48. Trust Deeds and Mortgages Recorded in Sacramento County, 1955-1960 227 49. Selected Recordings of Institutional Loans, Excluding Second Trust Deeds and Individual Loans by Institutions, Sacramento County, 1955-1960 228 50. Selected Sacramento County Real Estate Loans, by Lender, Number, and Volume, 1955-1960 ............................ 230 51. Number and Type of Institutions Originating Real Estate Loans in Sacramento County, 1954-1960 232 xvii TABLE PAGE 52. Commercial Banka, Estimated Real Eatata Loan Activity, by Type, Nuaber, and Voluae, 1955-1960 .................... 233 53. Savings and Loan Aaaociationa Eatiaated Real Eatate Loan Activity, by Type, Nuaber, and Voluae, 1955-1960 ........ 235 54. Life Inaurance and Mortgage Coapaniea Eatiaated Real Eatate Loans in Sacraaento by Type, Nuaber, and Voluae, 1955-1960 .................... 237 55. Trend of FHA Title II Section 203 Applicationa Received and New Hoaea Started in the Territorial Juriadiction of the Sacraaento Insuring Office, 1950-1960 ........................... 239 56. Eatiaated FHA Insuring Operations, Sacraaento County, 1951-1960 241 57. Schedule of NGI" Loans Made in Sacraaento County by Purpose of Loan, 1951-1960 . 242 58. Schedule of NGIM Loans Mado in Sacraaento County by Type of Lender, 1951-1960 . • 244 xviii TABLE PAGE 59. Distribution of Responses, by type of Lender, to the Question, "Has There Been a Tendency on the Part of Your Conpany to Restrict Loans Generally During the Last Six Months to a Year?" 251 60. Reasons Given for Restrictive or Nonrestrictive Policy, During tho Past Tvrelve Months, by type of Lender and Nuaber of Tines Mentioned........ 252 61. Nunber of Loans and Defaults in Sacraaento County, by Selected Lenders, and Selected Dates .......... 255 62. Distribution of Responses, by type of Lender, to the Question, "In What Years Have You Been Most Strict?".......... 261 63. Distribution of Responses, by type of Lender, Explaining the Strict Real Estate Lending Years or the Lack of Strict Years ....................... 262 64. Most Usual Intorest Rates Charged by Sacraaento Savings and Loan Associations in Selected Years, by Nuaber of Tiaes Mentioned............ 281 xix TABLE PAGE 65. Fees Charged by Sacraaento Savings and Loans for Loans Made in Selected Years, by Nuaber of Tiaes Mentioned........ 283 66. Sacraaento Savings and Loans' Most Usual Tern for Repayaent, and Loan-to-Sales Price Ratio in Selected Years, by Nuaber of Tiaes Mentioned............ 284 67. Distribution of Percentage of Sacraaento Savings and Loans' Conventional, FHA, and VA Lending Activity in Selected Years................................ 286 68. Most Usual Interest Rate Charged by Sacraaento Banks in Selected Years, by Nuaber of Tiaes Mentioned.... 298 69. Fees Charged by Sacraaento Banks for Loans Made in Selected Years, by Nuaber of Tiaes Mentioned............ 299 70. Distribution of Banks1 Loan-to-Sales Price Ratios in Selected Years, by Nuaber of Tiaes Mentioned ............ 301 71. Distribution, by Percentage Coabinations, of Sacraaento Banks Conventional, FHA, and VA Loans During Selected Years, by Nuaber of Tiaes Mentioned ............ 303 LIST OF FIGURES FIGURE PAGE 1. Percentage Changes in Regional Nonfarn Enploynent and Housing Starts, 1954-1960 ........................... 49 2. Percentage Changes in Nonfarn Enploynent and Housing Starts: Massachusetts, New Jersey, New York, and Pennsylvania, 1954-1960 ........................... 52 3. Percentage Changes in Nonfarn Enploynent and Housing Starts: Illinois, Michigan, and Ohio, 1954-1960 .................. 55 4. Percentage Changes in Nonfarn Enploynent and Housing Starts: Georgia, Maryland, and Virginia, 1954-1960 ........ 58 5. Percentage Changes in Nonfarn Enploynent and Housing Starts: California and Washington, 1954-1960 ................ 60 6. Percentage Changes in Nonfarn Enploynent and Housing Starts: Boston, Buffalo, New York-Northeast New Jersey, Chicago, Cleveland, and Detroit, 1954-1960 ... 64 7. Percentage Changes in Nonfarn Enploynent and Housing Starts: Atlanta and Baltinore, 1954-1960 67 xxi FIGURE PAGE 8. Percentage Changes in Nonfarm Employment and Housing Starts: Los Angeles, Sacramento, San Diego, and San Francisco-Oakland, 1954-1960 68 9. Total Annual Additional Employment and Housing Units Authorized: Sacramento Metropolitan Area, 1954-1961 107 10. Bond Prices and Bank Rates on Short-Term Business Loans ...................... 134 11. United States Government Bond Prices and Housing Starts, 1948-1961 ............ 137 12. Bond Price-Housing Starts Scatter Diagram . . ........................ 143 13. Member Banks' Net Reserves and Housing Starts, 1950-1960 .................... 146 14. Bond Prices and Regional Housing Starts, 1954-1961 ........................... 150 15. Bond Prices and Housing Starts in Boston, Buffalo, Chicago, Cleveland, Detroit, and New York, 1954-1961 .............. 154 16. Bond Prices and Housing Starts in Atlanta, Baltimore, Los Angeles, Sacramento, San Diego, San Francisco-Oakland, and Seattle, 1950-1961 156 PART A INTRODUCTION CHAPTER I PURPOSE, SCOPE, AND SIGNIFICANCE OF THE STUDY I. INTRODUCTION The total amount of outstanding residential mortgage debt in the United States more than doubled during the years 1950 to 1961. Nonfarm residential mortgage debt increased from $45.2 billion in 1950 to $153.5 billion outstanding at year-end 1961.* In percentage terms this represents a 240 per cent increase. The principal sources of real estate credit are financial institutions such as savings and loan associations, commercial banks, mutual savings banks, and insurance companies. These insti tutions, as a group, increased their share of the total nonfarn residential mortgage debt from 78 per cent in 1950 to 84 per cent at year-end 1961. How ever, the relative importance of these institutions as suppliers of real estate credit has changed significantly since 1950. For example, the share of the total nonfarm residential mortgage debt held by 1Savings and Loan Fact Book. 1962 (Chicago: United States Savings and Loan League, 1962), p. 71. The Federal Home Loan Bank Board ostimatea that total nonfarn mortgage debt outstanding increased to $168.3 billion in 1962. 3 savings and loan associations increased froa 29 per the share held by coaaercial banks decreased froa 21 per cent in 1950 to 13 per cent in 1961; also, life insurance coapanies' share of total residential aortgage debt decreased froa 19 per cent in 1950 to 17 per cent in 1961; and, the share of total residen tial aortgage debt held by autual savings banks increased froa 9 per cent in 1950 to 13 per cent in 1961.3 California savings and loan associations pro vided funds for 49 per cent of the state's total nonfara aortgage recordings or 22 per cent of the national aggregate in 1961.4 It should be noted that auch of the draaatic growth of California and the California savings and loan industry has taken place in southern California. An investigation of a fast growing urban area which elicits answers to questions of general iaportance relating to the aajor econoaic United wj and Loan League indicate that savings and loan associations aade 46 per cent of the total nonfara aortgage recordings undor 620.000 in 1963. cent in 1950 to 41 per cent in 1961 2 In contrast Preliainary estinates by the gIbid 4Ibid.. p. 46 4 factors which influence new residential construction would be helpful in the analysis of the deaand for and 5 the supply of real estate credit. The selection of the Sacraaento Metropolitan Area in northern California as an integral part of this investigation is justified on the basis that this urban area is aore aanageable for purposes of eapirical research and analysis than other largerf faster-growing areas. Furthermore, this area appears to present in aicrocosa aany of the For an excellent discussion of aortgage credit and new residential construction starts in terns of national aggregates see United States Congress, Senate, Coaaittee on Banking and Currency, Study of Mortgage Credit. Hearings before Subcoaaittee on Housing, SStETongress, 2d Session, Oeceaber 22, 1958 Trevised July 11, 1960; Washington: Governaent Printing Office, I960). See Edward E. Edwards, "Real Estate Econoaics. A Return to Fundaaentals." The Appraisal Journal. April, 1949, which describes the real estate narket as being local in character, highly disorganized, and in reality containing a large nuaber of separate aarkets wherein the effective deaand for real estate is dependent, for the aost part, on the availability and terns of financing. Weiner has suggested that the inpact of restrictive aonetary and credit policies has varied noticeably in the various sectors of the econoay: Arthur M. Woiaer, "Real Estate and the Business Outlook," Proceedings, The Sixteenth Stanford Business Conference. July 22-25."1 9 5 7 (sianfordx Stanford University, 1957), p. 3 3 . Also, see Junia Honnold, Richard U. Ratcliff, and Daniel B. Rathbun, Residential Finance. 1950 (New York: John Wiley and Sons, Incorporated, 1957), pp. 21-22: Honor Hoyt and Arthur Weiner, Principles of Real Estate (fourth edition; New York: The Ronald Press Coapany. 1960), p. 153; Maurice A. Unger, Real Estate Principles and Practices (Cincinnati: Souxh-Wesiern Publishing Coapany,1959), p. 375. 5 characteristics of urban growth of other areas as well as aany of the features which have typified the savings and loan industry in California, nanely, expansion of the economic base which resulted in rapid population growth and consequent large number of housing starts; aggressive savings and loan asso ciation activity in the residential financing sector; a rapid savings growth; formation of several new state-chartered associations; and, the entry of two holding companies into the local savings and loan area of operations. Thus, answers to questions of general iaportance should result froa an investigation of activity in this urban area. II. THE STATEMENT OF THE PROBLEM This study proposes to exaaine and analyze aajor econoaic determinants which have influenced nonfara residential construction in the various regions, selected states, and selected metropolitan areas in the United States. An integral part of this investigation is based upon the statistical analysis of published data related to selected local areas which have becoae available only within the last fifteen yecurs, and the Sacraaento Metropolitan Area data developed for this study. The hypothesis 6 underlying this study is that increased rates of nonfarm residential construction are influenced by (1) favorable changes in the economic base of an area, (2) availability of funds for real estate credit at savings and loan associations, and (3) the Federal Reserve System's monetary policies as they affect the availability of funds of nonbanlc financial inter mediaries. 111. SIGNIFICANCE AND SCOPE OF THE STUDY Importance of the Study The solution of real estate credit problems is of basic importance to the maintenance of sound and growing residential real estate markets. Knowl edge of real estate finance has improved greatly in the last twenty years; however, very little has been written about the extent and sources of equity funds, individual mortgage money markets, or the impact of mortgage lending practices and policies on real estate markets. If a mortgage money market is to function effectively its elements, in actual practice, must be carefully examined and accurately evaluated. Significant differences exist between geographic sections of the nation in relation to mortgage lending activity and money markets. Some are areas of surplus 7 funds with a net outflow of aortgage money. Others are deficit regions and nust look elsewhere for all or part of their mortgage funds. The differences are such as to indicate a need for a mortgage market study in the sense that national averages do not reveal the g underlying currents in any one market. It should therefore be fruitful to examine in detail the struc ture and functioning of the mortgage market in an important and rapidly growing section of the United States. Furthermore, the Sacramento Metropolitan Area is a large and active real estate market and events in this urban market could provide insight into national as well as other urban market trends. At the local level, this examination of economic base changes and the economic effects of mortgage lending by savings and loan associations may aid both in evaluating the past record and in determining the potential economic contribution to be made by these financial institutions in this and other metropolitan areas. g See Charles M. Torrance, "Gross Flows of Funds through Savings and Loan Associations," The Jonmai of Finance, XV (May, 1960), 163, where li is stated that "geographic variations in rates of return on hone mortgages cure quite pronounced and contribute to cloud ing the significance of national averages." 8 Limitations of the Study The empirical data developed in this study (1) are geographically restricted to the Sacramento Metropolitan Area; (2) focus upon the impact of savings and loan associations' lending activities on residential construction during the 1950-1961 period; (3) recognize the difficulty of isolating the net effect of changes in monetary policy, Federal Home Loan Bank regulations, and the economic base, from the many other factors acting simultaneously on the activity of savings and loan associations; (4) are dependent upon the accuracy and interpretation of certain published data; and (5) rely upon the relia bility and interpretation of data developed froa interviews, questionnaires, and field investigations. IV. REVIEW OF THE LITERATURE This section is devoted to (l) a review of research which directly relates to this study, and (2) a review of other related research and literature. Research directly related to this study may be broadly classified as aortgage credit studies which 7 The Bureau of Census defines a "Standard Metropolitan Area," except for the New England area, as a county, or a group of contiguous counties, con taining a city of 50,000 population or more. 9 include local Mortgage Market studies, junior Mortgage financing studies, and the cost of mortgage Money research projects. Related research and literature include testiMony given at Congressional Hearings, literature concerned with the effect of Monetary policy in general, and the specific effect of Monetary policy on savings and loan associations in particular. One local aortgage Market study which was com- pleted in 1956 concerned itself with the institutional structure of the Los Angeles County aortgage Market a during the five year period 1946-1951. It was found that during the study years Los Angeles lending insti tutions were able to expand alnost without Halt. That soae of these lending institutions aushrooned, while others reaained approxiaately static, was attributed priaarily to aanageaent policy. The changing percentage of the Mortgage Market held by each type of institution was exaained and related to conditions in the Market and to institutional restraints and regulations. During this period savings and loan associations absorbed an increasingly greater portion g Jaaes Gillies and Clayton Curtis, Institutional Bureau of Business and Econoaic Research, 1956). 10 of the mortgage market and at the end of the period were the most important lenders in Los Angeles County. It was concluded that primarily as a result of local conditions these institutions grew rapidly and that this growth was attributed to aggressive management policy and the large increase in shareholder savings. Also, the operations of the Los Angeles County mortgage market were compared with the operations in other local mortgage markets for which information was available. From this comparative analysis the relationship of local mortgage market structure to government insurance and guarantee programs was developed. The hypothesis that the structure of the local mortgage market is a major determinant of the effectiveness of government programs in influencing the volume of local residential mortgage lending was tested and substantiated. Another study indicated that in 1956, at least, the small mortgage market structure did not provide an accurate basis upon which to predict the impact of government insured and guaranteed programs on the Ann Arbor market, a nonmetropolitan area.9 This study q James C. T. Mao, "The Impact of Federal Mortgage Insurance Programs on Ann Arbor's Home Mortgage Market, 1956,'' The Journal of Finance. XIII (September, 1958), 412-416. tested the hypothesis that, since lending institutions operate with significantly different sets of attitudes toward FHA and VA lending, knowledge of the relative importance of the major institutions in a market can be used to predict the amount of FHA and VA lending that will be done in the local mortgage market. That is, in those local mortgage money markets where savings and loan associations are the predominant lenders, the proportion of insured lending will be lower than in areas where commercial banks and life insurance companies, if any, predominate. The reported con clusion was that the Ann Arbor major type lenders deviated significantly from their national patterns and suggested that other factors such as lending policies of local institutions, the home buyers' income and wealth positions, the level of local con struction costs, and the VA appraisal policy may have been operative in the 1956 Ann Arbor market.10 The authors of the 1956 Los Angeles mortgage market study replied in support of the hypothesis advanced by them, that their earlier article was based on data restricted to twenty-five major metropolitan areas, consequently it was impossible to conclude 12 whether or not the theses applied with equally satis factory results to nonmetropolitan areas.11 Further more, since their earlier article, additional data on two other small markets had been reported* Therefore, this information, together with the studies of others, may make possible the construction of a model of mortgage markets which will enable the prediction of events in this important area of capital formation. They concluded: If it may be assumed that small markets will generally have the smaller financial institutions, then it follows that the participation in the government programs will be significantly differ ent in the average small market than in the large* In general, we can expect small markets with the same institutional structure as the larger metropolitan markets to lend a significantly smaller portion of their funds on FHA and VA mortgages* This conclusion agrees with the findings for the three specific small markets for which we have complete information.12 One unpublished doctoral dissertation related the findings of a Middletown, Connecticut mortgage market study with an analysis of the findings of five James Gillies and Clayton Curtis, "A Note on the Small Mortgage Market," The Journal of Finance. XLV v »*-tember, 1959),“?X1-414. 12Ibid* * p. 414. 13 13 other local Mortgage Market studies. These studies analyzed the effects of different organizations and 14 procedures in different types of Mortgage Markets. Soms of the conclusions drawn froa a consolidation of the findings of the other studies were reported as follows: The great proportion of aortgage debt in the United States is still originated and held by Mortgagees located in the sane local Market area as the properties being Mortgaged. 13Williaa N. Kinnard, "The Mortgage Market in Middletown, Connecticut" (unpublished Ph.D. disserta tion, University of Pennsylvania, Philadelphia, 1956). The five studies analysed were: Paul M. Gregory, "The Vorchester Mortgage Market" (unpublished Ph.D. dis sertation, Clark University, WorChester, Massachusetts, 1942); Mathew M. Johnson, "The Philadelphia Mortgage Market: with Special Enphasis on Residential Mortgages" (unpublished Ph.D. dissertation, University of Penn sylvania, Philadelphia, 1951); John H. Cover, et al.. Residential Mortgage Financing: Hagerstown. Maryland (College Park, Maryland: University of Maryland, 1951); Paul F. Wendt and Daniel B. Rathbun, The San Francisco Bay Area Residential Mortgage Market, housing feosoarch Paper No. 2 0 (Washington: Governaent Printing Office, May, 1952); George B. Hurff, et aj,., Residential Mortgage Financing: Jacksonville. Florida: First Six Months of 1950. Housing Research Paper No. 23 (Washington: Governaent Printing Office, 1952). **The Hagerstown, Maryland Mortgage Market represented a snail city narket with a restricted total supply of funds; the supply of Mortgage funds in the San Francisco Bay Area was prinarily of local origin, but with a strategic Minor portion of aortgage noney at coapetitive rates entering the Market frea eastern sources; Jacksonville, Florida represented a large city Metropolitan aarket dependent nostly upon an inflow of outside aortgage funds to finance nass- production housing. 14 The demand for mortgage loan funds is a derived demand, stemming from the demand for real estate. Residential aortgage loan demand is relatively price-elastic as loans are made more favorable to the mortgagor. The local aortgage market is characterised by oligopoly on the lending or supply side. Loan terms vary significantly among lender groups within the market, and even within lender groups. Marked differences in interest rates, loan maturities, equity requirements, and service charges have been consistently observed. Lender groups within a market also vary considerably in the extent of their participation in the FHA and VA mortgage loan programs. A direct competitive relationship between mortgage loans and private securities, particularly those of public utilities, has been observed in the portfolios of institutional mortgagees. For institutional lenders, the supply of mortgage loan funds has tended to vary inversely with the attractiveness of alternating long-term investment outlets. Local restrictions and individual institutional policies have exerted an important influence on the supply of mortgage loan funds in the mortgage market. Institutional policy has been reflected in mortgage loan terms as well as in investment portfolios. It is held that mortgage loan rates have been more than adequate to compensate for the risks involved for the lender, although non-price competition has in some instances prevailed as a result of the oligopolistic character of the local mortgage market. Borrowers tend to be immobile in their borrowing behavior. When outside lenders are attracted to a mortgage market area on a large scale, the competition with local lenders gives rise to loan terms relatively favorable to the borrower. It is argued that such 15 outside competition also serves as a check against the abuse of oligopolistic power by local institu tional lenders.15 The remainder of this section is devoted to a review of other related research and literature. That is, studies which discuss the effects of monetary policy on mortgage credit in general and those primarily concerned with savings and loan associations in particular. / Testimony given at the Subcommittee on Housing of the Senate Committee on Banking and Currency Eighty-Sixth Congressional Hearings on mortgage 16 credit was reported in two publications. Outstand ing individuals from government, private industry, and educational institutions were invited to read papers on each of the component topics developed from the general question, "Does the decade 1961-1970 pose problems in the private housing and mortgage markets 15 Kinnard, op. cit., pp. 3-12. 16 United States Congress, Senate, Committee on Banking and Currency, Study of Mortgage Credit. Hearings before Subcommittee on Housing, Min Congress, 2d Session (revised July 11, I960; Washington: Government Printing Office, December 22, 1958); also, United States Congress, Senate, Committee on Banking and Currency, Study of Mortgage Credit. Hearings before Subcommittee on Housing, 86th Congress, 1st Session (Washington: Government Printing Office, May 14, 15, 19, 20, 21, 25, 26, 28, 29, 1959). 16 which require Federal legislation by 1960?" These component topics included questions related to credit terms and effects of monetary policies, among other things. One of the conclusions reported was: In an economic environment where general credit controls operate in conjunction with Federal housing programs characterized by relatively inflexible interest rates, the volume of credit flowing into mortgages is extremely sensitive to changes in monetary policy. The flow of home mortgage funds fluctuates more violently than capital flow to other sectors of the economy, and the movement upward and downward usually precedes the movement in other sectors. Thus, housing, as an industry, is relatively less stable than the economy in general.17 Dr. Warren Smith presented a paper at these hearings in which he concluded: . . . during the period since 1953, general credit controls have had a strong impact on residential construction through their ability to Influence the supply of funds available for Government- supported mortgages .... Presumably this will continue to be the case as long as the essential features of our present institutional structure and monetary controls are retained.^ Another published paper indicated that because it is clear that monetary policies have highly 17 Committee on Banking and Currency, Study of Mortgage Credit, 86th Congress, 2d Session, p. 434. 18Warren L. Smith, "The Impact of Monetary Policy on Residential Construction, 1948-1958," Committee on Banking and Currency, Study of Mortgage Credit. 86th Congress, 2d Session, p. 2(50. 18 mortgage credit conditions have significant effects on the volume of nonfarm residential construction from one short period to another. One of the con clusions was: . . . Ultimately, changes in general credit conditions are traceable to the policies of the Federal Reserve System. Thus, general credit controls affect general credit conditions, which, in turn, affect the volume of nonfarm residential construction.21 The second unpublished dissertation reviewed concluded: . . . the fact that credit is so important in real estate transactions suggests that a given over-all change in cost and availability of credit would tend to have a relatively greater impact on the real estate sector of the economy than on other sectors in which credit is less 2o important in the consummation of transactions. Literature primarily concerned with savings and loan associations included the published proceed ings of the annual conferences which discussed the basic economic theories underlying savings and Coldwell Daniel III* "Effects of General Credit Controls on Nonfarn Residential Construction," The Journal of Finance. XV (March, I960), 78. 22 David P. Eastburn, "Real Estate Credit Controls as a Selective Instrument of Federal Reserve Policy" (unpublished Ph.D. dissertation, University of Pennsylvania, Philadelphia, 1957), p. 10. 17 important effects in the mortgage market, monetary theorists should not limit their discussions to the impact of monetary policies upon commercial bank 19 credit and the bond market. After a review of the recent history of the impact of monetary policies on the mortgage market the conclusions reached were, in part, as follows: There can be little doubt that during the past several years monetary policies have had a highly significant impact upon the mortgage market as a whole and a differential impact upon component parts of the market. At best, general credit policies will have a difficult time coping with mortgage credit and housing expenditures because of the forward- commitment process and the long lag between commitment and actual construction. Finally, this suggests that careful and detailed research on the impact of monetary 2n policies on the mortgage market is sorely needed. Similar conclusions were reported in two unpublished doctoral dissertations which explored the effects of credit control policies of the Federal Reserve System. An abstract of one of these disserta tions indicated that changes in nonfarm residential James J. O'Leary, "The Effects of Monetary Policies on the Mortgage Market 1 The Journal of Finance. XII (May, 1958), 176. 20Ibid., pp. 185-187. 19 23 residential financing. These conferences were held each year in May, 1958, 1959, 1960, 1961, and 1962, when university educators actively engaged in teaching and research in the fields of aonetary theory, finance, and real estate were brought together to discuss the relationship of the basic economic theories and the aarket mechanisms of the savings and loan industry. The one real contribution of these conferences is that they have served to generate soae excellent discussions aaong the leading contributors to the literature on the effectiveness of aonetary policy. For exaaple, while noting the problem of distinguishing between demand deposits of banks on the one hand and all other types of liabilities of financial institutions on the other, Dr. Goldsmith stated: In the last three years we have been presented with two very interesting theoretical challenges in this field. One is the Gurley-Shaw theory which ultimately denies the specific difference between money and other liabilities of financial institutions. The other one is Milton Friedman's theory that there is no correlation at all between the level of income and the saving ratio.24 23 United States Savings and Loan League, Pro ceedings of the Conference on Savings and Residential Financing. lii»-1962 (Chicago: The League. 1&58-1962). 24 Raymond W. Goldsmith, "The Supply of Saving," United States Savings and Loan League. 1958 Proceedings. p. 121. 20 At the same conference Dr. Roland I. Robinson pointed out, "It has been argued that the influence of monetary policy on credit availability and on the supply of money is frequently reduced by a variety of 25 economic or institutional adjustments." He spoke of these as being monetary policy offsets and noted that one of the principal offsets indicated in recent economic debates was money substitutes supplied by nonmonetary financial intermediaries. That is, "A weakening of the influence of monetary policy through a supply of liquid assets by nonbanking financial 26 institutions." He concluded, however, that "theo retically, the possibility of an escape from monetary policy by use of the liability of nonbanking financial intermediaries is real. It could happen, but I find no evidence that it has yet become of material sig nificance."27 With reference to the Gurley-Shaw thesis, Dr. Jules Bogen concluded: 25 Roland I. Robinson, "The Impact of Monetary and Fiscal Policy on Residential Financing," United States Savings and Loan League, 1958 Proceedings, pp. 130-131. 26Ibid., p. 132. 27Ibid., p. 134 21 Gurley and Shaw want to do four things: First, they want to eliminate the distinction between money and financial assets. That is most fundamental. Second, they want to eliminate the distinction between money creation and the creation of other financial assets, such as insurance policies, savings and loan shares, deposits in mutual savings banks, pension fund claims and so on. Third, they argue that financial assets other than money have increased more rapidly through the years than money has. Fourth, they state that the monetary control measures sponsored by the Board of Governors of the Federal Reserve System have become decreasingly effective because an increasing segment of our finance system, the so-called finance interme diaries, is not subject to these controls.28 In his attempt "to show that in no important respect is it legitimate to single out commercial banks 29 and money for special analytical treatment,** Gurley summarized the main points of his theory as follows: 1. The mere act of creating financial assets does not distinguish commercial banks from other creators of financial assets, for each financial asset is quite clearly created by someone. 28 Jules I. Bogen. "Trends in the Institutional ization of Savings," United States Savings and Loan League, 1958 Proceedings, p. 156. 29 John G. Gurley, "Financial Institutions in the Saving-Investment Process (Part I)," United States Savings and Loan League, 1959 Proceedings. p. 25. 22 2. There is a great deal of differentiation of financial assets in our economy; each type differs in some respect froa all others. Money is a very important financial asset, but so are many others, judging by the demands for them. It is not obvious that one kind of differentiation should be con sidered holy, to the neglect of others. 3. In the general sense, multiple creation is found throughout the economy— in commercial banks, in other financial intermediaries, in business firms and so on. This does not set commercial banks apart from other creators of financial assets. In the legal sense, there is a multiple creation of money because commercial banks are subject to quantitative controls. In the same sense there would be multiple creation in non monetary intermediaries if they were subject to the same controls. 4. Commercial banks appear to be inherently more volatile than other financial intermediaries, but this appearance is largely due to the fact that banks are subject to quantitative controls and other intermediaries are not. The tables would be turned to a large extent if controls were placed on nonmonetary intermediaries and com mercial banks were not subject to them. 5. Both commercial banks and nonmonetary inter mediaries can add to inflationary pressure. Both can create an excess supply of money; both can create loanable funds, in the orthodox definition of the term. Dr. Ezra Solomon pointed out that parts of the issue of financial intermediaries and whether they should be brought under the direct network of quanti tative control are known as the Gurley-Shaw thesis which he summarized as follows: 30Ibid., pp. 24-25. 23 1. The size and variety of financial inter mediaries has grown rapidly since 1900. As a consequence the oldest intermediaries of all, commercial banks, have declined in relative importance. 2. In spite of this decline in the relative importance of banks, the mainstream both of theory and of control is directed exclusively at the banking system. 3. This procedure is ineffective because banks are simply one variant form of intermediary, and their liabilities— demand deposits— are simply one variant form of financial assets. To single out demand deposits for special analytical treat ment or to single out banks for special control makes very little sense. 4. Effective control would cover the assets of all financial intermediaries, not of commercial banks alone. This particular line of argument has given rise to more academic controversy than have the other three— and also to more theoretical confusion.31 Solomon contributed much of the confusion and diversi fied controversy over monetary policy to previously unresolved semantic tangles and "new semantic con fusions and definitional difficulties."32 After pointing out that portions of the Gurley-Shaw thesis are to a great extent dependent upon the author*' 31 Ezra Solomon, "Financial Institutions in the Saving-Investment Process (Part II)," United States Savings and Loan League, 1959 Proceedings. p. 31. 32Ibid., p. 32. 24 unusual definitions, Solomon stated that his con clusions are exactly the opposite of Gurley and Shaw. Solomon suggested that (1) intermediaries can tem porarily delay the impact of Federal reserve policy, and (2) intermediaries have a second effect which assists the Federal Reserve System in its process of monetary control by making interest rates higher and 33 bond prices lower. Solomon summarized his position as follows: . • • there is no need whatever for quantitative controls to be extended beyond the banking system. If anything, the time-deposit sector of the bank ing system should be segregated froa demand deposits and then removed completely froa quan titative control; that is, time depositories .. should be treated as a financial intermediary. While speaking of the problems of aonetary reform Or. Milton Friedman cited the Gurley-Shaw argument and concluded: I think this whole argument reflects a funda mental misconception of wherein the power of the Federal Reserve System lies. The detail control of banks has almost nothing to do with the essential power of the System. The essential power of the System is the power to determine the total stock of what we call high-powered money— pieces of paper we carry around in our pockets and the reserve balances which the banks hold at the Federal Reserve Banks.3* 33Ibid.. p. 40. 34Ibid., p. 41. 35Milton Friedman, "A Program for Monetary Stability (Part 1)," United States Savings and Loan League, 1962 Proceedings. p. 25. 25 In consequence, he thought that "there is no justi fication for extending control to financial inter mediaries."*^® Later, during the same conference, Dr. Abba Lerner referred to monetization and stated: This is a case where other devices have been developed for protecting the depositor, and we do not have to worry about the monetization side. Monetization comes on the other side, and there I find myself doing something which I reluctantly find myself doing elsewhere: attacking Friedman and defending him. Why not have 100% reserves to look after the liquidity side so that banks cannot spend money in ways that are not socially desir able? What I mean is that with 100% reserves, whoever is in charge of providing the reserves has the responsibility and may be able to do what needs to be done, whereas, without it, hundreds of banks are creating credit or not creating it for reasons which may not be very good reasons— they may be good reasons from the point of view of the bank, but not in the interest of the economy as a whole.37 Recently, the California savings and loan industry has been the subject of three special studies by Professors Shaw, Balderston, and Grebler 36Ibid. 37 Abba P. Lerner, quoted during the panel discussion of "Criteria of a Well-Functioning Financial System, Topic 1. The Segregation and Investment of Time and Savings Deposits of Commercial Banks," United States Savings and Loan League, 1962 Proceedings. p. 105. 26 38 and Brigham. The stated purpose of the Shaw report was to present an accurate picture of the economic forces at work within the industry and to suggest a structure of the industry that will accommodate still more riches with maximum benefit for savers, mortgage 39 borrowers, and guarantee stockholders. Shaw recommended the use of "selective merger as one technique (of structural development) to increase 40 operating efficiency." This technique "is one step toward retrieving the benefits of competition, for homogeneous products, between millions of buyers and sellers. Along with other measures, it can reduce 41 friction in savings and loan markets." Dr. Grebler indicated that the Shaw report "raised a number of important issues concerning the 3g Edward S. Shaw, Savinga and Loan Market Structure and Market Performance* A Study of (California State-LicensedSavings and LoanAssociations (Sacra mento: State of California Savings and Loan Commission, 1962); Frederick E. Balderston, "Revision of the Standards of Applications for New Charters and New Branches in the Savings and Loan Industry of California^' February 8, 1963), (Mimeographed); Leo Grebler and Eugene F. Brigham, Savinas and Mortgage Markets in California: The Position and Porf ormanco~br ihe Savings and Lean industry(Pasadena, California: California Savings and Loan League, 1963). 39 Shaw, op. cit., p. ii. 40Ibid.. p. 127. 41Ibid., p. 148. 27 scope and validity of its analysis and, consequently, the soundness of its reconmendations for public 42 policy." He continued with the suggestion that "the most constructive approach to dealing with these issues appeared to be a new investigation that would place the structure and operations of the California savings and 43 loan industry in a broader context." It is within this framework that Grebler attacked the Shaw report. For example, early in the study Grebler concluded that one of the failures of the Shaw report is "that it examines the California savings and loan industry in a vacuum, without any systematic reference to its role in the financial system or the market structure of 44 financial intermediaries generally." Shaw had acknowledged this early in his report by noting that "the omission we regret most is a close comparison of savings and loan in California with savings and loan elsewhere, with banking, and with other financial 45 enterprise." Shaw considered the California savings and loan industry an entirely passive beneficiary of 42 Grebler and Brigham, op. cit.. p. vii. 43Ibid. 44Ibid.. p. 1. 45 Shaw, on. cit.. p. 4. 28 46 California's growth. Evidence of Grebler's variance with this thesis is reflected in the following state ment : Because financial institutions are active forces in advancing a state's as well as a nation's economic growth, the spectacular expansion of savings and loan associations can also be credited with helping to increase and improve California's housing supply apace with the relentless surge of population.** Vith respect to issues of public policy and regulation of the industry, Grebler seems to be in general agreement with certain of the recommendations advanced by Professor Balderston. For example, Grebler stated: . . • the very wide divergencies in operating cost ratios of various associations which.appear to operate under fairly similar circumstances suggest that the operating efficiency of the industry could be vastly improved if the less efficient managements could be made to emulate efficient managements, and if branching and chartering policies tended to favor the more efficient organizations, as has been recommended in one of the recent reports prepared for the California Savings and Loan Commissioner.*8 Criteria for the licensing of branches of savings and loan associations and for mergers 46Ibid.. p. 118. 47 Grebler and Brigham, op. cit.. p. 100. 48Ibid.. p. 165. 29 should include preference to efficient Managements, provided that objective standards are developed for evaluating management performance.49 Other evidence of Grebler's general agreement with Balderston's recommendations is found in the sug gestion that the removal of the lending radius mileage 50 limitation would remove impediments to competition. Grebler*s summary of findings implies that the savings and loan industry in California operates in reasonably competitive markets and includes so many institutions that no single savings and loan or group of associa tions holds a dominant market position.®* One unpublished thesis which examined the effect of a restrictive monetary policy on savings and loan associations concluded, in part: . . . the major effects were (1) the narrowing of the spread between the rate paid to savers and the rate received on lending, (2) a restrictive 49Ibid.. p. 177. ®°Ibid., p. 180, citing Frederick Balderston, "Revision or the Standards for New Charters and New Branches in the Savings and Loan Industry of California,” February 8, 1963. (Mimeographed.) The latter study is out of print and unavailable for further review. 51 Grebler and Brigham, op. cit., pp. 160-166. 30 rather than an encouraging influence on savings, and (3) a decrease in the ability of associations to adjust to any future rate increases.$2 In a discussion of the competitive elements of the broad money market the question was raised as to whether or not the savings and loan associations' growth has not been achieved at the expense of other 53 financial intermediaries. The indicated conclusion was "that the total pool of savings, particularly institutional savings, has been much expanded by the associations' persistent efforts."54 However, the result of this aggressiveness was "that their 'imme diate' reinvestment of funds in mortgages for the construction and purchase of homes serves to channel the savings capital equivalents back into the spending stream, where the money can perform additional economic duties."55 Therefore: . . • the boom operations of savings and loan associations serve as a dramatic illustration of the constructive role played, to a greater 52 Jerry A. Saunders, "Impact of a Restrictive Monetary Policy on Los Angeles Metropolitan Savings and Loan Associations (unpublished Master's thesis, University of Southern California, Los Angeles, June, 1960), pp. 196-197. cm Torrance, op. cit.. p. 162. 54Ibid. 55Ibid. 31 or lesser degree, by financial intermediaries generally in the functioning of the economy— in this instance, the housing and related » sectors being the principal beneficiaries. In another recent paper which discusses the effectiveness of monetary policy the following hypothesis was advanced: . . • normally movements in credit availability and interest rates will reinforce each other (when rates increase, for example, credit will become "tighter"), but that under certain con ditions availability may move perversely, that is, to offset changes in interest rates .... The paper's author explained that this was the paradox of easy money and tight lenders or tight money and easy lenders where "a shift in lenders' risk functions takes place coincident with a shift in demand." The author concluded that the theory appears to be con sistent with the analyses made of general capital market conditions of the late 1920's and the early 1930's where "the perverse movement in availability was an important factor underlying the impotence of 59 monetary policy." He further stated that "the theory 56Ibid. 87Jack Guttentag. "Credit Availability, Interest Rates, and Monetary Policy," The Southern Economic Journal. XXVI (January, 1960), 2l$. 58Ibid.. p. 227. 59Ibid.. p. 228. 32 also appears to be consistent with the available facts . • • covering the post-World War II period, when, it is reasonable to assume, *normal' conditions 60 have generally prevailed." After indicating that the prevalence of these normal conditions in the postwar period may have been an important factor underlying the postwar revival of monetary policy he concluded that "it is well to bear in mind the possibility that availability may move perversely, 61 however, lest we lose perspective." One more recently published paper examines the role of interest rates, down payments, and contract maturities in the allocation of mortgage funds among 62 borrowers. The thesis advanced in this paper was that unlike the multiple-term hypothesis, which is one variant of the credit-rationing hypothesis, ... an explanation for the co-variation of mortgage interest rates and average contract terns can be found in explicit consideration of the fact that mortgages with different down-payments and maturities are really 60Ibid. 61Ibid. 62 Richard F. Muth, "Interest Rates, Contract Terms, and the Allocation of Mortgage Funds," The Journal of Finance. XVII (March, 1962), 63-80. 33 qualitatively different debt instruments and carry different contract interest rates. It was the author's conclusion that "the hypothesis would seem to be consistent with at least as much market behavior as the so-called credit-rationing or multiple-term hypothesis and more consistent with the economic analysis of markets other than credit mar kets."®* One of the authors cited earlier has reviewed the determinants of short-run fluctuations in residential construction during the 1946-1959 C K period. The thesis advanced was that "changes in the supply of mortgage credit were related in large part to changes in the level of general economic activity, so that fluctuations in residential con struction resulted from fluctuations in general 66 economic activity." One of the pertinent con clusions stated was that "the intricate mechanism through which business fluctuations generate counter- 63Ibid.. p. 64. 64Ibid.. p. 80. ®5Jack Guttentag, "The Short Cycle in Resi dential Construction, 1946-59," The American Economic Review. LI (June, 1961), 275-298. 66Ibid.. p. 291. 34 cyclical swings in the supply of mortgage credit 67 might develop kinks at one point or another." Also, the time series data presented in that paper "suggest the hypothesis that the response of housing to easy credit may be coming progressively later during recessions. This hypothesis warrants careful study."®® Another recent paper has as its purpose the 69 expos6 of the fixed-interest-rate theorists. This paper points out that fixed-rate theorists have concluded that had interest rates on VA and FHA mortgage loans been free to fluctuate in response to changing market forces, there would have boom little or no fluctuations in housing starts during the post-World War II period. It was the objective of this paper to show why fixed interest rates are not the key to the postwar housing cycles and to present an alternative explanation of the fluctuations 70 in housing starts. The author's basic thesis was: 67Ibid., p. 297. 68Ibid. 6®William W. Alberts, "Business Qydes, Residential Construction Qycles, and the Mortgage (EEVIsSifeauussi SS2B2at, 70Ibid.. p. 264. 35 • . . business cycles have generated housing cycles not because of fixed rates but primarily because the changes in incone associated with periods of business contraction and recovery have had a relatively snail direct effect on the denand for housing.71 It was concluded that four conditions linked the postwar fluctuations in aggregate spending with the fluctuations in the nunber of single-fanily housing starts. These were listed as follows: . . • (a) a high cross-elasticity of denand for nortgages with respect to yields on conpeting investnents on the part of lenders; (b) a denand schedule for nortgage funds that has been rela tively stable over the course of each recession and recovery; (c) an elastic supply schedule of new houses; and (d) an elastic denand schedule for nortgage funds.72 A review of the foregoing tends to indicate that econonists generally have been writing in terns of aggregate or nacroecononics. Therefore, as indicated earlier, it was felt that a particular contribution night be nade by an investigation which starts at the national level and then analyses the various econonic sectors at their respective lower levels. 71Ibid. 72Ibid., p. 281 V. THE METHOD OF INVESTIGATION 36 The underlying hypothesis is tested and analyzed through the use of graphic and statistical analyses, including correlation analyses of both the published data and the additional empirical Sacramento Metro politan Area data developed for this investigation. The general plan is to proceed from an analysis of the economic factors which influence the demand for real estate credit at the national level first, then the regional and selected states level, and finally the selected metropolitan area levels including the Sacramento Metropolitan Area. The relationship of employment, income, and the availability of credit for new residential construction is examined at the various levels. Statistical analyses, including correlation analysis, are made where appropriate. The relationship of the demographic factors, namely, population changes to new residential construction, is also examined at the various levels. Similarly, analyses are made of the factors which influence the supply of real estate credit at the national, regional, selected states, and selected metropolitan area levels. 37 VI. ORGANIZATION OF REMAINDER OF THE STUDY There are three parte in the remainder of the study. Part B, which contains Chapters II and III* discusses the denand for nonfarn residential con struction nortgage credit; Part C discusses the supply side of real estate credit in Chapters IV, V, and VI; and Part D, Chapter VII, enconpasses the sunnary and conclusions of the investigation. Chapter II, entitled "The Influence of Econonic Base Changes on New Nonfarn Residential Construction," discusses econonic base theory and the effect of econonic base changes on nonfarn residential con struction starts. Changes in enploynent levels and per fanily incone in the various regions, states, and netropolitan areas are utilized as indications of changes in the econonic base of each of the respective areas. These changes are then related to changes in the rate of new nonfarn residential construction at these various area levels in order to help bring this sector of econonic theory into its proper perspective. Correlation analysis is used to test the relationship between two variables, per fanily incones and nonfarn residential construction starts, in the four regions of the United States, eleven selected states, and 38 twelve selected metropolitan areas. The influence of selected demographic factors on the rate of new nonfarm residential construction is examined in Chapter III. This chapter discusses the relationship of population growth, migration, and the demand for new housing in the various economic areas of the United States. On the supply side of real estate credit, Chapter IV, entitled "The Influence of the Money Rate and Money Availability on Housing Starts," explores the relationships between new nonfarm residential construction activity, the cost of money, and the availability of funds for real estate credit. Correlation analysis is an integral part of this chapter. The independent variables, money rate and credit availability, and the dependent variable new nonfarm residential construction activity are examined for the existence of causal relationships. As in other chapters of this investigation the analyses of the relationships are developed for the regional, selected states, and selected metropolitan area levels. Chapter V, "Major Sources of Residential Real Estate Credit,” describes and analyses the important financial institutions which supply funds for new nonfarm residential construction in the United States. This is accomplished by a brief analysis of the structure and growth of the savings and loan industry in California and the United States. A comparative analysis of mortgage lending activity of the major financial institutions during the 1950-1960 period reveals the competitive shifts which have occurred in the relative importance of these institutions as suppliers of residential mortgage credit. This chapter also discusses the impact of credit avail ability at savings and loan associations on residential construction. The first portion of Chapter VI presents an analysis of empirical data which relates the impact of structural changes and growth of the savings and loan industry in the Sacramento Metropolitan Area on residential construction in the area. The latter portion of this chapter reports the findings of a survey which examined the effects of changes in the Federal Reserve Board's monetary policy on the mortgage lending activities of local financial institutions. The objective of Chapter VI is to test that portion of the hypothesis which states that the increased rates of residential construction are dependent upon the ability and willingness of savings and loan associations to provide funds for real estate credit. 40 The final part of this study, Chapter VII, presents the salient points discussed earlier in the investigation and summarizes the impact of credit availability at savings and loan associations on residential construction in the Sacramento Metro politan Area. Reported in the conclusions section of this chapter are inferences concerning the implica tion nationally of the findings. The investigation is concluded with some suggested areas of needed research. PART B THE DEMAND FOR NONFARM RESIDENTIAL CONSTRUCTION MORTGAGE CREDIT CHAPTER II THE INFLUENCE OF ECONOMIC BASE CHANGES ON NEW NONFARM RESIDENTIAL CONSTRUCTION I. INTRODUCTION The deaand for real estate credit is a derived deaand. That is, the deaand for reed, estate loan funds is derived fron the deaand for real estate. This is true because few housing units are purchased with total equity financing, and in aost instances borrowed funds aust be advanced in order to assist in naking the desire for real estate becoae an effec tive deaand. Generally, the deaand for real estate is influenced by factors which aay be classified into two broad categories, naaely, deaographic factors and econoaic factors. The chief deaographic factors which influence the deaand for real estate and, hence, the deaand for real estate credit are population changes and net household foraation. These are dis cussed in Chapter III. The iaportant econoaic factors which affect the deaand for real estate and hence the deaand for real estate credit are (1) eaployaent levels, (2) incoao levels, (3) the quantity and the quality of available housing, and (4) the availability of capital to finance the construction of new housing 43 units or to assist in the purchase of existing housing units* The influence of the availability of capital to finance new residential construction is discussed in Chapter IV. Subsequently, the following sections in the present chapter cover the effect of changes in enploynent and incone levels, both being key sectors in econonic base analysis, on regional, state, and local new nonfarn residential construction starts. Also, in this chapter enpirical data on the econonic base of the Saeranento Metropolitan Area are related to residential construction in the area. E c o n o a i c Base Theory By definition, the econonic base of an area, whether the area be a nation, region, state, or standard netropolitan area, is the sun total of all econonic activity in that area which produces enploy nent and incone. Briefly stated, the underlying concept of econonic base analysis is that a national, regional, or standard netropolitan area of relatively greater econonic opportunity will tend to draw popula tion away fron areas of relatively less econonic opportunity. That is, as basic econonic activity expands in an area, as opposed to stability or decline in other areas, workers are attracted fron areas of 44 lesser or declining econoaic opportunity. Thus, the urbanization of these workers in the dynaaic area leads to additional econoaic opportunities for other workers particularly in the nonbasic or service industries. These nonbasic workers will provide the additional goods and services required by the original population, the expanded basic industry eaployees and their fanilies. Consequently, as both basic and nonbasic eaployaent in the area expand, still aore population will be attracted to the area by the increasing nuaber of econoaic opportunities.^ Econoaic activity aay be aeasured in terns of eaployaent, incone, consuaption of goods and services, industrial developnent and its subsequent increases in land values, gross product, or value added by o Manufacture• In this analysis econonic activity is See Ernest M. Fisher and Robert II. Fisher, Urban Real Estate (New York: Henry Holt and Coapany, ldS4), Chapter XII; also, Richard U. Ratcliff, Urban Land Econoaics (New York: McGraw-Hill Book Coapany, Incorporated, 1947), Chapter II. o The tera "value added by Manufacture” desig nates a neasure of value that is derived by subtracting the cost of raw Materials, senifinished parts and coapenents, supplies, fuels, purchased electric energy, and contract work fron tho value of shipnents. Neither gross product nor value added by Manufacture aeasure- aents are utilised in this analysis because of the extreae difficulty of obtaining the necessary data on a local, state, or evon a rogional basis. 45 measured, first, in terms of employment, which is perhaps the most widely used measure at the local levels, and second, in terms of income. While these measures leave much to be desired as accurate indi cators of the changes in the relative economic importance of basic to nonbasic industry employment in the economic base of an area, they are generally the most available data. Both employment and income data express growth or decline in terms of job oppor tunities. Furthermore, "it is the expansion of such opportunities which leads to immigration and population growth; job reductions result in unemployment, emigra tion, declining local income, and falling property values." In terms of the broad aggregates for which data are available, considerable economic growth has been reported in various econoaic areas in the United States. The reported econoaic growth has been, in part, at the expense of the slow-growth econonic areas. This may be explained by the fact that the most significant factor which determines the distribu tion of population is ocononic opportunity, and the most significant measures of econoaic opportunity are 3Richard U. Ratcliff, Real Estate Analvaia (New York: McGraw-Hill Book Company, Incorporated, 1961), p. 248. 46 employment and income opportunities. Ultimately, changes in the economic base will affect the rate of new nonfarm residential construction in the area. However, it appears that economic doc trine is unclear as to the extent of the influence of changes in the economic base on residential construc tion. If the economic base is the link between the nation, regions, states, and the differentiated local residential construction markets, then changes in the economic base would be expected to result in changes in the rate of local residential construction. According to prevailing theory, differences in the demand for housing in an area are explained by changes 4 in basic industries in specific areas. The strengths and weaknesses of the basic industries and the degree of diversification cure considered as being factors in the differentiated local, state, and regional resi dential construction markets. However, in the past, economic theory has paid primary attention to long-run national aggregative data, and as a result housing policy has been established with little consideration for local forces which influence local residential *C. Rapkin, L. Winnick, and D. Blank, Hamln^ Market Analysis. Housing and Home Finance Agency (.Washington: Government Printing Office, 1953), Chapter V, pp. 41-56. 47 construction activity. This chapter utilizes changes in both eaployaent levels and per faaily incoae levels on a regional, state, and aetropolitan area basis as indicators of changes in their respective econoaic bases and relates these changes to changes in the rate of new nonfarm residential construction at these various levels in order to help bring this sector of econoaic theory into its proper perspective. II. REGIONAL, SELECTED STATES, AND SELECTED METROPOLITAN AREA EMPLOYMENT LEVEL CHANGES COMPARED WITH NONFARM RESIDENTIAL HOUSING STARTS One of the significant characteristics of aodern econoaic activity is the high degree of specialization which has occurred within the various econoaic areas of the United States, as well as the other industrialized countries of the world. Basic to a high degree of specialisation has been the developaent of an extensive network of enterprises which facilitate the flew of goods and services. As a result, the econoaies of these various econoaic areas have becoae highly interdependent. The aggre gate of all the interdependent econoaies in the United States has produced a single national economy 48 involving complementary interrelated parts. Conse quently, national developments influence all areas of the nation in varying degrees over tine. It is unlikely that any one area can insulate itself completely from the effects of major national cyclical swings in general business activity. Regional Responses to Changes in the Economic Base Evidence of the inability of any one area to be completely free from major national cyclical swings in general business activity may be found in available regional data. Pertinent regional employment data may be utilized to test the econoaic base theory relating changes in regional employment with changes in regional nonfarn residential housing starts during two recessionary periods which weakened the econoaic bases of many areas in the United States in the 1954- 1960 period. Figure 1® graphically illustrates that 5 Sources for Figures 1 to 8, inclusive, are as follows* Employment data are computed from United States Department of Commerce, Bureau of the Census, Statistical Abstract of the United States (Washington: GovernmentPrinting OfTice, 195^-1961). Starts data are frem United States Department of Commerce, Con struction Review (Washington: Government Printing 0?Hce7 1955-15*0). 49 +40 +30 +20 +10 0 -10 -20 -30 -40 ia ID I*- 00 O i O IA < D r - 00 O i O IA IA IA IA IA < D IA IA IA IA IA (0 i 1 1 1 1 1 1 1 1 1 1 1 IA (0 00 Oi IA (0 00 O i IA IA IA IA IA IA IA IA IA IA IA IA Northeast North Central South Wait Figure 1. Percentage changes in regional nonfarn euploynent and housing starts, 1954-1960. Qfisploynent | Nonfarn Residential Starts 50 in the four regions of the United States, during and isusediately following contractions in general business activity in both the 1953-1954 and the 1957-1958 recessions, the nuaber of nonfarn residential housing starts increased over the nunber of starts reported for the years innediately prior to the peak in general business activity. On the basis of the reported year- end data, the first conplete year after the trough in general business activity had been reached reflected the greatest positive changes in both eaploynent and starts in all four regions. In general, the two regions which reported the greatest percentage fluctuations in eaployaent also reported the greatest percentage changes in housing starts during the 1954-1960 period. That is, it appears that the Northeastern and the North Central Regions, in teras of percentage changes in eaployaent, reported greater volatility in the nuaber of housing starts than did the other regions. On the other hand, the Southern and Western Regions, which reported auch less deviation in their respective eaployaent levels, reported auch less volatility in their respective housing starts levels. It nay be noted, therefore, that a degree of correlation exists between changes in regional eaployaent and regional nonfara residential 51 construction starts. Housing Starts Rssnonso to Econonic Bass Changes in Sslsctsd States An examination of the relevant data at the state level reveals that a similar correlation exists between changes in employment and nonfarm residential con- struction starts in each of the selected states. These data are shown graphically in Figures 2, 3, 4, and 5 wherein changes in the level of employment and housing starts in several states, within each of the four regions, are plotted. Selected states in the Northeastern Region. Each of the four selected states in the Northeastern Region have reported fluctuations in both levels of employment and housing starts during the 1954-1960 period (Figure 2). Following the 1953-1954 business recession, increases in nonagricultural employment occurred at decreasing rates of growth each year until 1958, when each of the four states reported a fi The selected states were chosen on the basis of those important states within the region for which the most complete and comparable data were available for the several analyses made in this study. 52 r - O i O IO <0 r - 3 IO ft IO $ IO IO IO i i I T 1 1 1 1 <o to p» IO 3 O i IO 3 IO IO <0 IO S ) 8 o> w ■HO Massachusetts New Jersey- New York Pennsylvania Figure 2• Percentage changes in nonfare eeploy- eent and housing starts: Massachusetts, New Jersey, New York, and Pennsylvania, 1954-1960. QEeployeent |Nonfare Residential Starts 53 decrease in the level of eaployaent. Coincident with increases in eaployaent at decreasing rates, housing starts in each of the four states declined* Three of the states— Massachusetts, New Jersey, and New York— reported increases for the first tine since 1955 in their state's level of housing starts at year-end 1958. Pennsylvania, which had the largest negative percentage change (-5*2 per cent) in eaployaent of the four selected northeastern states, did not report a positive increase in annual housing starts until 1959* The two states, New York and New Jersey, which reported nega tive changes in annual eaployaent in only one out of the six years, had the strongest residential construc tion Markets in teras of annual housing starts changes; whereas Massachusetts and Pennsylvania, which had both reported decreases in annual eaployaent levels twice during the 1954-1960 period, had the weakest residential construotion aarkets Measured in teras of the greatest fluctuations in annual housing starts. Selected states in the North Central Region. Except for the first full year after the 1953-1954 recession when percentage change increases in both annual oaployaent and housing starts in Illinois, Michigan, and Ohio wero reported, tho three selected 54 states in the North Central Region recorded changes of different magnitudes and direction in their respective levels of annual employment and housing starts. As shown in Figure 3, Ohio, which did not report any negative change in the level of annual employment during the six year 1954-1960 period, had the least amount of fluctuation in annual residential construction starts. However, for the years in which the greatest positive percentage changes in annual employment occurred, Ohio reported its strongest residential construction markets, measured in terns of positive percentage increases in housing starts. Conversely, for the three years during the period when Ohio had its weakest employment markets, Ohio also reported its weakest residential construction starts activity. Employment in Ohio increased 3.7 per cent in 1955, 7.1 per cent in 1958, and 4.8 per cent in 1959. During the same years housing starts in Ohio increased 11.0, 12.0, and 17.0 per cent, respectively. During each of the years in which eaployaent increases did not equal or exceed the previous year's increase, housing starts declined in Ohio. That is, in 1956 eaployaent increased only 2.9 per cent, whereas a 3.7 per cent increase was reported in 1955; in 1957 the Increase was only 55 c <0 r- 8 Oi tO n IO to 1 I i l 1 s to (O t* 00 to in to to ? ? O t to to <o s at to to to to 1 1 1 1 i x 0 < 0 t * 9 0 to to 0 Illinois Michigan +20 +10 0 -10 -20 IO <0 t ' - GO 0> Q 0 IO to <6 0 <5 1 I i i i T 0 to r- ao w 0 0 0 0 0 0 Ohio Figure 3. Percentage changes in nonfara eaployaent and housing starts: Illinois, Michigan, and Ohio, 1954-1960. [jEaployaent | Nonfara Residential Starts 56 0.3 per cent, compared with the 2.9 per cent increase of 1956; and in 1960 the reported increase was only 1.3 per cent, which was considerably less than the 4.8 per cent increase reported in 1959. Meanwhile, housing starts in Ohio decreased 16.0 per cent in 1956, 16.0 per cent in 1957, and 11.0 per cent in 1960. The pattern of changes in eaployaent and housing starts activity in Illinois reflected that of the North Central Region, which is understandable, since activity in Illinois does have a considerable influence on the total activity in the region. Thus, changes in the economic base of Illinois and the North Central Region, as measured by changes in annual employment levels, were coincident with percentage changes in housing starts. In contrast, Michigan, which reported the greatest fluctuations of the three states in terns of annual employment, had negative percentage changes in housing starts in five of the six years immediately following the 1953-1954 reces sion. This may reflect the fact that Michigan had not fully recovered from the effects of that recession. Selected states in the Southern Resion. The pattern of the contracydical behavior of nonfarm residential construction activity during the 1954—1960 57 period, which was evident both at the regional level and in the six states, discussed earlier, is also discernible in the states of Georgia, Maryland, and Virginia. As shown in Figure 4, Maryland and Virginia appear to have had the greatest fluctuations in annual eaployaent percentage changes as well as the greatest swings in annual residential construction starts. Georgia, whose annual eaployaent percentage changes were negative in two out of the six years, reported negative percentage changes in housing starts in four of the saae years; however, the fluctuations in its annual housing starts appear to be the least of the three southern states. Virginia, which also had two years when negative annual eaployaent percentage changes were reported, had greater increases and decreases in its annual housing starts percentage changes. Selected states in the Western Region. Both California and Washington influence the Western Regional pattern. California, on the one hand, has boon relatively aore stable in both annual eaployaent and annual housing starts. Washington, on the other hand, being loss diversified in eaployaent activities, reported both saaller positive and larger negative 58 •a to r* to C5 1 0 t S O) Q 1 0 <5 I I I I I to to oo e» to to to w to r> oo a to to to IO to to I f ) I I I I I S IO tO I * - 00 I f ) I f ) I f ) I f ) ? O) to >30 >20 >10 0 -10 -20 -30 Georgia Maryland n to h oo w to to 55 i i r i e* to to i* IO I f ) I f ) I f ) at o n to I at to Virginia Figure 4. Percentage changes in nonfara eaployaent and housing starts: Georgia, Maryland, and Virginia, 1954-1980. QEaployaent | Nonfara Residential Starts 59 eaployaent percentage changes and greater positive and negative housing starts percentage changes than California. Figure 5 graphically illustrates these coaparisons. Following the 1954 recession, eaployaent in California in 1955 increased 4.4 per cent and housing starts increased 7.0 per cent. In Washington eaployaent increased only 3.7 per cent and housing starts decreased 2.0 per cent in 1955. The following year eaployaent in California and Washington increased 8.3 and 3.2 per cent, respectively; whereas housing starts dropped 17.0 per cent in California and 28.0 per cent in Washington. With the start of the 1957- 1958 recession, eaployaent increased 3.0 per cent in California and 2.5 per cent in Washington during 1957. Housing starts decreased in both states in 1957, 4.0 per cent in California and 6.0 per cent in Washington. At year-end 1958, annual eaployaent level deereases of 0.6 and 1.5 per cent were reported by California and Washington, respectively. Meanwhile, 1958 housing starts in California and Washington increased 16.0 and 36.0 per cent, respectively. In 1959 both California and Washington had positive percentage changes in eaployaent and housing starts. During the 1960 recession, California reported its second highest percentage increase in eaployaent (5.8 per cent) for 60 IO <0 s 0» © IO IO IO IO <5 1 1 1 1 1 1 3 IO IO <0 IO I* IO 8 0» IO +20 +10 0 -10 -20 California +40 +30 +20 +10 0 -10 -20 -30 -40 Washington Figure 5. Percentage changes in nonfara eaployaent and housing starts: California and Washington, 1954-1960. Q Eaployaent | Nonfara Residential Starts 61 the study period and housing starts in that state decreased only 18.0 per cent. In 1960 annual eaploy- ■ent in Washington increased only 2.1 per cent, which was less than one-half the percentage increase in California's annual eaployaent; however, housing starts decreased 36.0 per cent in Washington, twice that of the decrease in California housing starts. Housing Starts Response to Econoaic Base Changes in Selected Metropolitan Areas The econoay of the United States aay be con sidered, for purposes of analysis, to be a closed econoay coaposed of aany saaller econoaic areas, each of which is considered to be an open economy. Thus, the econoaic activity of the whole is equal to the sua total of the econoaic activity of all its parts. To arrive at the total, however, increases or decreases in the econoaic activity of one area aay be ainiaised by the offsetting decreases or increases of another coaparable area at the aetropolitan area level, state level, or regional level. The discernible pattern of fluctuations in annual eaployaent levels and changes in annual residential construction starts which was visible at the regional and selected states levels continues to exist at the selected Metropolitan area 7 level. Figures 6, 7, and 8 illustrate the percentage changes in eaployaent and housing starts in selected Metropolitan areas within the Northeastern and North Central, the Southern, and the Western Regions, respectively. The analysis afforded by these figures appears to be consistent with econoMic base theory. That is, both the regional and selected states sectors of the econony are cowposed of Many Metropolitan area bases, and hence local fluctuations in econonlc activity are conbined with the fluctuations of econoMic activity of other Metropolitan areas within the state and the region. However, changes in the econoMic base of any one Metropolitan area will have a greater effect on the annual housing starts within that saae area. Consequently, changes in eaployaent levels aay lead to considerable changes in annual housing starts during the short cycle of business activity. Therefore it is to be expected that greater fluctuations aay occur at the local level than on a state or regional level. 7 As with the selected states, the decision to include a particular Metropolitan area in this study was based on (1) the influence of the local area within the state and, subsequently, the region; and (2) the availability of coaparable data at the local level. 63 Metropolitan area reaponeea in the Northeaatern and North Central Redone. Three aetropolitan areaa in each of the two northern regiona were aelected for thia analysis: Boston, Buffalo, and New York-north- eastern New Jersey in the Northeaatern Region; and Chicago, Cleveland, and Detroit in the North Central Region (Figure 6). During the 1954-1960 period Boston, Buffalo, and New York, which had succeedingly snaller percentage increases in annual eaployaent froa 1954-1958 when percentage decreases were reported in all three areas, reported a downward aoveaent in residential construction starts. In fact, the Buffalo Metropolitan Area reported percentage decreases in the nuaber of annual housing starts in each of the last five years of the six year period. This indicates that a strong degree of correlation aay exist between changes in annual eaployaent levels and housing starts in the Buffalo Metropolitan Area. Also, the New York Metropolitan Area, which had the weakest eaployaent aarket in teras of positive percentage changes, and the only one of the three areas with two negative changes in the six year period, had the greatest fluctuations in housing starts of the three areas. Data for the highly industrialised Detroit Metropolitan Area, which had considerable influence NORTHEAST to <0 o - oo a o to to to IO IO <0 I I I I I I to <0 I ' - 00 A IO IO IO IO IO IO Boston IO IO Is- 00 A O IO IO IO IO IO ID I I t i l l * io «o r>» oo a IO IO IO IO IO IO Buffalo io io r> oo a o io IO IO IO IO IO i l l i t i S IO <0 t * > 00 A io io m io . to Nov York- Northoast Nov Jersoy +20 Chicago NORTH CENTRAL Clovoland Dotroit Figaro 6. Poreentago changes in nonfara eaployaent and housing starts: Boston, Buffalo, Nov York-Northeast New Jersey, Chicago, Cleveland, and Detroit, 1954-1960. Q Eaployaent fNonfara Residential Starts on the econoay of the state of Michigan in the North Central Region, effectively illustrate that a strong degree of correlation aay be evident between a down ward shift in annual eaployaent levels and a downward shift of annual residential construction starts. During each of the first of the six years in the 1955-1960 period, Detroit reported decreases in annual eaployaent. It was not until 1960 that a 1.02 per cent increase in annual eaployaent was reported in Detroit; however, in none of the six years was a percentage increase in housing starts reported. The range of annual eaployaent percentage change was froa -13.45 to +1.02 per cent. In the saae period the range of annual housing starts percentage change was froa -31.0 to -1.0 per cent. The pattern of fluctuations in both the aore diversified Chicago Metropolitan Area and Cleveland Metropolitan Area was aore siailar to that of Boston and Now York. Of the foraer two areas, Chicago appears to have had a stronger eaployaent aarkot in teras of percentage changes in annual eaployaent, and as a result enjoyed a stronger residential con struction narket in terns of less volatile percentage changes in annual housing starts. 66 Metropolitan aroa reaponsea in the Southern and Western Regions. Atlanta and Baltimore were chosen for the analysis in the Southern Region. Also, the data for four Metropolitan areas— Los Angeles, San Oiego, San Francisco-Oakland, and Sacr&aento— were plotted for the Western Region analysis. These data are shown in Figures 7 and 8. Finally, the pertinent eapirical data of the Sacraaento Metropolitan Area are discussed in the last section of this chapter. In the Southern Region of the United States neither the Atlanta nor the Baltiaore Metropolitan Areas appear to have developed eaployaent and housing starts fluctuations patterns siailar to the discernible pattern of the region. However, Atlanta, which appears to have had the greatest positive annual eaployaent percentage changes, also had the greatest positive annual housing starts percentage changes. The Balti aore eaployaent narket appears to be the weaker of the two in teras of positive percentage increases, and it follows that the housing starts narket would also be the weaker of the two as evidenced by the housing starts percentage change perfornance in that area. The range of Baltiaore annual eaployaent per centage change is froa +4.7 to -2.9 per cent, whereas that of Atlanta is +7.1 to -2.2 per cent. Consistent 67 SOUTH io <o t>. o> n IO IO So IO i I 1 1 1 IO (O 00 IO IO IO IO IO +30 +20 +10 O -10 -20 -30 Atlanta +10 0 -10 -20 -30 -40 Baltiaore Figure 7. Percentage changes in nonfara eaployaent and housing starts: Atlanta and Baltiaore, 1954-1960. ^Saployaent |Nonfara Residential Starts 68 WEST io <0 00 < n o IO (0 00 0> IO IO IO IO to <o IO IO IO IO to i 1 1 1 i i 1 1 1 1 1 3 IO IO <0 IO r* IO 8 o> IO s IO IO (0 IO I* IO 8 o> IO +110 +100 +90 +80 +70 +60 +50 +40 +30 +20 +10 0 -10 -20 -30 -40 Los Angeles Sacraaento +40 +30 +20 +10 0 -10 -20 -30 -40 -50 San Diego San Francisco- Oakland Figure 8. Percentage changes in nonfara eaployaent and housing starts: Los Angeles, Sacraaento, San Diego, and San Francisco-Oakland, 1954-1960 Q Eaployaent £Nonfara Residential Starts 69 with these reported ranges were the housing starts percentage change ranges. In the Baltiaore Metro politan Area the range of housing starts percentage change is froa -37.0 to +5.0 per cent (note that there were five negative percentage changes during the six year period), and in Atlanta the range is froa -20.0 to +35.0 per cent. Of the four selected aetropolitan areas in the Western Region the San Francisco-Oakland area pattern of fluctuations in eaployaent and housing starts appears to be aost consistent with that of the region. However, both San Francisco-Oakland, except for one year, and Los Angeles reported negative percentage changes in annual housing starts during the years in which the annual eaployaent levels did not equal that of the prior year. That is, in the Los Angeles area annual eaployaent levels were lower in 1956 than in 1955, in 1957 than in 1956, in 1958 than in 1957, and in 1960 than in 1959; and in each of those years Los Angeles reported negative percentage changes in annual housing starts. Siailarly, in the San Francisco- Oakland area eaployaent and housing starts levels were lower in 1957 than in 1956, and In 1960 than in 1959. San Diego, on the other hand, had greater percentage increases in both eaployaent and housing 70 starts throughout almost all of the six year period. Furthermore, San Diego reported the greatest decrease in annual employment percentage increases of all three areas in 1960; and it also suffered the greatest decrease, more than 50.0 per cent, in housing starts in 1960. In Los Angeles the range of annual employment percentage change was from +8.3 to >1.5 per cent during the six year period. The range of housing starts percentage change in that area was from -14.0 to +12.0 per cent. During the same period in the San Francisco- Oakland area the percentage change in annual employment ranged from +3.9 to -2.2 per cent. Housing starts percentage change in San Francisco-Oakland ranged from +34.0 to -34.0 per cent. The range of annual employ ment percentage change in San Diego was from +16.9 to -0.9 per cent, compared with the range of housing starts percentage change of +40.0 to -52.0 per cent during the 1954-1960 period. Econoaic opportunities in the Sacramento Metro politan Area, unlike any other selected area in the study, increased in each of the years between 1950 and I960. Housing starts in the Sacramento Metropolitan Area, as in all other areas, however, were subject to wide fluctuations during this same period. More specifically, the narrow range of Sacramento annual 71 employment percentage change was from +3.9 to +9.5 per cent during the shorter study period years of 1955 to 1960. During these years housing starts percentage changes ranged from -39.8 to +108.0 per cent in the Sacramento area. As shown in Figure 8, page 68, the direction of annual housing starts per centage changes in the Sacramento area coincide with the direction of annual housing starts percentage changes in both the Western Region and the state of California; the one difference, however, being in the magnitude of the annual housing starts percentage change. As pointed out earlier, it is to be expected that greater fluctuations may occur at the local level than on a state or regional level. The discernible pattern of fluctuations in annual enployment levels and changes in annual residential construction starts, which was evident at the various regional, selected states, and selected metropolitan areas levels, is also apparent in the Sacramento Metropolitan Area. That is, except for 1956, in each of the years when annual employment percentage level changes were higher than the previous year, housing starts increased. During the years when employment level percentage changes did not increase as much as the prior year, housing starts decreased. 72 In summary, annual housing starts in the selected metropolitan areas, as in the selected states and regions of the United States, reflected diversified magnitudes of reactions to changes in their economic bases, as measured in terms of changes in the level of annual employment in the area. A conclusion may be drawn from the available data that variations in the economic base have had a prominent role in the determination of the level of annual housing starts during the short-term cycle. III. THE RELATIONSHIP OF NONFARM HOUSING STARTS AND PERSONAL INCOME The purpose of this section is to discuss the effect of income on nonfarm housing starts. The improvement in real personal income has brought home ownership within reach of an increasing percentage of nonfara families during the last decade. Increased incomes have made it possible for many aore families to qualify for either conventional or government- underwritten mortgage loans. This growth in real income or purchasing power during the 1950's has been traced to an advancing technology and to the unprece- 73 Q dented prosperity of the United States. Of the several Measures of incoae available for housing Market analyses, faMily incoae data are More iMportant than per capita incoae data because Most houses are bought by faMilies. It is the incoMe of the faMily spending unit which generally influences the effective deaand for housing. For analytical pur poses , however, it is difficult to quantify the rela tionship between personal incoae, aore specifically per faaily incoae, and nonfara housing starts. One of the statistical tools available for econoaists to coapare the degree of relationship between a pair of variables is correlation analysis. The coefficient of correlation (r) is an abstract Measure indicating the degree of relationship between two variables, where unity indicates perfect correlation and sero indicates the absence of any correlation. This statistical technique, therefore, was used to test the relationship between the two variables, per faaily incoae and the nuaber of nonfara housing units author ised (building peraits issued data), for the four regions of the United States, the eleven selected a Preston Martin, Real Estate Principles and Pract|c|S (New York: The Macaillan Coapany 719557, 74 states, and the twelve selected Metropolitan areas. The results of these correlation analyses, as was expected, failed to develop any discernible pattern of relationship between the two variables, per faaily incoae and nonfara housing units authorized, in any of the various areas. The findings of the correlation analyses are reported in Table 1. An earlier review of the pertinent literature had led to the conclusion that the findings night not indicate a direct cause-and-effect relationship but that the relationship night be aore of an indirect Q nature. One noted authority has pointed out that changes in per faaily incoaes produce certain secondary factors which influence housing deaand in the short-run. These secondary factors involve the doubling and undoubling of households as well as "deaand filtering." Furthernore, data plotted on a scatter diagraa which related private nonfara residential construction expenditures and disposable personal incoae, both on a per household basis, in 1930, 1940, and 1947-1957, led to the conclusion that this conparison "reveals no very stable relationship."10 This authority also 9Ibid., pp. 39-41. 10John P. Lewis, Business Conditions Analysis (New York: McGraw-Hill Book Coapany, Incorporated, 1959), p. 441. TABLE 1 CORRELATION COEFFICIENTS OF PER FAMILY INCOME AND NONFARM HOUSING UNITS AUTHORIZED, SELECTED AREAS, 1954-1960 Rations r Selected States r Selected Metropolitan Areas r Northeast -.26 Massachusetts -.30 Boston .04 North Central -.39 New Jersey -.77 Buffalo -.97 South .23 Nee York .19 New York-New Jersey -.23 West .36 Pennsylvania -.66 Chicago -.54 Illinois -.55 Cleveland -.47 Michigan -.85 Detroit -.68 Ohio -.53 Atlanta .36 Georgia .18 Baltimore -.87 Maryland -.44 Los Angeles -.46 California .18 San Diego • 36 Washington -.41 San Francisco-Oakland .14 Sacramento .22 Source: Computed froa data provided by United States Department of Commerce, Bureau of the Census and Sales Management (annually). 76 concluded that for short-run forecasting purposes it would be nore reasonable . . . to view disposable personal incone as a general conditioner of housing deaand which, within the space of a year or so, is unlikely to exort any markedly distinctive influence of its own oxcopt on thoso occasions whon a narked change of course in inconos or incono expectation appears to be in store.11 Further insight into this apparent lack of a stable relationship between changes in incones and nonfara residential construction outlays nay be gained fron the statenent that "in consideration of the high incones and the relatively lower preference for housing, it does seen reasonable to conclude that average con- 12 suaers are underinvesting in their hones." Earlier 13 a tean of authorities had suggested that the long- tera decline in real capital investment per new housing unit night be explained in terns of the incone elas ticity of families and the price elasticity for housing expenditures. Therefore, it night be concluded that, 11Ibid. 12 E. Bryant Phillips, Consumer Economic Problems (New York: Henry Holt and Conpany, lfes?), p. 294. *3Leo Grebler, David II. Blank, and Louis Winnick, Capital Formation in Boaidantial Beal Estate (Princeton, NewJersey: Princeton University Press. 1956), Chapter VIII. 77 •specially in the short-run, "there is little point in trying to generalise about the degree of responsive ness of residential construction to year-to-year incoae 14 changes." Consequently, the absence of statistical evidence which relates changes in incoae levels to changes in nonfara residential starts does not invali date the hypothesis. However, it does require that greater reliance be given to changes in eaployaent levels. This was shown to be consistent with econoaic base theory in the first section of this chapter. IV. CHANGES IN THE ECONOMIC BASE OF THE SACRAMENTO METROPOLITAN AREA RELATED TO CHANGES IN LOCAL RESIDENTIAL CONSTRUCTION STARTS Econoaic growth characterizes the secular trend of the national econoay. Many areas of the nation have experienced iapressive econoaic expansion, par ticularly since World War II. This is true of the Sacraaento Metropolitan Area where, especially since 1950, industrial, coaaercial, and governaent activities have expanded greatly. As a result of the investaent which has taken place in a great variety of activities within the Sacraaento Metropolitan Area, eaployaent 14 Lewis, op. cit., p. 442. 78 and incoae opportunities have been created in the economic base of the area. faninvint Growth in the Sacraaento Econoaic Base Figures on eaployaent are the aost relevant for analysing the growth of the Sacraaento economic base. As the data in Table 2 indicate, the Sacraaento labor force increased by 76,000 workers, from 122,600 to 199,300, or 61.8 per cent during the 1950's; whereas the labor force in the state of California increased only 40.4 per cent, froa 4,700,000 in 1950 to 6,600,000 in I960. In effect, the Sacraaento labor force increased at a rate of one and a half tines that of the state as a whole. It is interesting to note that in this sane tine period the Sacraaento labor force, as a percentage of the aetropolitan area population, decreased froa 44.3 to 39.6 per cent; whereas the total labor force in the state of California declined froa 42.6 to 40.2 per cent of the population. Further insight into the labor force and the econoaic base is gained by an analysis of Sacraaento's basic industry and nonbasic industry eapleyaent data. Basic industries are defined as those enployaents which bring incoae into an econoaic area* The expan sion of these basic industries is a priaary cause of TABLE 2 CALIFORNIA AND SACRAMENTO METROPOLITAN AREA LABOR FORCE AS A PERCENTAGE OF POPULATION, 1950, 1953, 1956, 1959, AND 1960 Year Californiaa Sacramento Metrooolitan Areaa Labor Force PoDUlation Per Cent Labor Force Ponulation Per Cent 1950 4,700 10,586 42.6 123 277.1 44.3 1953 5,141 12,168 42.3 144 340.3 42.3 1956 5,664 13,594 41.7 164 404.8 40.5 1959 6,100 15,280 39.9 189 480.7 39.3 1960 6,600 15,717 40.2 199 502.8 39.6 1950-•1961 Labor Force Increase 40.4 61.8 aLabor Force and Population data in thousands. Source: United States Bureau of the Census; State of California, Division of Labor Statistics and Research, Department of Industrial Relations, Research and Statistics Section; Department of Employment. -a 80 the area*8 econoaic growth. Eaployaents classified as nonbasic or service-type industries are largely those that are attracted into an area after the basic industries have indicated an expansion is under way. For purposes of this study basic industries in Sacraaento include agriculture, aanufacturing, federal civilian eaployaent, the ailitary living in the county, state eaployaent, persons engaged in travel trade, and Sacraaento State College eaployaent. Table 3 offers an insight into the changing nature of the econoaic base of Sacraaento. Basic industry eaployaent, as a percentage of total Sacra aento eaployaent, has increased steadily froa 37.8 per cent in 1940 to 42.4 per cent in I960; and con versely, nonbasic industry eaployaent has been decreasing as a percentage of total Sacraaento eaployaent. In absolute nuabers, basic industry eaployaent increased froa 23,900 workers in 1940 to 80,000 in 1960; whereas eaployaent in nonbasic industry grew froa 40,400 in 1940 to 108,800 in I960. Consequently, total Sacraaento eaployaent increased froa 64,300 in 1940 to 188,800 in 1960, or a 192 per cent increase. More revealing than either percentage figures or absolute nuabers are basic to nonbasic industry TABLE 3 SACRAMENTO METROPOLITAN AREA BASIC AND NONBASIC INDUSTRY EMPLOYMENT, 1940, 1950, AND 1960 Year Basic Industry Nonbasic Industry Total Employment Number Per tient Number Per Cent Number Per Cent 1940 23,900 37.8 40,400 62.2 64,300 100.0 1950 44,600 38.9 70,000 61.1 114,600 100.0 1960 80,000 42.4 108,000 57.6 188,800 100.0 Source: Derived froa data provided by the California Department of Eaployaent, Department of Industrial Relations. 82 employment ratios and the changes in these ratios, which lead to the establishment of employment multi pliers and employment-multiplier-change-factors. Employment ratios are useful in identifying past trends; whereas employment multipliers reflect the impact of changes in basic and nonbasic employment as a result of business impulse reverberations throughout the national, regional, and local economies. Therefore, employment multipliers are useful in the projection and prediction of population and employment figures. The ratio of basic to nonbasic industry employ ment, based on total employment, is a quantitative measure of an economy at any one point in time. Sacramento employment ratios reported in Table 4 indicate that in 1940, for each worker in basic employ ment, there were 1.7 workers employed in nonbasic industry employment. In 1950 there were 44,600 basic industry employment workers and 70,000 nonbasic industry employment workers, or a 1:1.6 ratio, which indicates a decrease in the ratio since 1940. This ratio had further decreased to 1:1.4 in 1960, when for each basic industry worker there were 1.4 nonbasic industry workers employed. The more significant enploynent-ratio-change- factor, which indicates the change in the Sacramento 83 TABLE 4 SACRAMENTO METROPOLITAN AREA BASIC AND NONBASIC INDUSTRY EMPLOYMENT RATIOS, 1940, 1950, AND 1960 Year Industry *W>lownt Ratio iAin Baa1c Nonbasic 1950 Monbaslc 1960 NonEa.lc 23.900 46 ',fS6 1:1.7 44.600 w t m 1:1.6 80.000 i6&*m 1:1.4 Sourcs: Derived froa data provided by the California Departaent of Eaployaent, Departaent of Industrial Relations. 84 economic base over time, can be established from the data in Table 5. That iu, the Sacramento basic industry employment-ratio-change-factor, which involves the bulk of the government employment and almost all of the manufacturing employment (Table 6), indicates that for each additional employee hired in the basic sector the nonbasic sector hired only 1.07 additional employees during the 1950 to 1960 time period. When compared to the 1940 to 1950 employment- ratio-change-factor, each additional employee hired in that time period led to 1.5 additional employees in the nonbasic sector; the 1950-1960 factor of 1:1*07 reflects an actual decrease in this important ratio. Thus, greater insight into the economic forces at work on the economic base can be realized by employment- ratio-change analysis than can be obtained from the simple percentage analysis. Sacramento's growth during the 1950's may be explained partially in terns of the increase in government and defense activities in the metropolitan area. As Table 6 indicates, city, county, state, and federal government employment increased from 32,700 in 1950 to almost 60,000 in 1960, which represents a 77.1 per cent increase. Meanwhile, the total number of manufacturing workers increased from 9,100 in 85 TABLE 5 SACRAMENTO METROPOLITAN AREA BASIC AND NONBASIC INDUSTRY EMPLOYMENT CHANGE RATIOS, 1940-1950 AND 1950-1960 n Eiployatnt Eaployaent Year______Industry______Differential______Ratio 1940-1950 flolbiiflc I S 1:1-5 1 9 5 0 - 1 9 6 0 M o S S S i l c §f$& 1:l-07 Source: Derived froa data provided by the California Departaent of Eaployaent, Departaent of Industrial Relations. 86 TABLE 6 WAGE AND SALARY WORKERS IN GOVERNMENT AND MANUFACTURING IN THE SACRAMENTO METROPOLITAN AREA 1950, 1959, AND 1960 1950 1959 1960 1950- 1960 Increase Government 32,700 57,900 59,900 77.1% Manufacturing 9,100 26,800 28,900 217.5 Source: Stato of California, California Labor Statiotica Bulletin Area Supplement. December. 19507 87 1950 to 28,900 in I960, a 217.5 per cent increase. These two sectors, governaent and manufacturing, which account for 44.6 per cent of the 1960 Sacraaento labor force, are discussed in detail in the appendix. In suaaary, although aanufacturing, due to the location of the defense industry, reflected a relative increase in representation in Sacraaento, it was still relatively lacking in representation by 1959. In general, there has been a trend away froa specializa tion toward relative diversification. Governaent, although still a potent force in the local economy in 1959, was relatively less iaportant than in 1953. Sacraaento Econoaic Base Inceae Level Changes Incoae is considered to be an iaportant indi cator of the econoaic activity of an area. Although estiaates of national and state incoae have been available for years, it is only recently that esti aates of incoae for saaller political and econoaic units have becoae available. Two sources, The California Statistical Abstract and the California Chaaber of Coaaerce, have provided estiaates for the state and the aetropolitan area, as reported in Table 7. The data indicate that the 1959 total TABLE 7 CHANGES IN TOTAL PERSONAL INCOME, UNITED STATES, CALIFORNIA, AND THE SACRAMENTO METROPOLITAN AREA, 1950 AND 1955-1959 1950 1955 1956 1957 1958 1959 United States Personal Income. Dollar Increase b Percentage Increase 1950-1959 Increase 9227.1 67.396 9305.9 9 78.8 34.7 9326.9 9 21.0 6.9 9348.4 9 21.5 6.6 9359.0 9 10.6 3.0 9380.1 9 21.1 5.9 Calif omiaa Personal Income^ Dollar Increase b Percentage Increase 1950-1959 Increase 9 19.6 108.596 9 30.2 9 10.6 54.1 9 33.3 9 3.1 10.3 9 35.6 9 2.3 6.9 9 37.2 9 1.6 4.5 9 40.9 9 3.7 10.0 Sacramento0 Personal Incomeb Dollar Increase b Percentage Increase 1950-1959 Increase 9546.1 139.0% 9867.3 9321.2 58.9 9972.4 9105.1 12.1 91.042.5 9 70.1 7.2 91,143.0 9100.5 9.6 91,305.3 9162.3 14.2 *Reported in billions of dollars. bPrior year bass. cRsportsd in Billions of dollars. Soares: Dsrivsd fron data in: California Statistical Abstract. 1961; Federal Reserve Bulletin. 1955, 1957, 1&&; and California Chamber of Coaaerce. 89 personal income in Sacramento was approximately two and a half times that of 1950. From $546,137,000 in 1950, personal income in the area rose to $1,305,340,000 by 1959, for a 139 per cent increase over 1950. Total personal income in California also increased substan tially, from approximately $19.6 billion in 1950 to almost $41 billion in 1959, for an increase of 108.5 per cent. The United States, by comparison, showed a relatively small increase. Personal incoae in the nation, $227.1 billion in 1950, increased only 67.3 per cent to slightly over $380 billion by 1959. Between 1950 and 1955 total personal income rose 34.7 per cent in the nation, 54.1 per cent in the state, and 58.9 per cent in Sacramento. The United States showed an annual average of 6.9 per cent increase in personal income; whereas the data for California and Sacramento indicate annual growth rates in income during the 1950-1955 period of 10.8 and 11.8 per cent, respectively. Thus, Sacramento reported a relatively higher annual growth rate in personal incoae than either the state or the United States. Froa a 6.9 per cent increase in 1956 over 1955 personal incoae in the nation declined signifi cantly, as a result of the business recession in 1958, to record only a 3.0 per cent increase over 1957. The 90 state reflected the sane pattern, as personal incone dropped fron a 10.3 per cent gain in 1956 to a 4.5 per cent gain in 1958. Sacraaento also reflected a declining rate of growth in personal incone, fron a 12.1 per cent increase in 1956 to a 7.2 per cent increase in 1957. The growth in the nation's personal incone in 1959 had not regained its 1950-1955 average of 6.9 per cent annual personal incone; however, California's 1959 personal incone growth rate was nearly as high as its 1950-1955 average of 10.8 per cent average annual increase. Sacraaento's 1959 growth rate, in contrast, had exceeded its biggest year of 1956 and indicated an annual average growth rate of 14.2 per cent, which was nuch higher (20.3 per cent) than the 1950-1955 average of 11.8 per cent. In the nain, both California and Sacraaento have had a relatively greater growth in total personal incone than the United States as a whole. Wanes and salaries. Wages and salaries averaged 64.6 per cent of total personal incone in California for the ten years 1950-1959. In Sacraaento wages and salaries contributed a relatively higher 68.8 per cent of the total personal incone during the sane period. The data in Table 8 indicate the relative percentages 91 TABLE 8 WAGES AND SALARIES AS A PER CENT OF TOTAL PERSONAL INCOME, CALIFORNIA AND SACRAMENTO METROPOLITAN AREA, 1950 AND 1955-1959 Year California Sacramento 1950 61.596 67.596 1955 64.7 66.7 1956 65.5 68.6 1957 64.9 69.8 1958 64.3 69.7 1959 64.8 70.1 1950-1959 Average 64.496 68.896 Source: Derived fron data provided by the California State Chamber of Commerce. 92 of total personal incoae in California and Sacraaento that were attributable to wages and salaries during the 1950's. The data indicate that wages and salaries coaprised a relatively higher proportion of total per- sonal incoae in Sacraaento in each selected year of the 1950's than did wages and salaries in the state. Whereas the wages and salaries in the state ranged froa 61.5 to 65.5 per cent of total personal incoae, the range in Sacraaento during the saae 1950-1959 period was 66.7 to 70.1 per cent, which indicates the relatively greater iaportance of wages and salaries in the Sacraaento economy. Table 9 data indicate the average wage paid per worker for uneaployaent-insurance-covered industries in the nation, state, and Sacraaento. These figures exclude the effect of governaent payrolls and other noncovered industries. In each of the three years reported the average wage paid in Sacraaento was lower than the average wage paid in the state. In both 1953 and 1956 the average wage in Sacraaento approxiaated that paid in the United States aore so than the average wage paid in California. B{y 1959, however, due to the Inpact of relatively high paying defense industries, Sacraaento's average wage alaost equaled that of the state, which was $588 aore than the United States TABLE 9 AVERAGE WAGE PER WORKER, UNITED STATES, CALIFORNIA, AND THE SACRAMENTO METROPOLITAN AREA, 1953, 1956, AND 1959 Year United States California Sacraaento Data as a Per Cent of United States California Sacraaento 1953 $3,408 $3,792 $3,524 111.2 103.4 1956 3,844 4,316 3,948 112.3 102.7 1959 4,208 4,796 4,756 114.0 113.0 aBased upon uneaployaent-insurance-covered industries, i.e., excluding governaent. Note: Yearly averages are based upon first quarter figures. Source: United States Departaent of Coaaerce, County Business Patterns (Washington: Governaent Printing Office, 1953, 1956, 1959). to 01 94 average wage. Although the average wage per worker in Sacraaento who was employed in the nongovernment sector had increased by 1959, in almost all industries the average wage was lower in Sacraaento than in California. Table 10 records the average annual wage per worker by major industry classification in the nation, the state, and Sacramento. In 1953 the average annual wage per worker in all industries except finance, insurance, and real estate, and retail trade was higher in California than in Sacraaento County. By 1959, except for the change in average wages paid in manufacturing industry, all industries, including the two mentioned above, were below the state average. Although all manufactur ing indicates a higher average wage than that of the state, a breakdown of the manufacturing industry into subdivisions indicates that, of the thirteen selected manufacturing industries in Sacramento County, eight pay relatively lower wages than their counterparts pay generally throughout the state (Table 11). Only the printing, chemical, and fabricated metal industries, and those classified under the inclusive heading of "miscellaneous manufacturing" reported higher wages. The food processing industry, basic to the Sacramento TABLE 10 ESTIMATED AVERAGE ANNUAL WAGES PAID IN SELECTED INDUSTRIES, UNITED STATES, CALIFORNIA, AND THE SACRAMENTO METROPOLITAN AREA, 1953 AND 1959 1953 1959 United States California Sacraaento United States California Sacraaento Mining $4,160 $4,836 $4,280 $5,220 $6,300 $5,184 Contract Construction 3,836 4,236 3,928 4,520 5,404 5,068 Manufacturing 3,800 4,196 3,936 4,856 5,576 6,616 Public Utilities 3,644 4,036 3,684 4,840 5,436 5,036 Wholesale Trade 3,948 4,280 3,960 5,000 5,396 5,288 Retail Trade 2,508 3,100 3,148 2,948 3,612 3,560 Finance, Insurance, and Real Estate 3,220 3,480 3,888 4,204 5,606 4,508 Services 2,576 3,224 2,724 3,148 3,808 3,252 Source: United States Departaent of Coaaerce, County Business Patterns (Washington: Governaent Printing Office, 1953, 1956, I555JT 96 TABLE IX ESTIMATED ANNUAL AVERAGE WAGES PAID IN SELECTED MANUFACTURING INDUSTRY CLASSIFICATIONS, CALIFORNIA AND THE SACRAMENTO METROPOLITAN AREA, 1959 ■ I ■ H i . . — — — — — — ..... ............................ — — — — p— — I Industry Title______________ California Sacramento Food & Kindred Products 94,964 94,464 Apparel & Other Fabric Products 3,276 3,116 Luaber & Wood Products 4,696 4,372 Furniture & Fixtures 4,684 3,956 Printing, Publishing, & Allied Industries 5,560 5,724 Gheaicals & Allied Products 5,800 6,748 Leather & Allied Products 4,608 4,588 Stone, Clay, & Glass Products 5,152 4,984 Primary Metal Industries 5,900 4,780 Fabricated Metal Products 5,384 5,584 Machinery (except Electrical) 5,732 5,144 Transportation Equipment 6,240 6,766' Miscellaneous Manufacturing Industries 4,428 4,540 aCoaplled by the Sacraaento State College Roal Estate Research Bureau. Source: United States Departaent of Coaaerce, County Business Patterns (Washington: Governaent Printing Office, First Quarter, 1953, 1956, 1959). 97 economy, paid approximately 10 per cent lower salaries in the county than was generally paid throughout the state. The defense industries, employing approximately 60 per cent of the workers engaged in manufacturing industry in the county in 1959, paid the highest aver age wage. In 1959 the average wage, as indicated in Table 10, page 95, was $6,616. In that same year the defense industries paid wages averaging over $6,765 per worker. Except for defense as cited, and a few other industries as previously mentioned, manufacturing industry in Sacramento paid relatively lower wages than manufacturing industry in the state in general. The annual average wage paid by government was relatively consistent with the average wage paid in the private sector. The 1959 data in Table 12 indicate the annual average wages paid by federal, state, and local government in Sacramento. The annual average wage paid in the private sector in 1959 was $4,756 (Table 9, page 93). The average annual wage paid by the federal government during the same period to its employees in the Sacramento Metropolitan Area was approximately 10.0 per cent higher, or $5,196. State employees during 1959 received an annual wage also 10.0 per cent higher than that typical in Sacramento. Wages paid to local, city, and county employees were 98 TABLE 12 AVERAGE ANNUAL WAGES PAID TO GOVERNMENT EMPLOYEES IN THE SACRAMENTO METROPOLITAN AREA, 1959 Average Annual Waire As a Per Cent of Sacraaento Averace Federal $5,196 109.3 State 5,202 109.4 Local 4,824 101.4 Source: Derived froa data provided by the State of California, Division of Labor Statistics and Research, Departaent of Industrial Relations, Research and Statistics Section; California Depart aent of Eaployaent. 99 only 1.4 per cent higher than the Sacraaento average. Per Capita Incoae Per capita personal incoaes frequently are calculated to allow coaparisons between incoae in one area and that of another. It is iaportant to note, however, that the per capita incoae of an area is siaply the division of the total personal incoaes by the total nuaber of people in the area. That is, if every person in the area earns an incoae of 92,000, the per capita incoae will be 92,000. However, the per capita incoae can be 92,000 even though nany people in the area earn incoaes auch aore and auch less than 92,000. Consequently, even though the per capita incoae figures in both areas are the saae, the econoaic status of the people of the two areas is not the saae. Therefore, it becoaes obvious that per capita incoae does not necessarily represent the incoae of the "typical" individual. It is necessary to know the distribution of the total incoae aaong all the people in order to coapare with certainty the econoaic status of the people in the areas being coapared. Available data indicate that per capita incoae in the United States increased 44.5 per cent during 100 the 1950-1959 period. Per capita incone in Sacraaento rose relatively less, only 37.8 per cent in the saae period. The state of California, as reported in Table 13, had the highest percentage increase in per capita incoae as conpared with that of the United States and Sacraaento. The 1950-1959 per capita incoae increase was 46.4 per cent for the state. Between 1950 and 1955 per capita incoae in the United States and California grew at conparable rates. The nation's per capita incoae increased 25.7 per cent froa $1,495 to $1,879, and the per capita incoae in the state increased 27.7 per cent froa $1,869 to $2,386. Sacra- aento, in contrast, showed a relatively low 16.3 per cent increase in per capita incoae during the saae period. In 1950 per capita incoae in Sacraaento was 132 per cent of the nation's per capita incoae, and 106 per cent of that of California. However, Saera- aento's percentage declined substantially in relation ship to both the state and the nation; i.e., in 1959 per capita incoae in Sacraaento was 126 per cent of that of the nation's and only 99 per cent of the state's. TABLE 13 PER CAPITA INCOME, UNITED STATES, CALIFORNIA, AND THE SACRAMENTO METROPOLITAN AREA, 1950 AND 1955-1959 Year United States California Sacraaento Sacraaento Per Capita Iacoae as a Per Cent of California United States 1950 51,495 51,869 51,971 106 132 1955 1,879 2,386 2,294 96 122 1956 1,967 2,512 2,402 96 122 1957 2,031 2,573 2,429 94 120 1958 2,070 2,585 2,533 98 122 1959 2,164 2,736 2,716 99 126 1950-1955 Percentage Increase 25.7 27.7 16.3 1950-1959 Percentage Increase 44.5 46.4 37.8 Source: Unittd States Statistical Abstract and the California State Chaaber of Coaaerce. TOT 102 The Iapact of Aero. 1 et-General Corporation on the Sacraaento Metropolitan Area Econoaic Base Defense-oriented industry was cited earlier as being responsible for the increase in the average wage in the aanufacturing sector of the local econoay. Alaost all of the increase nay be related to the location of one new fira in the Sacraaento Metropolitan Area. This fira, Aerojet-General Corporation, which had purchased land in Sacraaento County as early as 1951, eaployed relatively few workers as late as 1955. During that year only 604 workers were on its local payroll. This basic industry eaployaent had been expanded to alaost 18,000 by year-end 1962. Data furnished by Aerojet-General indicate the iapact it has had on the econoaic base of the Sacraaento Metro politan Area. Estiaates provided by Aerojet-General report that, in 1958, 7,000 of their 13,500 eaployees were attracted froa areas outside of the county. The local labor pool supplied only 6,500 Aerojet-General workers in 1958. The iaaediate effect of the 13,500 workers was 20,000 new local residents in 7,000 households. It was estiaated that out of a $64 aillion annual payroll the workers and their faailies spent $30 103 million in local retail outlets, paid $2.5 million in property taxes, and created an additional $17.5 million in bank deposits at local institutions. Local expendi tures for material and supplies by the company in 1958 amounted to $19 million. Also, in the same year, $15 million was spent by Aerojet-General for local construction projects. Furthermore, it was estimated by the company that this expansion of basic industry employment led to an increase of 6,250 nonbasic 15 industry employment. During 1959 the local labor market provided an additional 1,500 workers, which brought the total number of Aerojet-General1s employees at the Sacramento facility to 15,000 with an annual payroll of more than $101 million. In 1960 an additional 1,000 workers were provided by each of the local and nonlocal labor markets. Thus, in 1960, there was a total of 17,000 workers, 9,000 of which had been hired in the local labor market and 8,000 of which were attracted from other economic areas. Also, in 1960 the annual payroll had risen to $118 million; local outlays were $11 15 Aerojet's estimate of increases in nonbasic industry employment must be considered as being a conservative estimate in view of the employment ratios developed in Tables 4 and 5, pages 83 and 85, respec tively. 104 Billion for material and supplies and $12.2 Billion for construction at the local facility. The accumu lated total investment in buildings and equipment had risen to $112 million. By the end of 1961 approximately 25,000 new local residents in 8,000 households, paying an esti mated annual property tax of $3 million and spending more than $30.6 million of the $132 million annual payroll for retail goods and services, were directly related with Aerojet-General basic industry. By mid year 1962 another 500 employees were hired by the company. This brought their total employment to '17,500. The local labor market provided 9,300 workers, and the nonlocal labor market provided 8,200 workers. The company estimates that $16 million was spent locally for material and supplies and $12.2 million for construction at the local plant. To date, Aerojet-General has paid more than $2.3 million in property taxes locally. In terms of nonbasic industry employment it has been estimated that more than 7,500 jobs have been created in other fields as a result of Aerojet-General's expansion in Sacramento. There fore, there can be no doubt that this firm has had a significant impact on the economic base of the Sacramento Metropolitan Area. 105 More specifically, each new job at Aerojet has (1) had a Multiplier effect of 1.5 jobs in the nonbasic sector of the local econosy, and (2) had an effect on the rate of new residential construction starts as evidenced by building peraits issued. Also, the rela tively high paying basic eaployaent at Aerojet has placed the eaployee in a better position to aake his deaand for housing an effective deaand. Table 14 was developed to report the iapact of Aerojet basic eaploy aent on the over-all increases in coaaunity eaployaent. Froa 509 eaployees in 1954, Aerojet expanded to 17,000 eaployees in 1960, at which level it reaained during 1961. Increases were reported each year until the leveling-off at 17,000 in 1961. The total annual additional eaployaent, based on the basic to nonbasic eaployaent ratio of 1:1.5 developed in Table 4, page 83, ranged froa none in 1961 to 16,373 in 1958. More than 25,000 nonbasic jobs have been created in the local econoay during the 1954-1961 period by the increases in the basic sector at Aerojet. The iapact of the favorable changes in the local econoaic base on the rate of residential construction starts nay be seen in Figure 9, which was developed by plotting the total annual additional eaployaent data and the total nuaber of housing units authorised by 106 TABLE 14 MULTIPLIER EFFECT OF ADDITIONAL AEROJET EMPLOYMENT, 1954-1961 Year Number of Aerojet Emnlovees Additional Annual Employment Created by Aerojet basic Cat Aerojet) Nonbasic Total Annual Additional Emnlovment 1954 509 — — — 1955 604 95 142 237 1956 4,637 4,033 6,050 10,083 1957 6,951 2,314 3,471 5,785 1958 13,500 6,549 9,824 16,373 1959 15,000 1,500 2,250 3,750 1960 17,000 2,000 3,500 5,500 1961 17,000 None None None aMultiplier effect ratio of 1:1.5 developed from Table 4. Source: Derived from original data provided by Aerojet-General Corporation. 107 (thousands) 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 \ Building \ Peraits 1954 1955 56 57 58 59 60 61 Figure 9* Total annual additional eaployaent and housing units authorised: Saeraaento Metropolitan Area, 1954-1961. 108 16 the building permits issued. The resultant figure ■ay be interpreted to show that a causal relationship exists between the increases and decreases of the total annual additional employment and the increases and decreases in the number of building permits issued. That is, Figure 9 indicates that a major shift (increases) in total annual additional employment levels during the 1955-1958 period led to a major shift (increases) in the number of building permits issued. Also, during the 1959-1961 period a major downward shift in the total annual additional employ ment levels led to a major downward shift in the number of building permits issued. Data reflecting the relatively high paying basic employment at Aerojet are presented in Table 15, which reports the trend of the average annual wage paid to Aerojet employees. Beginning in 1957, when $6,593 was the average annual wage, there has been an almost uninterrupted rise in the average wage bill each year until 1963. The estimated average annual wage for 1963 was $8,156, an increase of more than 16 Figure 9 is based on original data provided by Aerojet-General Corporation and Sacramento City- County Chamber of Commerce. 109 TABLE 15 AVERAGE EMPLOYMENT AND AVERAGE ANNUAL VAGE PAID BY AEROJET-GENERAL CORPORATION, SACRAMENTO, 1957-1962 Payroll Average Avoraga in Annual Annual Yaar_______ Mil Ilona___________oyant___________Wage 1957 $ 38.2 5,794 96,593 1958 64.0 10,225 6,259 1959 101.5 14,250 7,123 1960 118.0 16,000 7,375 1961 132.0 17,000 7,765 1962 146.5 18,000 8,139 1963 (eat.) 161.5 19,800 8,156 — — — - ' — — — — —— — — — ■ Sources Robert K. Coe, Sacraaento State College, baaed on Aerojet-General Corporation'a year end figurea. The average eaployaent ia conatrued aa the nean of the two reapective year-end figurea; for inatance, the mean of 17,000 and 19,000 ia 18,000. 110 $1,000 over the $7,123 paid in 1959. In contrast, in 1959 the average wage per worker in Sacramento was only $4,756 (Table 9, page 93), the typical employee in the manufacturing sector of the local economy was paid $6,616 (Table 10, page 95), and government employees in the Sacramento area received an average wage of $5,202 or less (Table 12, page 98). V. SUMMARY In general, this chapter has discussed the effects of changes in the economic base of various selected areas on nonfarm residential construction starts in those areas. It was shown that a degree of correlation may exist between changes in employment levels and changes in the number of nonfarm housing starts. Also, it was shown that no stable relationship exists, especially in the short-run, between personal incomes and nonfarm housing starts. Section IV of this chapter discussed Sacramento Metropolitan Area empirical data, changes in its economic base, and the influence which a new major employer has had on the economic base of the area. CHAPTER III THE INFLUENCE OF SELECTED DEMOGRAPHIC FACTORS ON NEW NONFARM RESIDENTIAL CONSTRUCTION It was stated early in Chapter II that the deaand for real estate is, in general, influenced by factors which may be classified into two broad cate gories: (1) economic factors, and (2) demographic factors. The economic factors were discussed in that chapter. The purpose of this chapter is to discuss the influence of selected demographic factors on new nonfarm residential construction. Demographic factors refer to all those ecological elements which affect the size of the household population.1 This chapter concerns itself with the relationship between popula tion distribution, growth, and migration, and new residential construction activity. The basic influence on the new housing sector of the over-all demand for housing is population growth. Furthermore, as will be shown in this chapter, it is the mobility aspect of population growth which is the key factor in the The household is the social unit occupying the housing unit and is the basic demand unit for housing. The basic supply unit in a housing market analysis is the housing unit. See Preston Martin. Real Estate Principles and Practices (New York: The Macmillan Company, 19^9), pp. . 112 demand for new housing. That is, population growth, principally through migration, is the underlying force in the residential construction market. Regional Distribution of Population and New Nonfarm Residential Construction In general, the distribution of new nonfarm residential construction in the United States on a regional basis during the 1950-1960 period was fairly equal; and, as the data in Table 16 suggest, a strong degree of correlation may exist between the percentage of United States population living in the region at year-end and the percentage of total nonfarm housing starts which took place in that region during the year. The Northeastern Region of the United States had approximately one-fourth of the United States population during the 1950-1960 period. There was, however, a slight decrease in its share of the United States population during almost every one of the eleven years. Even though the population of the region increased during the observed time period, the percentage of United States population living in that region declined froa 26.2 per cent in 1950 to 25.0 per cent in I960. Also, the percentage of total TABLE 16 PERCENTAGE ESTIMATES OF CIVILIAN POPULATION AND NONFARM HOUSING STARTS BY REGIONS OF THE UNITED STATES, 1950-1960 Year Northeast North Central South West Popu lation Starts Popu lation Starts Popu lation Starts Popu lation Starts 1950 26.2% 23% 29.5% 24% 31.0% 32% 13.3% 21% 1951 26.0 23 29.5 24 31.1 33 13.4 20 1952 25.9 22 29.4 23 31.0 33 13.7 22 1953 25.8 23 29.5 24 30.7 30 14.0 23 1954 25.9 20 29.6 27 30.3 29 14.2 24 1955 25.6 21 29.6 27 30.4 29 14.4 23 1956 25.3 20 29.5 27 30.5 30 14.7 23 1957 25.1 19 29.4 25 30.6 33 14.9 23 1958 25.1 18 29.2 24 30.6 34 15.1 24 1959 25.1 18 29.1 24 30.5 33 15.4 25 1960 25.0 19 29.0 23 30.5 33 15.5 25 Source: United States Department of Commerce, Bureau of the Census, United States Census of Population: I960. Volume I; United States Department of Commerce, Bureau ol-*the Census, Current Population Reports. Series P-25, No. 229, May 22, 1961; United States Department of Labor, Bureau of Labor Statistics, Construction Review. 114 housing starts, which occurred in the older North eastern Region of the United States, gradually declined froa 23 per cent of United States total in 1950 to 19 per cent in 1960. Regional housing starts in the latter year reversed the alnost continuous trend downward which had occurred during the prior ten years. The relationship of population distribution and housing starts in the North Central Region is less clear than that of the Northeastern Region. That is, during the 1950-1960 period the percentage of United States population living in the North Central Region fluctuated froa a high of 29.6 per cent in 1954 and 1955 to a low of 29.0 per cent in 1960. The secular trend is, however, downward. Housing starts in the North Central region also have fluctuated froa a high of 27 per cent of total housing starts in 1954, 1955, and 1956 to a low of 23 per cent of the total in 1952 and I960. In general, a siailar positive relationship of population distribution and housing starts in the North Central Region during the 1950-1960 period existed as in the Northeastern Region. The Southern Region, which contained the largest share of United States population, enjoyed the largest share of total United States housing 115 starts during the 1950-1960 period. However, the secular trend of population living in the Southern Region is, as in both the Northeastern and North Central Regions, downward. The range of percentage of total population is froa a high of 31.1 per cent in 1951 to a low of 30.3 per cent in 1954. The beginning and ending population percentage figures are 31.0 per cent in 1950 and 30.5 per cent in 1960. Southern housing starts as a percentage of total starts were fairly stable, being 30 per cent or aore of the total in all but two of the eleven years. During the 1950-1960 period the Western Region of the United States reported an upward trend, con trary to the other three regions, in both population and housing starts shares. The percentage of total United States population living in the Western Region increased .2 per cent or wore every year since 1950. In the aeantiae, housing starts in the region ranged, as a percentage of total United States starts, froa a low of 20 per cent in 1951 to a high of 25 per cent which occurred in both 1959 and 1960. Froa the above it aight safely be concluded that a degree of correla tion exists between population and housing starts in the Western Region. 116 In suaunary, the data in Table 16 are evidence that a strong degree of correlation exists between the year-end share of regional population and the share of total annual housing starts which occur in the region. Conoarison between the 1950 to 1960 Growth and Distribution of Population and Housing Units in Selected States Couplets historical population estimates are available for each of the selected states covered in this study. However, housing starts estimates for each of the states in the United States are available annually froa 1954 only; consequently the analysis in this section is based on housing units changes which have taken place in the housing stock between 1950 and 1960. This analysis, therefore, considers the net change in the total housing inventory of the various states. New construction (starts), denolitions, and conversions are thus taken into consideration. Data in Table 17 indicate that the population increase in the United States averaged 18.5 per cent during the 1950 to 1960 period. The increase in the United States housing stock was 27.0 per cent during this saae period. Nine of the fourteen selected 117 TABLE 17 POPULATION AND HOUSING UNITS GROWTH, THE UNITED STATES AND SELECTED STATES, 1950 TO 1960 Population Incraaaa Housing Units Increase United Stataa Average 18.5% 27.0% Florida 78.7 86.7 Nevada 78.2 79.0 Arizona 73.7 72.6 California 48.5 52.2 Maryland 32.3 35.7 New Jersey 25.5 33.1 Michigan 22.8 29.3 Ohio 22.1 26.6 Waahington 19.9 24.7 Illinois 15.7 22.6 Georgia 14.5 21.0 New York 13.2 22.9 Massachusetts 9.8 20.8 Pennsylvania 7.8 18.0 Source: Coaputed froa: Unltod Statos Dopartaont of Coaaarca, Buraau of tha Canaus, Currant Population Reports. Sariaa P-25, No. 227, "Prelininary Esilaates of tha Coaponanta of Population Change, by Stataa: 1950 to 1960," April 26, 1961, p. 4; Unitad Stataa Dapartaant of Coaaarca, Buraau of tha Canaua, Statiatical Abatract of tha Unitad Stataa, 1955, p. ?84 and1&62, p. 739. 118 states had above average population increases. Florida led the nation and the selected states with a 78.7 per cent population increase. Nevada and Arizona were a close second and third with population increases of 78.2 and 73.7 per cent, respectively. Washington, with a 19.9 per cent population increase, ranked ninth and was the last of the selected states which reported population increases above the national average. Thus, the above average population increases ranged from 19.9 to 78.7 per cent. Population growth in the remaining five selected states did not equal and hence were below the United States average popula tion increase. Among the selected states which were below the average United States population increase were the heavily populated states: Illinois (15.7 per cent), New York (13.2 per cent), Massachusetts (9.8 per cent), and Pennsylvania (7.8 per cent). These data become more revealing when compared with the housing units increase data reported for the various selected states. For example, each of the first seven selected states which reported above average population increases also had above average increases in housing units. The housing units increases in Florida and Nevada exceeded the popula tion increases of both states. That is, Florida's 119 housing units increase was 86.7 per cent; its popula tion increase was 78,7 per cent. Nevada reported 79.0 and 78.2 per cent increases in housing units and population, respectively. Arizona's percentage increases were almost identical in both population (73.7 per cent) and housing units (72.6 per cent). Also, the housing units increases in California (52.2 per cent), Maryland (35.7 per cent), New Jersey (33.1 per cent), and Michigan (29.3 per cent) all exceeded their respective population increases. That is to say, each had above average increases in both housing units and population. Only two, Ohio and Washington, of the fourteen selected states reported above average population increases without reporting above average increases in housing units. The housing units increases in these two states were 26.6 and 24.7 per cent, respec tively. The remaining five selected states did not report above average increases in either population or housing units. Thus, the conclusion that a strong degree of correlation exists between above average increases in population and above average increases in housing units is reaffirmed. Furthermore, the data in Tables 18 and 19 offer additional evidence that a good correlation TABLE 18 POPULATION ESTIMATES OF SELECTED STATES: 1960 AND 1950 April 1, 1960 (Census) April 1, 1950 (Census) April 1, 1950 to April 1, 1960 Increase Amount Per Cent United States 179,323,175 151,325,798 27,997,377 18.5 Florida 4,951,560 2,771,305 2,180,255 78.7 Nevada 285,278 160,083 125,195 78.2 Arizona 1,302,161 749,587 552,574 73.7 California 15,717,204 10,586,223 5,130,981 48.5 Maryland 3,100,689 2,343,001 757,688 32.3 New Jersey 6,066,782 4,835,329 1,231,453 25.5 Michigan 7,823,194 6,371,766 1,451,428 22.8 Ohio 9,706,397 7,946,627 1,759,770 22.1 Washington 2,853,214 2,378,963 474,251 19.9 Illinois 10,081,158 8,712,176 1,368,982 15.7 Georgia 3,943,116 3,444,578 498,538 14.5 New York 16,782,304 14,830,192 1,952,112 13.2 Massachusetts 5,148,578 4,690,514 458,064 9.8 Pennsylvania 11,319,366 10,498,012 821,354 7.8 Source: United States Department of Commerce, Bureau of the Census, Current Population Reports« Series P-25, No, 227, "Preliminary Estimates of the Components of Population Change, by States: 1950 to 1960," April 26, 1961, p. 4. 120 TABLE 19 HOUSING UNITS, UNITED STATES AND SELECTED STATES: 1950 AND 1960 1950 1950-1960 1960 Increase (In Thousands) 1950 1960 (Per Cent of United States) United States 45,938 58,326 12,388 100.00 100.00 New York® 4,634 5,696 1,062 10.09 9.77 California 3,591 5,466 1,875 7.82 9.37 Pennsylvania 3,037 3,582 545 6.61 6.14 Illinois8 Ohio8 2,672 3,276 604 5.82 5.62 2,403 3,041 638 5.23 5.21 Michigan 1,972 2,549 577 4.29 4.37 New Jersey 1,502 1,999 497 3.27 3.43 Massachusetts 1,400 1,691 291 3,05 2.90 Georgia 967 1,170 203 2.11 2.01 Florida 952 1,777 825 2.07 3.05 Washington 810 1,010 200 1.76 1.73 Maryland 689 935 246 1.49 1.60 Arizona 241 416 175 0.53 0.71 Nevada 57 102 45 0.12 0.18 aStates which suffered losses in their respective share of total United States housing units. Source: Conputed froa United States Department of Conaerce, Bureau of the Census. Statistical Abstract of the United States. 1955. n. 784 and 1962, p. 759. 121 122 exists between the distribution of population and the share of housing units in the various selected states. Table 18 indicates, by percentage increase rank, the population increase of the various selected states in terns of both absolute nunbers and percentage increase. New York, the most populated state in both 1950 and 1960, reported alnost two nillion additional residents during the period. This 13.2 per cent increase was, nevertheless, below the 18.5 per cent national average. Florida's slightly nore than two nillion additional residents resulted in a 78.7 per cent increase and the nunber one rank in terns of percentage increase growth. California, which reported the nost growth in terns of absolute nunbers, ranked fourth in per centage increase. These data becone nore neaningful when considered along with the changes in the selected states share of United States housing units. The conparative Table 19 was developed to show the increase or decrease in the various selected states share of the total United States housing units. Thus those seven states— California, Michigan, New Jersey, Florida, Maryland, Arizona, and Nevada— which increased their respective share of total United States housing units also ranked in the top half of the selected states which had above average population increases (Tables 123 17 and 18, pages 117 and 120, respectively). The bottom half of the selected states, ranked in terms of relative population Increases (Tables 17 and 18), had smaller shares of total United States housing in 1960 than they did in 1950. In summary, the above data indicate that in the various states: (1) above average increases in population were accompanied with above average increases in housing units; (2) the top seven states, in terms of relative population growth, increased their respective shares of total United States housing units; (3) the last seven states, in terms of relative population growth, decreased their respective shares of total United States housing units; and (4) the foregoing are consistent with the conclusion that a strong degree of correlation exists between the year- end share of regional population and the share of total annual housing starts which occur in the region. The Influence of Migration on Housing Unit Growth The purpose of this section is to discuss the mobility aspect of population growth and housing unit growth in selected states. Population growth does not necessarily lead to a proportionate growth 124 in housing units, as evidenced by the data in Tables 18 and 19, pages 120 and 121, respectively. Household composition and age distribution of the population are significant demographic factors in the demand for housing. Another important factor influencing the demand for housing, and hence the housing unit growth, is the net total migration to an area. The data in Tables 18 and 20 relate the impor tance of migration to population growth in the various selected states during the 1950-1960 period. For example, approximately one-half of the population growth in the states of Florida, Nevada, Arizona, California, Maryland, and New Jersey is attributable to net migration. That is, in six of the nine selected states which had greater than average population increases (Table 17, page 117) during the 1950 to 1960 period, approximately one-half of the population growth is attributable to net migration. Net migration in five of the selected states— New York, Illinois, Massachusetts, Pennsylvania, and Georgia— did not equal and hence were below the United States net total migration figure of 1.8 per cent. The data in Table 19, page 121, indicate that these five states suffered losses in their respective share of total United States housing units during the 1950 125 TABLE 20 NET MIGRATION, UNITED STATES AND SELECTED STATES: 1950 TO 1960 Number Per Cent United States +2,660,000 + 1.8 Florida +1,617,000 +58.3 Nevada + 86,000 +53.8 Arizona + 330,000 +44.0 California +3,145,000 +29.7 Maryland + 320,000 +13.7 New Jersey + 577,000 +11.9 Ohio + 409,000 + 5.1 Washington + 88,000 + 3.7 Michigan + 156,000 + 2.5 New York + 210,000 + 1.4 Illinois + 124,000 + 1.4 Massachusett s - 93,000 - 2.0 Pennsylvania - 475,000 - 4.5 Georgia 214,000 - 6.2 Source: United States Department of Commerce, Bureau of the Census, Current Population Reports. Series P-25, No. 227, "Preliminary Estimates of the Components of Population Change, by States: 1950 to 1960," April 26, 1961, p. 6. 126 to 1960 period. On the other hand, in each of the six states— Florida, Nevada, Arizona, California, Maryland, and New Jersey— where net Migration was a major factor in the population growth of that state, increases in their respective share of total United States housing units were reported during this sane period. Migration. Employment. and Housing Demand Additional evidence to support the thesis that there is a relationship between migration and employ ment is offered in Table 21. The selected states are ranked in this table above and below the United States rate (1*8 per cent) of population change attributable to migration during the 1950-1960 period. The result is that there are nine states above and only five states below the United States rate. The second column of figures reports the percentage increases in total nonfarm employment for selected states during the sane period. It is important to note that six of the nine states which were ranked above the United States migration figure reported much greater employ ment increases on a percentage basis than the United States percentage increase of 17.9 per cent in total 127 TABLE 21 SHARE OF POPULATION GROWTH ATTRIBUTABLE TO NET MIGRATION AND NONAGRICULTURAL EMPLOYMENT INCREASES, UNITED STATES AND SELECTED STATES, 1950-1960 Migration aa a Par Cant of Eaployaent Population Pareantaga &ange___________IffcrWi- Florida +58.3 89.3 Nevada +53.8 7.4 Arisona +44.0 108.1 California +29.7 53.2 Maryland +13.7 20.9 Now Jarsay +11.9 22.2 Ohio + 5.1 14.1 Washington + 3.7 21.5 Michigan + 2.5 10.7 Unitad Statas + 1.8 17.9 Naw York + 1.4 10.0 Illinois + 1.4 9.2 Massachusatts - 2.0 10.5 Pennsylvania - 4.5 5.0 Georgia - 6.2 32.3 Sourca: Unitad Stataa Dapartnant of Labor, Buraau of Labor Statistics, R»piaymmnt and Earnines. Annual Supplsasnt Issue. Voluae Vlll, No. 5, Noveaber, 1961; and Table 20. nonfara eaployaent during the years between 1950 and I960. Only one state, Georgia, which had lost popula tion as a result of Migration, reported a greater percentage increase in total nonfara eaployaent than the coaparable percentage increase in United States eaployaent. The underlying concept of the econoaic base theory is that an area which has a rapidly expanding econoaic base with high eaployaent rates would attract population to the area. "Under such conditions the supply of space would probably not be sufficient to o aeet the increased deaand." In other words, "a coaaunity's aggregate need for housing . . . will increase aore froa a Migration of 20,000 people con sisting of 6,150 households than froa a natural increase in population of 20,000." Increased deaand would lead to "bidding for the existing space Cwhich] 4 would then raise prices." Consequently, if the prices were to reaain high, "builders would be 2 Frederick E. Case, Real Estate (Boston: Allyn and Bacon, Incorporated, 106&), p. 103. 3 Martin, op. cit•, p. 38. 129 encouraged to build houses to be sold at these prices. Ibid. For an excellent illustrated discussion of the funciions of the market see pages 103-104, wherein the author states, "the influx of population in response to better jobs results in curve D2, with price rising to Pg. Continued influx of population results in a new demand indicated by Dg, with resulting price Pg. The higher prices with their accompanying higher profits will result in more properties being supplied and a new supply curve, S2. The new supply reduces the price pressures created by rising demand, so that prices drop to P^. As the supply demand rela tionships continue to change new prices will be pro duced, Pg, which is the point at which prices had previously come to rest.n $100 o •H £ m e * 3 a < 2 D 0 Units Supplied 100 PART C THE SUPPLY OF REAL ESTATE CREDIT CHAPTER IV THE INFLUENCE OF THE MONEY RATE AND MONEY AVAILABILITY ON HOUSING STARTS Changes in the voluae of new nonfara housing starts have been associated with changes in the ease of borrowing or the cost of borrowing. Certain policies of the federal governaent aay be singled out as evidence in support of this theory. Thus, in its effort to stiaulate the econoay, and eaploying residential construction as the priaary stiaulant, the federal governaent has, froa tine to tine, aggressively pursued a policy of liberalizing the FHA and VA mortgage prograas as well as expanding periodically the secondary narket activities of the Federal National Mortgage Association.^ The relationship between new nonfara housing starts and the cost or ease of borrowing is conplex. The conplexity of the interrelationship is increased by the familiar difficulty of aeasurenent. Quantify ing changes in the relative ease or stringency of borrowing aortgage funds is highly coaplieated. Also, it is extreaely difficult to aeasure the true *First National City Bank, Monthly Econoaic Letter, March, 1959, pp. 30-31. 132 cost of borrowed aoney. Furthermore, there is neither a national index of mortgage interest rates nor a general interest rate. With the above difficulties in mind this chapter attempts no more than to present data which explore the relationships between new nonfarm resi dential construction activity and the influence of the money rate and aoney availability, as indicated by the amount of "free reserves" available for lending and investing. In the absence of a national index of mortgage interest rates, the discussion in this part of the study is based on a general interest rate as represented by the Federal Reserve System's index of treasury bond prices. It is generally recognised that bond price fluctuations are representative of the fluctuations in loanable funds in the money markets. Another index of credit conditions, the banks' prime rate on business loans, was considered for this analysis. The banks' prime rate is the only interest rate charged by banks which is both widely publicized and uniform throughout the country.^ However, this index was not utilised because this o Albert M. Wojnilower and Richard E. Speagle, "The Prime Rate," Federal Reserve Bank of New York Monthly Review, April, 1§62, p. 54. 133 interest rate (1) is the banks? charges for their most credit-worthy business customers, and (2) is for short-tera business loans. Furthermore, as shown in * | Figure 10, the direction of the movements of the long-term treasury bond prices (yields, inverted) and the prime rate charged in the various cities closely parallel each other. Therefore, for the purpose of this section of the study, the bond price data were felt to be adequate. I. THE COST OF MONEY INFLUENCE ON NEW NONFARM RESIDENTIAL CONSTRUCTION Housing market economists agree that "tight" or "easy" money market conditions contain elements which influence housing costs. These important elements affect the rate of new construction and the supply of 4 housing; however, their influence was considered to 3 Source of data for Figure 10 is Federal Reserve Board of Governors. Federal Reserve Bulletins (Wash ington: Government Printing Office", ' 193511711). 4 The demand for housing is also affected by money market conditions. See Paul F. Wendt and Daniel R. Rathbun, "The Role of Governaent in the San Francisco Bay Area Mortgage Market," The Journal of Finance. December, 1951, p. 395, where they siatei" that "ease of financing is, of course, only one of the many factors influencing housing demand. The level of incomes, the prices of other items entering into the Bank Rates 1950 '51 *52 '53 '54 *55 '56 '57 '58 *59 *60 Figure 10. Bond prices and bank rates short-tern business loans. X35 5 be minor during the 1889-1952 period. In the years between 1950 and 1959 the high level of econoaic activity assisted in maintaining a high level of demand for new residential construction, and the home- building industry considered tight aoney periods as being the major restrictive element to an ever- expanding rate of new residential construction. The available data appear to indicate that in the 1950- 1959 decade the rate of housing starts generally moved in close relationship to activity in the money market. That is, periods of tight aoney were asso ciated with decreased housing-starts rates. Con versely, periods of easy money were associated with g accelerated housing-starts rates. This relationship 4 (continued) cost of living, the prices and avaiJability of rental housing, and the investment outlook are among the other important variables influencing the volume of construction and the prices in the residential dwelling market." ®Leo Grebler, David M. Blank, and Louis Winnick, Capital Formation in Residential Real Estate (Prince- ton, New Jersey: PrTnceton University PressT 1956), p. 224. g See Federal Reserve Board of Governors, Forty-Fifth Annual Report (Washington: Government Printing Office, 1958), p. 5, which states that "the Increased availability of mortgage funds at lower cost, together with the maintenance of personal income, was promptly reflected in a step-up of builder activity in constructing new houses." 136 ■ay be shown by (1) using the plotted Federal Reserve average price of United States long-tera governaent bonds to represent the aoney aarket trend, and (2) plotting the seasonally adjusted annual rate of housing starts with a six aonth lag to reflect the reaction of the residential construction industry to bond price 7 changes. The tiae series lines of Figure 11 indicate that a parallel downward aoveaent in both housing a starts and bond prices was evident in 1948. The 7 Reactions to changes in the aoney aarket are not iaaediately reflected in the housing starts data. This is true because of the use of either advance coa- aitaents or stand-by coaaitaents as well as the tiae necessary to prepare plans, to obtain building peraits, and to secure the aforeaentioned financing. The selec tion of six aonths as the lag period is based upon nuaerous references to that effect. See David L. Wlckens, Residential Real Estate: Its Econoaic Position as Shown by Values. Ranis. Faaily Incoaes. Financing. and Construction. Together wlih Eatiaatea for Ail Real Estate (New York: National bureau of Econoaic Research, 1941), p. 46; Geoffrey H. Moore (editor), Business Cycle Indicators (Voluae I; Princeton: Princeton Uni versity Press, 1961), pp. 56, 678; Doris G. Phillips, "Econoaic Dualisa in Classical Econoaic Thought" (paper read at the Western Econoaic Association Annual Meeting, San Francisco State College, San Francisco, California, August 22, 1963); George Parker, "The Savings and Loan Business," United States Savings and Loan League, Procssdingg of the Conference on Sayings and Residential Financing. lgH (Chicago: TEe League, 1958-1462), p. 15§; Nathaniel fi. Rogg, "Another Look at Soae Factors in Deteraining Housing Voluae," United States Savings and Loan League, 1960 Proceedings. p. 35. A Sources for Figures 11 to 16, inclusive, are Federal Reserve Board of Governors, Federal Reserve Bulletins. 1950-1961. Nonfara Housing 1948 '49 '50 *51 '52 '53 »54 '55 '56 '57 '58 '59 '60 '61 Figure 11. United States governaent bond prices and housing starts, 1948-1961. 137 138 money market switched to a condition of easy money during the 1948-1949 recession; bond prices edged upward in 1949, and housing starts surpassed the record 1925 housing starts for the first time to build up to a record number of 1,352,000 (old series) nonfarm housing starts in 1950. During 1950 bond prices turned downward again; and with the tightening of the money market after "The Accord" in March 1951, when the Treasury and the Federal Reserve agreed to the unpegging of govern ment security prices, bond prices drifted lower for three years into 1953. After the record number of nonfarm housing units were built in 1950 housing starts drifted downward until 1952, when they moved upward for approximately twelve months. Then both bond prices and housing starts paralleled each other downward in 1953. When bond prices rose in 1954, under the influence of an inventory recession, housing starts accompanied the rise to reach a peak in January 1955. For a second time in the post-World War II period both bond prices and housing starts swung upward together during a recession and starts were carried to new peaks before swinging downward again. Early in January 1955 the price of quality governaent bonds declined; simultaneously the rate 139 of housing starts drifted downward. These parallel aoveaents continued until the 1958-1959 recession. Bond prices reached a new low (new series) in the winter of 1957, as did housing starts five aonths later in March, 1958. At those points in tiae both variables started upward again and aaintained the saae approxiaate tiae lag. The peak in bond prices, however, was reached in six aonths before heading downward again in 1958, whereas housing starts peaked out in nine aonths at the end of 1958. Again, for the third tiae during a recessionary period, both bond prices and housing starts had swung upward together. Consequently, it seeaed only natural for soae indi viduals, both in private industry and in high govern aent positions, to look toward the liberalizing of terns in the governaent-insured sector of the housing aarket to lead the econoay out of the 1960-1961 recession. During the aost recent recession the apparent close relationship of bond prices and housing starts, however, did not continue as it had in the three prior recessions. Bond prices aoved slightly upward in nid-year 1960, dipped, then leveled off late in the year before aoving upward a few points at year end. The 1959 downward shift in the rate of housing 140 starts continued until the low point was reached in December 1960. The 1960 bond price high was followed by the housing starts high in October 1961, fourteen ■onths later. Therefore, the housing starts rate shifted away froa the apparent close relationship which had existed during the earlier years of the period under review. One possible inference which Bight be drawn from Figure 11, page 137, and Table 22 is that there Bight be a direct causal relationship between bond price increases and decreases and housing starts activity. The data in Table 22 have been arranged to show the bond prices and housing starts on the peak and trough dates, as established by the National Bureau of Econoaic Research, as well as on the date aidpoint between the peak and trough. The housing starts date and data have been lagged six aonths to allow tiae for the reaction to changes in bond prices to be reflected. The data indicate (1) that the bond price average percentage increase during the four recessions was 6.8 per cent, and (2) that the average percentage increase in the housing starts seasonally-adjusted annual rate was 28.2 per cent. Therefore, the inference aight be Bade that for each 1.0 per cent increase in bond prices during a TABLE 22 PERCENTAGE INCREASES IN BOND PRICES AND HOUSING STARTS, SELECTED DATES Bonds Starts Percent Seasonally Percent- Business Qyde age Date (six- Adjusted age Item Reference Dates Price Increase month lac) Annual Rate Increase 1 Nov. 1948 P $100.79 May 1949 911 2 April 1949 101.65 Oct. 1949 1,149 3 Oct. 1949 T 103.90 April 1950 1,382 Peak-Trough Increase 3.11 3.1 51.7 4 July 1953 P 100.03 Jan. 1954 1,051 5 Jan. 1954 106.16 July 1954 1,220 6 Aug. 1954 T 111.50 Feb. 1955 1,324 Peak-Trough Increase 11.47 11.4 25.9 7 July 1957 P 91.31 Jan. 1958 1,020 8 Nov. 1957 92.87 May 1958 1,039 9 April 1958 T 98.23 Oct. 1958 1,303 Peak-Trough Increase 6.92 7.6 27.7 10 May 1960 P 84.39 Nov. 1960 1,206 11 Sept. 1960 88.57 March 1961 1,262 12 Feb. 1961 T 88.74 Aug. 1961 1,301 Peak-Trough Increase 4.35 5.1 7.8 Key: P«Peak, T-Trough. Source: Computed froa data provided by Federal Reserve Balletins; and, United States Department of Coaaerce, Bureau of the Census, Business Cycle Developments. December, 1962. 142 recession, an increase of 4.1 per cent is to be expected in nonfarn housing starts. However, addi tional statistical analysis fails to support this precise relationship. For analytical purposes Table 22 data were plotted in Figure 12. Housing starts were plotted on the x-axis and bond prices on the y-axis as item one to twelve (Table 22), inclusive. The resultant scatter diagram of the data for the four recession periods indicates that a cause-and-effect relationship night exist between bond price increases during a recession and housing starts. This hypothesis was not validated, however, when subjected to the Pearson- ian r coefficient of correlation formula. The coefficient of correlation of the four-recession- period data is .15. Therefore, where unity (1.0) indicates perfect correlation and zero (0.0) indicates the absence of any correlation, .15 indicates that the apparent parallelisms lack any significant degree of correlation and could have happened simply by 9 chance. Therefore it appears that, for the period under review, the influence of the cost of money on 9 For complete calculations see Appendix Note, Table 23 at the end of this chapter. 143 115 110 105 ' 100 11 12 90 10 85 950 1,050 1,150 1,250 1,350 1,450 900 1,000 1,100 1,200 1,300 1,400 1,500 Figure 12. Bond price-housing starts scatter diagram. 144 the number of housing starts, on a nationwide basis, has not been a factor which can be shown to be statistically significant by the usual tests. However, the statistical record does support the residential construction industry opinion, contrary to that which economic theory has thus far denied, that it is the availability of mortgage money which leads to an increase in housing starts.10 II. THE RELATIONSHIP BETWEEN MONEY AVAILABILITY AND NEW NONFARM RESIDENTIAL CONSTRUCTION One measure of money availability, more commonly referred to as the availability of credit, is the net reserve position of commercial banks. The commercial banking system's nationwide supplies of money imme diately available for lending and investing are known as "free reserves." Commercial banks are required to set aside as reserves certain percentages of their Substituting the "free reserves" data for the bond price data in the Pearsonian r coefficient of correlation formula results in a coefficient of corre lation of .47. By comparison, therefore, there appears to be a higher degree of correlation between free reserves and housing starts than bend prices and hous ing starts. These results suggested checking the degree of correlation between bond prices and free reserves. The coefficient of correlation between these two variables was found to be .68. 145 deposits. These reserves nay be held either as deposits at Federal Reserve banks or as vault cash. An expansion in loans and investments increases deposits which, in turn, will require a larger amount of reserves. The capacity of a bank to extend credit is influenced by its reserve position. At any point in time it is probable that some banks may have reserves that exceed their requirements, while other banks nay have to borrow from the Federal Reserve banks. The margin by which aggregate excess reserves exceed total borrowings constitutes free reserves, or immediately available money for lending or investing. Free reserves indicate that member banks, as a whole, have excess reserves to support additional credit expansion. Free reserves reflect a relatively "easy" reserve position. The higher the level of free reserves, the greater the capacity of the banking system to extend credit. Figure 13 traces the net reserve position of member banks for the period 1950 through I960. This includes the period since the Treasury-Federal Reserve accord of March 1951 which made possible a return to a flexible monetary policy. Between April 1951 and the end of 1960 member banks reported a net deficiency of reserves in fifty-seven months and a net excess of reserves in sixty months. Million* of Dollars 1,000 800 600 400 200 0 -200 -400 -600 -800 -1,000 Thousands 1,700 1,600 1,500 1,400 1,300 1,200 1,100 1,000 Free R Housing Starts i i t i t i i ■ i t i 50 51 52 53 54 55 56 57 58 59 60 61 ,700 ,600 ,500 ,400 ,300 ,200 ,100 ,000 Figure 13. Menber banks' net reserves and housing starts, 1950-1960. o> 147 The test of that portion of the underlying hypothesis which relates the availability of funds for real estate credit to increased rates of resi dential construction aay be nade by plotting the seasonally adjusted voluae of new private peraanent nonfara housing units started on a chart with the nenber bank's net reserve data. The housing units data are plotted with a six aonth lag in Figure 13, which also traces the net reserve position of aeaber banks— the key institutions in the aoney aarket. An analysis of the tiae series lines indicated that a high degree of correlation aay exist between aoney availability and increased rates of residential construction since the return to a flexible aonetary policy in 1951. More specifically, aajor shifts (increases) in free reserves during the years since the accord were accoapanied by aajor shifts (increases) in the nuaber of housing units started. Also, during this sane period, aajor downward shifts in free reserves led aajor downward shifts in the nuaber of housing units started. Therefore, it nay be concluded that during the ten years, 1952-1961, a positive relationship existed between increases and decreases in the availability of aoney (free reserves) and increases and decreases in the voluae of housing units started at the national level. 148 111. COMPARISON OF BOND PRICE CHANGES AND CHANGES IN RESIDENTIAL CONSTRUCTION IN FOUR GEOGRAPHIC REGIONS This section of the study focuses on the housing starts activity in the four geographic regions of the United States rather than the nationwide basis, as discussed earlier. The section following this describes the housing starts activity on an even snaller geographic area, i.e., the standard Metro politan areas. Thirteen Metropolitan areas' housing starts activities are discussed in the concluding section of this chapter. Regional and local data, unfortunately, are not as plentiful as national data; consequently the period under review in the following two sections is reduced to the nuwber of years between 1953 and 1962. Another linitation at the regional and local levels is the lack of seasonally adjusted data which eliMinate seasonal construction patterns due to local clinatic conditions. In the absence of seasonally adjusted data, Monthly data have been converted to thirteen Month Moving totals and plotted with a six Month lag to allow tine for reaction to changes in the noney Market. The use of thirteen 149 ■onth aoving totals, or thirteen month ending figures as they are more properly called, eliminates much of the seasonal element which exists in all new construc tion sectors and reduces the influence of nonrecurring activity, thereby enabling the true trend to become discernible. Figure 14 illustrates the swings in both bond prices and regional nonfarm housing starts for the period under review. It is interesting to note that in six out of seven years a close relationship was apparent in the time series lines of each of the plotted variables. This was equally true in two out of the three recession periods which occurred during the observed tiae period. It was not until the seventh year, 1961, and the third recessionary period that regional housing starts did not swing in the same direction as did bond prices. The regional housing starts tiae series lines followed those of bond prices upward in 1953 and 1954 and downward in 1955. How ever, at this point in tiae divergencies began to appear in the regional housing starts patterns. It appears that, starting in 1956, the two deficit capital regions with traditionally high inter est rates reacted more spontaneously to changes in the money market. That is, the housing starts rates in 150 Bond Prices u£lng Starts Nonfara He Northeast North Central South West 1954 '55 '56 '57 '58 '59 *60 '61 Figure 14. Bond prices and regional housing starts, 1954-1961. 151 the Southern Region and Western Region of the United States began a three year cliab to their respective peaks in 1959. Meanwhile the older geographic region, which has aaple capital and traditionally low rates of interest, deaonstrated less responsiveness to bond price changes in the aoney aarket. Thus the North eastern Region of the United States, in which the aoney aarket capital of the world, New York City, is located, aaintained a close relationship between bond prices and housing starts until aid-year 1960. Housing starts in the Northeastern Region failed to cliab upward during the third recessionary period under review and continued the downward trend which had started in the latter part of 1958. However, this dip in housing starts was aore gradual than that of the other three regions. This was particularly true when the sluggish reaction of the housing starts tiae series line of the Northeastern Region is coapared with that of the Southern Region, which displayed greater aaplitude of fluctuations. The upward shift in the rate of housing starts in the South had started early in 1956, approxiuately ten aonths prior to that of the Northeastern, North Central, and Western Regions, and cliabed even higher than the other three regions. Subsequently, the sharpest decline in 153 period and converted to thirteen month ending figures. These data have bden plotted, by regions, for com parison with changes in bond prices as was done at the national and regional levels. It is clear from the figures drawn to illustrate these data that fluctua tions of the greatest magnitude occurred in the most heavily populated metropolitan areas, and if trend lines were plotted through these data a close downward relationship would appear. The data plotted in Figure 15 indicate that the New York, Chicago, and Detroit Metropolitan Areas housing starts fluctuatedmore than the Boston, Buffalo, and Cleveland Metropolitan Areas. During the 1953-1954 recession all of these metropolitan areas responded to the shift to an easy money policy to record increases in housing starts. However, four of the six cited areas reached their housing starts peak prior to the bond price peak. Only in two areas, Chicago and Cleveland, did housing starts peak out at the same approximate time that bond prices reached their high point during that recession. Housing starts in the above mentioned metro politan areas appear to have followed bond prices upward during the 1957-1958 recession. Actually, the shift in the rate of housing starts in the New York, 152 regional housing starts was recorded in the South. It is interesting to note that the narrowest anplitude of fluctuations appeared in the stable Northeastern Region, where noney is plentiful in relation to the denand, and the capital deficit Western Region of the United States, where the denand for new residential construction has been second only to the South. Thus it can be concluded fron the foregoing that regional differences do exist, even after seasonal elenents have been withdrawn fron the data, and that the differences cannot be explained entirely on the basis of the cost of noney. IV. COMPARISON OF BOND PRICE CHANGES AND CHANGES IN RESIDENTIAL CONSTRUCTION IN THIRTEEN METROPOLITAN AREAS Data available at the local level, as at the regional level, are not as conplete as data at the national level. Hence, only a United analysis can be nade of the reaction of housing starts in selected uetropolitan areas to changes in bond prices. Monthly data in twelve netropolitan areas have been converted to thirteen nonth ending figures for the 1954-1961 period. Also, enpirical data for the Sacranento Metropolitan Area have been developed for the 1950-1961 154 Bond Prices Nonfara HonbIqb Starts New York Buffalo Boston Chicago 1954 '55 '56 '57 ’58 • i i Figure 15. Bond prices and housing starts in Boston, Buffalo, Chicago, Cleveland, Detroit, and New York, 1954-1961. 155 Boston, and Chicago areas had started upward prior to the increase in bond prices late in 1957, and continued to clinb for several Months after bond prices had reached their peak early in 1958. Detroit and Cleve land housing starts appear to have been more responsive to the 1957 bond price increase since housing starts in those areas started upward late in 1957, and started to drop after bond prices did in 1958. Buffalo, on the other hand, showed little reaction to changes in the noney Market, and the trend in housing starts there continued the decline which had started in 1955. The pattern of housing starts in the above six Metropolitan areas is not clear for the 1960-1961 recession. Housing starts in New York, Chicago, and Boston, to a lesser degree, appear to have started upward with bond prices prior to the May 1960 National Bureau of Econoaic Research business cycle peak. New York and Boston housing starts continued to cliub at the trough date of February 1961; whereas housing starts in Chicago had started to drop several Months earlier. In the neantiae, housing starts in Buffalo, Cleveland, and Detroit continued to drift downward during the period since the previous recession. Data for seven Southern and Western Metropolitan Areas have been plotted in Figure 16, which records 156 Bond Pricey Nonfarm HouailSg' Star Sacrc ne isco- San Los as 1951 *52 '53 '54 '55 '56 '57 '58 '59 '60 '61 Figure 16. Bond prices and housing starts in Atlanta, Baltimore, Los Angeles, Sacraaento, San Diego, San Francisco-Oakland, and Seattle, 1950-1961. 157 the fluctuations in bond prices and the housing starts (thirteen month ending figures) for Atlanta, Baltimore, Los Angeles, San Diego, San Francisco-Oakland, and Seattle for the 1953-1961 period. Also, the 1950-1961 empirical data for Sacramento are therein reported. This graphic analysis illustrates that the tradition ally high interest areas responded as a whole in a positive manner to increases in bond prices during the first two of the three recessions in the period under review. The most heavily populated areas, Los Angeles and Baltimore, to a lesser degree, indicated the closest relationship between increased availability of money and housing starts not only in all three recessionary periods but also during the expansion periods between the business cycle contractions. In the Atlanta Metropolitan Area housing starts appear to have followed bond price changes during the first half of the 1954-1961 period and leveled off since the 1957-1958 recession, contrary to the downward shift of bond prices, fiasy money during the 1960-1961 recession was followed by upswings in housing starts in four out of the seven areas. Of the three areas which did not reflect increases in housing starts during the 1960-1961 recession two, namely San Diego and Sacramento, had suffered tremendously in 1960 158 fron an overhang of unsold houses, many of which had been built during the peak 1958-1959 period and had never been sold or occupied at any tine. Consequently, lenders in those two fast growing areas were extremely cautious in screening new residential construction loan applications. Fron the foregoing it night be concluded that there is a honogeneity of the new residential con struction industry in the metropolitan areas within the geographic regions of the United States. That is, individual metropolitan areas generally conform in their regional housing starts patterns. However, the magnitude of fluctuations and the time lag between bond price changes and changes in the housing starts rate are nonhomogeneous between metropolitan areas, regions, and nationwide. Thus, the local nature of the housing market is reaffirmed. The purpose of this chapter was to present data which explore the relationships between (1) the cost of money, as represented by the Federal Reserve System's index of treasury bond prices; (2) money availability, as indicated by the amount of "free reserves" available for lending and investing; and (3) new nonfarm residential construction starts rates. Analyses of the relationship between the bond prices 159 and housing starts variables led to the conclusion that the influence of the cost of money on the number of housing starts, during the study period, has not been a factor which can be shown to be statistically significant by the usual tests. Subsequent analysis of other pertinent data indicated that a higher degree of correlation may exist between free reserves and increased rates of housing starts than the correlation between bond prices and housing starts rates. V. APPENDIX NOTE As indicated earlier, correlation analysis is one of the statistical tools available for the measure sent of the degree of relationship between two variables. This technique was utilized to evaluate the degree of causal relationship between bond prices and housing starts (seasonally adjusted annual rates, six months lag) on selected dates. The dates were selected on the basis of the National Bureau of Economic Research designated business cycle turning dates. Table 23 reports the statistical analysis of the data found in Table 22, page 141. Other tests of the relationship between bond prices and housing starts were made, and the results of these analyses are reported in Table 24. Varying tine periods were TABLE 23 CORRELATION OF HOUSING STARTS AND BOND PRICE DATA, SELECTED DATES X Y X2 Y2 XY 1 911 100.79 829,931 10,158.6 91,819.7 2 1,149 101.65 1,320,201 10,332.7 116,795.9 3 1,382 103.90 1,909,924 10,795.2 143,589.8 4 1,051 100.03 1,104,601 10,006.0 105,131.5 5 1,220 106.16 1,488,400 11,269.9 129,515.2 6 1,324 111.50 1,752,976 12,432.3 147,626.0 7 1,020 91.31 1,040,400 8,337.5 93,136.2 8 1,039 92.87 1,079,521 8,624.8 96,491.9 9 1,303 98.23 1,697,809 9,649.1 127,993.7 10 1,206 84.39 1,454,436 7,121.7 101,774.3 11 1,262 88.57 1,592,644 7,844.6 111,775.3 12 1,301 88.74 1,692,601 7,874.8 115,450.7 Total 14,168 1,168.14 16,963,434 114,447.2 1,381,100.2 Standard deviations: x2 - X2 - t y 2 - 16,963,434 - 200,732,224 _ 16|963 |434 _ 16,727,685 - 235,749 2 y2 - Y2 - * 114,447 - 1>3^i55^ . 114,447 - 113,712 - 735 (Continued) TABLE 23 (CONTINUED) < r x - /g? - . /14.W4 . 1 4 0 m j " j * /5T72 *7.8 Cross products: xy - XI - ( Y) - 1,381,100 - 1 6 - 1,381,100 - 1,379,184 - 1,916 Coefficient of correlation: xt 1.916 1.916 " “ -15 161 162 TABLE 24 CORRELATION COEFFICIENTS OF BOND PRICES AND HOUSING STARTS: UNITED STATES, REGIONS OF THE UNITED STATES, SELECTED STATES, METROPOLITAN AREAS, AND DATES -.17 Stltcttd Mtropolitan Ar»a» 1954 Regions 1551-1961 Northeast North Central South West Massachusetts New Jersey New York Pennsylvania Illinois Michigan Ohio Georgia Galifornia Washington .15 .26 -.11 .21 .15 .38 .14 .33 .29 .67 .21 >•18 .08 .16 Boston Buffalo New York- New Jersey Chicago Cleveland Detroit Atlanta Baltimore Los Angeles San Diego San Francisco Sacranento 1950-1961 .01 .62 .01 .31 .28 • 72 • 23 .53 .28 .24 .08 .35 •■ployed on the basla of availability of data for the specified area. For exaaple, the analysis for the United States as a whole was for the years 1948 through 1961; the tiae period for the regions, the selected states, and the Metropolitan areas was 1954 through 1961; and the tine period for the Sacraaento Metropolitan Area was 1950 through 1961. CHAPTER V MAJOR SOURCES OF RESIDENTIAL REAL ESTATE CREDIT I. INTRODUCTION The hypothesis underlying this study is that increased rates of nonfara residential construction are influenced by (1) favorable changes in the economic base of an area, (2) availability of funds for real estate credit at savings and loan associations, and (3) the Federal Reserve System's Monetary policies as they affect the availability of funds of nonbank financial intermediaries. The first part of the underlying hypothesis has been dealt with in Part B, The Demand for Residential Construction Mortgage Credit. It is the purpose of this chapter, therefore, to analyse the role of savings and loan associations and other major financial institutions which provide residential mortgage credit. This is accomplished by a brief analysis of the structure and growth of the savings and loan industry and major assets and lia bilities of the associations, at both the national and the state (California) levels. Following this is a comparative analysis of the mortgage lending activi ties of the important financial institutions in the United States and California. The significant 165 competitive shifts which have occurred in the relative importance of these institutions as suppliers of residential mortgage credit during the 1950-1960 period are noted. 11. THE SAVINGS AND LOAN INDUSTRY IN THE UNITED STATES Structure and Growth The number of savings and loan associations in operation in the United States has grown steadily during the 1950-1960 period. As the data in Table 25 indicate, the moderate annual growth amounted to a net gain of 328 which represents a 5.5 per cent increase in the number of associations during the period. Federally chartered associations increased from 1,526 in 1950 to 1,873 in 1960, a 22.7 per cent increase. The total number of state chartered asso ciations actually decreased by nineteen during this same period; however, those state chartered associa tions which were insured by the Federal Savings and Loan Insurance Corporation (FSLIC) grew 66.7 per cent in number, or from 1,334 in 1950 to 2,225 in I960. Noninsured state chartered associations decreased steadily in number from 3,132 in 1950 to 2,222 in 1960, a decrease of 29.1 per cent. This does not TABLE 25 THE STRUCTURE AND GROWTH OF THE UNITED STATES SAVINGS AND LOAN INDUSTRY: NUMBER OF ASSOCIATIONS BY TYPE OF CHARTER, SELECTED DATES Year Insured Total Insured Noninsured Grand Total State Chartered Federally Chartered State Chartered 1950 1,334 1,526 2,860 3,132 5,992 1955 1,871 1,683 3,554 2,517 6,071 1956 1,927 1,739 3,666 2,470 6,136 1957 2,000 1,772 3,772 2,397 6,169 1958 2,074 1,807 3,881 2,327 6,207 1959 2,138 1,841 3,979 2,245 6,223 1960 2,225 1,873 4,098 2,222 6,320 Source: Insured associations data are fron the Federal Hone Loan Bank Board Annual Combined Financial Statements, 1950-1960; noninsured associations data are fron the United States Savings and Loan League 1961 Fact Book. 167 necessarily aean that all of the 910 associations involved in the decrease ceased to exist. Many, if not aost, of thea undoubtedly becaae insured associa tions upon qualifying for insurance of accounts by the FSLIC and therefore were included in the insured association category the following year. The growth in the volune of business done by savings and loan associations, in teras of asset growth, has been even aore pronounced them the gain in actual nuaber of associations. Total assets grew aore than $54 billion, which represents a 323.3 per cent increase, during the 1950-1960 period (Table 26). Federally chartered associations recorded the greatest absolute growth in assets during this period. The assets of these associations increased aore than $30 billion in the ten year period. Insured state chartered association assets grew by aore than $24 billion, which represents a phenoaenal 456.9 per cent increase in assets. Therefore, in teras of relative growth, the insured state chartered associations have grown the aost in both nuaber of associations and asset volune. It is inportant to note, however, that the federally chartered associations' total assets increased fron less than $8.5 billion, which was approxiaately 50.0 per cent of the nation's savings TABLE 26 THE STRUCTURE AND GROWTH OF THE UNITED STATES SAVINGS AND LOAN INDUSTRY: ASSETS OF ASSOCIATIONS BY TYPE OF CHARTER, SELECTED DATES (IN MILLIONS OF DOLLARS) Year (December 31) State Insured Chartered Noninsured Total Federally Chartered Grand Total 1950 $ 5,192 $ 3,202 $ 8,394 $ 8,452 S 16,846 1955 14,039 3,458 17,497 20,036 37,533 1956 16,381 3,443 19,824 22,957 42,781 1957 18,729 3,594 22,323 25,730 48,053 1958 21,675 3,667 25,342 29,636 54,978 1959 25,193 3,851 29,173 34,357 63,401 1960 28,919 3,884 32,803 38,511 71,314 1950-1960 Dollar Increase 23,727 672 24,409 30,059 54,468 1950-1960 Percentage Increase 456.9 20.9 290.8 355.6 323.3 Source: Insured associations data are froa the Federal Home Lean Bank Board Annual Combined Financial Statements, 1950-1960; noninsurad associations data aro froa the United States Savings and Loan League 1961 Fact Book. 168 169 and loan aggregate at year-end 1950, to $38.5 billion or 54.0 per cent of the total asseta in 1960. Further more, as shown in Table 27, federally chartered associations held higher average asset amounts than did the other two types of associations. Federally chartered associations held average asset balances of $20.6 million at year-end 1960, as compared with both the $13.0 million average assets reported by the insured state chartered associations and the low $1.8 million average assets of the noninsured state chartered associations. Ma.ior Assets and Liabilities The principal assets of a savings and loan association are its mortgage loans which, except for a minor portion of approximately 7.0 per cent, repre sent loans to finance one-to-four-family housing unit construction, purchases, or refinancing.^ This is not unusual since savings and loan associations are hone financing specialists. Total mortgage loans, which comprised 84.0 per cent of the total assets in 1960, have grown steadily, as shown in Table 28, froa $13.7 billion in 1950 to $59.9 billion as of December United States Savings and Loan League, Savinga and Loan 1961 Fact Book (Chicago: The League, I55TJ7" p. 85l 170 TABLE 27 AVERAGE ASSETS OF UNITED STATES SAVINGS ASSOCIATIONS BY TYPE OF CHARTER DECEMBER 31, 1960 (IN MILLIONS OF DOLLARS) ^sssxa5ss5ssss= = = sasxsss5sssaE csac3S E aB saB aaaE S S S E saB 8B s& naB sasE S S sasss5ss Number of Associa- Assets_____ Typa of Chart ar_____________tlona______fotal Average State Chartered Noninsured 2,222 5 3,884 5 1.8 Insured 2.225 28.919 13.0 Total State Chartered 4,447 532,803 5 7.4 Federally Chartered 1,873 538.511 520.6 Grand Total 6,320 571,314 511.3 Source: Derived from Federal Home Loan Bank Board Annual Coablned Financial Statements, 1950- 1960 and the United States Savings and Loan League 1961 Fact Book. TABLE 28 TOTAL ASSETS OF ALL SAVINGS ASSOCIATIONS (IN MILLIONS OF DOLLARS) Year (Decem ber 31) Mortgage Loansa U.S. Government Securities Cash on Hand and in Banks Real Estate Onrnedb FHLB Stock Other Assets Total Assets 1950 $ 13,714 $ 1,489 $ 951 S 21 * 177 S 494 f 16,846 1951 15,610 1,606 1,082 13 263 590 19,164 1952 18,416 1,791 1,306 21 309 742 22,585 1953 21,957 1,923 1,500 20 362 876 26,638 1954 26,088 2,005 1,962 24 429 1,000 31,508 1955 31,354 2,319 2,085 33 507 1,235 37,533 1956 35,719 2,743 2,116 42 597 1,564 42,781 1957 39,969 3,154 2,163 51 678 2,038 48,053 1958 45,478 3,785 2,571 74 762 2,308 54,978 1959 52,990 4,471 2,211 104 858 2,767 63,401 1960 59,932 4,616 2,669 158 978 2,961 71,314 aGross, not adjusted for Mortgage-pledged shares. ^Excludes office buildings. Source: Insured associations data are froa the Federal Home Loan Bank Board Annual Combined Financial Statements, 1950-1960; noninsured associations data are froa the United States Savings and Loan League 1961 Fact Book. 31, I960. It is significant to note the importance of the conventional, or in other words, nongovernment- underwritten, loans in the aggregate mortgage portfolio of the associations. The upward trend of conventional loans is revealed by the fact that this type of loan increased from 58.0 per cent of year-end assets in 1950 to 69.0 per cent in I960. Meanwhile, in the government-underwritten sector FHA-insured loans remained approximately the same (5.0 per cent of total assets) during the entire period, whereas VA-guaranteed mortgages decreased in relative importance from 18.0 per cent in 1950 year-end assets to only 10.0 per cent in I960.2 The liquidity needs of savings and loan associa tions are provided by assets invested in three other types of investments. These are cash on hand and in banks, United States government securities, and obliga tions of government agencies. The largest item of the three liquidity type assets includes holdings of government's direct issues, bonds, bills, notes, and certificates comprised of $4.6 billion in 1960. Cash on hand and in banks, which includes the small amounts of vault cash, the second largest of the liquidity 173 type assets, is available for meeting cash demands. These resources amounted to $2.7 billion at year-end 1960. Nine hundred seventy-eight million of the $71.3 billion total assets was invested in holdings of Federal Home Loan Bank stock. Savings capital, the major liability item, had a net gain of $7.6 billion during I960. As reported in Table 29, savings capital supplied by private sources amounted to more than $62 billion at year- end 1960. This accounted for 87.0 per cent of the total liabilities. In only two other years during the 1950-1960 period was savings capital higher relative to total liabilities. That is, in both 1957 and 1958 the savings sector accounted for 87.1 per cent of aggregate liabilities. In relative terms the low point during this period was 83.0 per cent at year-end 1950. Thus, this proportion has typically fluctuated within a narrow range from year to year. The 1960 percentage was "the third highest within a decade and reflects the relatively easy credit J I Ibid., p. 88. This source reports that no savings originate from the United States Treasury or other agencies of the government. Individuals, including husband and wife joint accounts, are the principal account holders. Also, trust funds and organisations of various types have only recently begun to gain in importance as association savings account holders. TABLE 29 TOTAL LIABILITIES OF ALL SAVINGS ASSOCIATIONS (IN MILLIONS OF DOLLARS) Year (Decem ber 31) Savings Capital FHLB Advances Reserves and Undivided Profits Loans in Process Other Liabilities Total Liabilities 1950 $ 13,978 $ 880 $ 1,280 $ 403 $ 305 $ 16,846 1951 16,073 884 1,453 419 335 19,164 1952 19,143 934 1,658 501 249 22,585 1953 22,778 1,014 1,895 560 391 26,638 1954 27,164 932 2,180 791 441 31,508 1955 32,058 1,522 2,534 920 499 37,533 1956 37,073 1,343 2,912 899 554 42,781 1957 41,856 1,373 3,321 864 639 48,053 1958 47,894 1,427 3,796 1,146 715 54,978 1959 54,583 2,344 4,353 1,322 799 63,401 1960 62,142 2,184 4,922 1,203 863 71,314 aIncludes $200 million owed to other institutions, mainly commercial banks. Source: Insured associations data are from the Federal Home Loan Bank Board Annual Combined Financial Statements, 1950-1960; noninsured associations data are fron the United States Savings and Loan League 1961 Fact Book. 174 175 situation which prevailed in 1960. At such tines 4 less use is aade of FHLB advances." It is noteworthy that Federal Hone Loan Bank (FHLB) advances decreased fron the record high of nore than $2.3 billion at year-end 1959 to less than $2.2 billion at year-end I960. This najor liability iten, which represents borrowings by the associations fron their regional banks and a relatively snail aaount fron other institutions (nainly connercial banks), has developed an erratic but, none the less, upward trend in its year-end totals. That is, this najor liability iten reflected increases in nost of the years during the 1950-1960 period. There were only three years— 1954, 1956, and 1960— when decreases fron prior year-end total FHLB advances were reported. However, at no tine during this period did FHLB advances exceed the 5.2 per cent of total liabilities reported in 1950. The low point in this latter rela tionship was 2.6 per cent in 1958. The reserves and undivided profits iten is the second largest najor liability of the savings and loan associations. The reserve accounts in the fern of reserves, surplus, and undivided profits grew $569 nillion to record a new high of alnost $5 billion in I960. This enabled the associations to raise their 4Ibid. 176 composite reserve position to 6.9 per cent of total assets. III. THE CALIFORNIA SAVINGS AND LOAN INDUSTRY Structure and Growth California savings and loan associations reached a record nuaber and all-tiae peak in their growth by year-end 1960. The seven new associations which were chartered in 1960 raised the total nuaber of associations in California to 244. During the 1950-1960 period insured associations grew aore than 54.8 per cent, froa 155 to 240 in nuaber, a net gain of eighty-five. More important, however, was the growth which occurred in the nuaber of insured state- chartered associations, as reported in Table 30. The latter group aore than doubled its nuaber froa eighty- one in 1950 to 171 in 1960. Furthermore, this upward growth trend was uninterrupted during the period. Meanwhile, federally chartered associations displayed a downward but not uninterrupted trend in the saae decade. The nuaber of federally chartered associa tions in operation decreased 6.8 per cent, froa seventy-four in 1950 to sixty-nine in 1960, for a net loss of five. The decline in the nuaber of these associations had reached its low point in 1955, when TABLE 30 177 THE STRUCTURE AND GROWTH OF THE CALIFORNIA SAVINGS AND LOAN INDUSTRY: NUMBER OF ASSOCIATIONS BY TYPE OF CHARTER, 1950-1960 Ytar (D«cn- State Chartered Federally Grand ber 31> Insured Woninsurefl Total Chartered Total 1950 81 19 100 74 174 1951 89 13 102 74 176 1952 92 13 105 73 178 1953 101 8 109 71 180 1954 114 8 122 69 191 1955 131 7 138 66 204 1956 136 7 143 69 212 1957 147 6 153 70 223 1958 160 4 164 69 233 1959 165 4 169 68 237 1960 171 4 175 69 244 .950-1960 Change .950-1960 Percentage Change (Decrease) 90 (15) 75 (5) 70 111.1 (78.9) 75.0 (6.8) 40.2 Source: Federal Hone Loan Bank Board Annual Conblned Financial Statements, 1950-1960. 178 only sixty-six federally chartered associations were operating in California. Noninsured state chartered associations, on the other hand, decreased steadily froa nineteen in 1950 to only four at year-end 1960, a decline of alaost 79.0 per cent, or a net loss of fifteen associations of this type. The growth in the total assets of savings and loan associations located in California was even aore draaatic them the net gain in the nuaber of California associations. Total assets of these associations grew froa Si.5 billion in 1950 to alaost $11 billion in 1960. This represents a 625.0 per cent increase, as shown in Table 31. The aost significant growth in total assets, however, was reported by insured state chartered associations. Total assets of the latter associations increased froa $573 Billion in 1950 to aore than $6 billion in 1960. The actual net gain of $5.6 billion represents a phenoaenal growth of 977.3 per cent. Meanwhile, federally chartered asso ciations reported a net gain of $3.7 billion for a aodest 428.8 per cent increase* Noninsured state chartered associations, on the other hand, declined 56.3 per cent in total assets froa $48 Billion in 1950 to only $21 aillion in 1960, a net decrease of $27 aillion. It is worthy to note that federally 179 TABLE 31 THE STRUCTURE AND GROWTH OF THE CALIFORNIA SAVINGS AND LOAN INDUSTRY: ASSETS OF ASSOCIATIONS BY TYPE OF CHARTER, 1950-1960 (IN MILLIONS OF DOLLARS) Year (Deeen- State Chartered Federally Grand taer 31) Insured Noninsured Total Chartered Total 1950 8 573 * 48 8 621 8 862 8 1,483 1951 695 36 731 968 1,669 1952 917 42 959 1,219 2,178 1953 1,215 19 1,234 1,440 2,674 1954 1,599 21 1,620 1,707 3,327 1955 2,150 22 2,172 1,961 4,133 1956 2,642 24 2,666 2,303 4,969 1957 3,286 25 3,311 2,648 5,959 1958 4,011 19 4,030 3,216 7,246 1959 5,045 20 5,065 4,023 9,088 1960 6,173 21 6,194 4,558 10,752 1950-1960 Dollar Change $5,600 (27) 85,573 83,696 8 9,269 1950-1960 Percentage Change (Decrease) 977.3 (56.3) 897.4 428.8 625.0 Source: Federal Hone Loan Bank Board Annual Contained Financial Statenente, 1950-1960. 180 chartered associations in California held higher average asset amounts than did the other two types of associations. This was also true at the national level. Data in Table 32 indicate that the average asset amount held by federally chartered associations in California in 1960 was $66.1 million as compared with the $44.1 million average assets of all associa tions in California. Average assets of insured state chartered associations ($36.1 million) were slightly lower than the statewide average, and noninsured state chartered associations were even considerably lower at $5.3 million. The competitive shifts in relative importance which occurred among the associations operating in California during the 1950's are effectively shown in Table 33. The most important competitive shift was the change which occurred in the role of the insured associations. By 1960 insured state chartered associa tions had obtained almost 58.0 per cent of aggregate association assets in California. Their share of total assets in 1950 was less than 39.0 per cent of the aggregate. In contrast, federally chartered associa tions in California, which held more than 58.0 per cent of total assets in 1950, reported their share as being slightly more than 42.0 per cent in 1960. Noninsured 181 TABLE 32 AVERAGE ASSETS OF CALIFORNIA SAVINGS ASSOCIATIONS BY TYPE OF CHARTER, DECEMBER 31, 1960 (IN MILLIONS OF DOLLARS) Tvds of Charter Number of Associa tions Assets Total Averaxe State Chartered Noninsured 4 $ 21 $ 5.3 Insured 171 6.173 36.1 Total State Chartered 175 $6,194 $ 35.4 Federally Chartered 69 $4,558 $ 66.1 Grand Total 244 $10,752 $ 44.1 Source: Derived from Federal Home Loan Bank Board Annual Combined Financial Statements, 1950-1960. 182 TABLE 33 CHANGE IN SHARE OF TOTAL ASSETS OF THE CALIFORNIA SAVINGS AND LOAN INDUSTRY BY TYPE OF ASSOCIATION CHARTER, 1930 AND 1960 Decrabtr 31, Deceaber 31, Type of Charter________________1950____________1960 State Insured Noninsured Federal Total 38.6% 3.2 58.2 100. 0% 57.4% .2 42.4 100. 0# Source: Derived froa Federal Hone Loan Bank Board Annual Coabined Financial Stateaents, 1950-1960. 183 state chartered associations reflected a decrease froa 3.2 per cent of total assets to considerably less than 1.0 per cent of aggregate assets in California associations. In summary, therefore, it was shown that (1) the insured state chartered associations in California reported the greatest growth in both number and total assets, (2) the numerically fewer federally chartered associations are relatively larger in terms of size of average assets per association, and (3) the non insured state chartered associations in California are relatively unimportant. Ma.lor Assets and Liabilities of California Associations The two major asset items of California savings associations are the mortgage loans and the combined account, cash, and United States government obliga te tions. Total mortgage loans, which were 85.1 per K California savings associations assets and liabilities figures in this section refer to savings and loans insured by the FSLIC. The uninsured asso ciations' data are not included in this section, therefore there is an insignificant understatement of the data. The assets are understated 0.28 per cent and the total savings are understated 0.2 per cent when all associations are to be considered. See California Savings and Loan League, California Savings and Housing 1962 Data Book (Pasadena: tbe League, 19o2), pp.l6-l8. 184 cent of total assets In 1960, had grown more than 645.6 per cent during the decade of the 1950's. In dollar aaounts the net gain was alaost $8 billion, which is the largest absolute gain reported in any of the iteas in Table 34. The reported growth in the coabined itea "cash and United States governaent obligations" was a aodest 514.3 per cent froa $161 aillion in 1950 to alaost $1 billion in 1960. The aost dramatic growth, in relative terns, in any of the najor assets or liabilities iteas was reported in the savings account itea. During the 1950-1960 period savings accounts in insured California associations increased without interruption alaost seven tines for a phenoaenal 693.3 per cent increase. The savings accounts growth, for the aost part, aay be attributed to aggressive nanagement policies which successfully competed for the public's savings dollars. The latter were the result of a continued high level of prosperity and employment in the nation in general and, aore specifically, in California. The trend of another najor liabilities itea, "FHLB advances and other borrowed aoney," is particu larly worthy of being noted. The upward trend of this itoa, as shown in Table 34, has not been without interruption. That is, decreases froa prior year-end TABLE 34 MAJOR ASSETS AND LIABILITIES <F INSURED SAVINGS ASSOCIATIONS OPERATING IN CALIFORNIA, 1950-1960 (IN MILLIONS OF DOLLARS) Major Asset Iteas Major Liability Iteas Year Tetal Mortgage Loans Cash and U.S. Governaent Obligations Savings Accounts fhlb Advances and Other Borrowed Monev Reserves and Undivided Profits 1950 9 1,226 1 161 9 1,120 9 108 9 99 1951 1,412 186 1,351 83 120 1952 1,820 232 1,708 137 138 1953 2,278 280 2,198 135 176 1954 2,835 348 2,725 128 220 1955 3,516 421 3,382 201 281 1956 4,214 493 4,182 155 336 1957 5,068 582 4,959 281 441 1958 6,045 823 6,080 278 513 1959 7,624 944 7,335 615 626 1960 9,141 989 8,885 598 759 1950-1960 Dollar Increase « 7,915 9 828 9 7,765 9 490 9660 1950-1960 Percentage Increase 645.6 514.3 693.3 453.7 666.7 Source: Federal Hose Loan Bank Board, as reported in California Savings and Housing 1962 Data Book (Pasadena: California Savings and Loan League. msy, pp . ittt? : ----- 186 figures, which would indicate repayments of advances or other loans, were reported in six of the ten years. Modest increases were reported in three years, namely, 1952, 1955, and 1957. However, the 1959 increase over 1958 amounted to $337 million, which represents an increase of more than 121.0 per cent in this figure. The importance of these data is explained by the fact that the Federal Home Loan Bank System acts in a reserve credit capacity for these associations. As g such, certain monetary theorists have made the charge that FHLB activities can and at times do result in actions contrary to the stated monetary policy of the Federal Reserve Board. These theorists point out that associations can increase their lendable funds composed of net savings increases and loan amortisation payments by borrowing from their regional FHLB. The data con cerning the insured California associations' actions during four periods of alternating ease and restraint may be construed to support the charge. That is, in the years 1951, 1952, and 1953, which were generally considered to be years of restraint, insured California associations borrowed $29 million net (the difference g Among the more notable are John G. Gurley and Edward S. Shaw, Money in a Theory of Finance (Wash ington: The Brookings Tnstitution,"T960). 187 between $108 Billion as of December 31, 1950 and $137 Billion as of year-end 1952) during 1951 and 1952 and repaid $2 Billion (net) during 1953. Also, during another period of general restraint conditions, 1956-1957, net borrowings by insured California asso ciations amounted to $126 million. These data, therefore, would appear to support the charge. How ever, the remaining evidence weakens the seriousness of the charge since the other two times when net borrowings were made occurred during periods of general ease. More specifically, net borrowings increased $73 million in 1955, and $337 million in 1959. The latter net increase represents a 121.2 per cent increase over the 1958 volume. These activities, therefore, tend to reduce the seriousness of the charge's implications at the state level. The effect of aonetary policy changes on the mortgage lending activities of financial institutions in the Sacramento Metropolitan Area is discussed in Chapter VI. 188 IV. INSTITUTIONAL LENDERS'ROLE IN PROVIDING REAL ESTATE CREDIT IN CALIFORNIA AND THE NATION Institutional Mortran Landing Activitiaa in tha Unitad Stataa In tha past tha only statistical sarias which reported tha total volume of residential mortgage financing was that compiled by the Federal Home Loan Bank Board's Operation Analysis Division. This monthly series, known as the "Nonfara Mortgages Recorded of $20,000 or Less," reports both the number and dollar amount of recordings by six types of lenders. These data represent fairly accurately the extent of financing of one-to-four-family nonfarm housing units, even though loans above $20,000 are excluded. The latter omissions are offset by the inclusion of an approximately equal amount of loans on property other than residential properties. Total nonfarm mortgage recordings of $20,000 or less, as indicated in Table 35, amounted to aore than $29 billion in 1960, second only to 1959 when over $32 billion nonfarm mortgage loans were recorded. The 9.0 per cent decrease reflected the "falling off in new dwelling units put in place . . • and a market TABLE 35 NONFARM MORTGAGE RECORDINGS OF $20,000 OR LESS, IN THE UNITED STATES, BY TYPE OF LENDER, 1950-1960 (IN MILLIONS OF DOLLARS) Year Savings and Loan Associations Mutual Savings Banks Connercial Banks Insurance Cooroanies Indi viduals Other Mortgagees Total 1950 $ 5,060 $1,064 $3,365 $1,618 $2,299 $2,773 $16,179 1951 5,295 1,013 3,370 1,615 2,539 2,573 16,405 1952 6,452 1,137 3,600 1,420 2,758 2,651 18,018 1953 7,365 1,327 3,680 1,480 2,841 3,054 19,747 1954 8,312 1,501 4,239 1,768 2,882 4,272 22,974 1955 10,452 1,858 5,616 1,932 3,362 5,264 28,484 1956 9,532 1,824 5,458 1,799 3,558 4,918 27,088 1957 9,217 1,430 4,264 1,472 3,554 4,307 24,244 1958 10,516 1,640 5,204 1,460 3,434 5,134 27,388 1959 13,094 1,780 5,832 1,523 3,946 6,060 32,235 1960 12,158 1,557 4,520 1,318 4,001 5,787 29,341 Source: Federal Hoae Loan Bank Board Annual Savings and Hoae Financing Source Book, 1950-1960. 190 for older properties that was somewhat more sluggish than in 1959."7 The competitive aspects of the sources of mortgage credit are revealed by the data which Indicate that savings and loan associations continue to supply the greatest amount of residential mortgage financing in terms of dollar amounts, share of total dollar amount, and the absolute number of loans recorded. For example, savings and loan associations supplied more than $12 billion for financing residential mortgages in I960. Other mortgagees, mainly mortgage bankers, furnished $5.8 billion of mortgage credit in 1960. Commercial banks provided $4.5 billion; whereas mutual savings banks recorded $1.6 billion and insur- Q ance companies supplied only $1.3 billion. The dollar volume of funds provided by the savings and loan associations has ranged froa a low of $5.1 billion in 1950 to a high of $13.1 billion in 1959. Mutual savings banks recorded a much narrower range (slightly less than $1 billion) during the 1950-1960 7 United States Savings and Loan League, op. eft., p. 37. 8 Since insurance companies purchase mortgages originated by mortgage bankers and others, the $1.3 billion figure nay be an understatement. 191 period. The dollar voluae range of nonfara recordings nade by conmercial banks approxinated $2.4 billion during this sane period. The range of the insurance conpany recordings during the period was approxiaately one-half billion dollars; in contrast, other aort- gagees' recordings revealed a range of $2.5 billion which, in effect, was the second largest fluctuation reported by the institutional type lenders. The aore inportant coapetitive shifts in the relative iaportance of the suppliers of residential nortgage financing which have occurred during the 1950-1960 period are shown in Table 36. Savings and loan associations have increased, alaost without interruption, their share of the total nonfarn nortgage recordings. These associations have been successful in increasing their share of the aggregate voluae froa 31.3 per cent in 1950 to 41.4 per cent in I960. This was the highest percentage ever recorded by any one of the types of lenders. "The total lending voluae of associations reflects . . • not only the rise in the deaand for hoae nortgage credit but also the increasing share of the total o accounted for by associations." g United States Savings and Loan League, op. cit., p. 39. TABLE 36 NONFARM MORTGAGE RECORDINGS OF $20,000 OR LESS, IN THE UNITED STATES, BY TYPE OF HINDER, 1950-1960 (PERCENTAGE DISTRIBUTION BASED ON DOLLAR VOLUME) Year Savings and Loan Associations Mutual Savings Banks Connercial Banks Insurance Coauanies indi viduals Other Mortcacees Total 1950 31.3 6.6 20.8 10.0 14.2 17.1 100.0 1951 32.3 6.2 20.5 9.8 16.1 15.1 100.0 1952 35.8 6.3 20.0 7.9 15.3 14.7 100.0 1953 37.4 6.7 18.6 7.5 14.4 15.4 100.0 1954 36.2 6.5 18.5 7.7 12.5 18.6 100.0 1955 36.7 6.5 19.7 6.8 11.8 18.5 100.0 1956 35.2 6.7 20.2 6.6 13.1 18.2 100.0 1957 38.0 5.9 17.6 6.1 14.7 17.7 100.0 1958 38.4 6.0 19.0 5.3 12.5 18.8 100.0 1959 40.6 5.5 18.1 4.7 12.2 18.0 100.0 1960 41.4 5.3 15.4 4.5 13.6 19.8 100.0 Sourest Federal Hoee Loan Bank Board Annual Savings and Hoae Financing Source Book, 1950-1960. 193 Since 1954 the "other Mortgagees" type of lender increased in iMportance and provided approxi mately 18.0 per cent or aore of the total nonfara mortgage recordings. Commercial banks, on the other hand, have decreased their share of the aggregate hone financing. From a 1950 high of almost 21.0 per cent of the total, commercial banks have shown a steady decrease to a low of 15.4 per cent in 1960. Mutual savings banks and insurance companies combined accounted for approximately 10.0 per cent to 16.0 per cent of the annual nonfarm mortgage recordings. How ever, the relative share of funds provided by these two types of institutions has shown a definite downward trend. During the 1950-1960 period the absolute number of nonfara mortgage recordings remained conatant between three and four aillion recordings a year (Table 37). In contrast, the aggregate dollar volume amount loaned ranged froa $16 billion to $32 billion (Table 35, page 189). In both respects savings and loan associations led all the other types of lenders throughout the period. Furthermore, in each of the years since 1952 the savings and loan associations recorded more than one aillion loans annually. Con sequently, the savings and loan associations made almost one-half of all the loans recorded during the TABLE 37 NONFARM MORTGAGE RECORDINGS OF $20,000 OR LESS, IN THE UNITED STATES, BY TYPE OF LENDER, 1950-1960 (IN THOUSANDS OF RECORDINGS) Year Savings and Loan Associations Mutual Savings Banks Connercial Banks Insurance Coananies Indi viduals Other Mortgagees Total 1950 935 165 628 227 610 466 3,031 1951 902 145 583 207 631 410 2,878 1952 1,027 153 600 171 671 407 3,029 1953 1,109 168 599 168 682 438 3,164 1954 1,198 183 661 186 675 554 3,457 1955 1,373 210 783 191 726 631 3,914 1956 1,202 197 724 169 734 575 3,601 1957 1,122 153 587 134 736 514 3,246 1958 1,213 165 668 126 692 577 3,441 1959 1,378 171 704 125 757 647 3,782 1960 1,260 149 574 105 756 628 3,472 Source: Federal Hoae Loan Bank Board Annual. Savings and Hoae Financing Source Book, 1950-1960. 194 195 period. There can be little doubt as to the importance of the mortgage lending activities of the savings and loan industry in the United States. Institutional Mortgage Lending Activities in California The monthly series compiled by the FHLB known as the "Nonfara Mortgages Recorded of $20,000 or Less" is particularly useful for describing the competitive shifts that have occurred among the financial institu tions which extend credit in the dynamic California real estate markets. Total nonfarm mortgage recordings of $20,000 or less in California have been dominated by savings and loan associations in the years between 1950 and 1960, as reported in Table 38. California savings associations loaned $666 million in 1950 and advanced increasing amounts each year until a peak of more than $2.6 billion was reached in 1959. Savings and loan associations and commercial banks reported (Table 39) alaost identical shares of the nonfara mortgage recordings of $20,000 or less during 1950, when their respective shares were 27.8 and 27.3 per cent. During the following ten years, however, the savings association share increased almost steadily until 1963, when 43.0 per cent of the total was advanced by them. In contrast, the commercial banks' TABLE 38 NONFARM MORTGAGE RECORDINGS OF $20,000 OR LESS, IN CALIFORNIA, BY TYPE OF LENDER, 1950-1960 (IN MILLIONS OF DOLLARS) Year Total Savings Associations Insurance Companies Coaaercial Banks Individuals Other Lenderi 1950 $ 2,392 $ 666 $ 345 $ 653 $ 473 I 255 1951 2,455 655 375 626 521 278 1952 2,633 854 262 647 599 271 1953 3,005 940 279 671 687 428 1954 3,436 1,046 297 795 717 581 1955 4,776 1,497 288 1,251 990 750 1956 4,952 1,517 259 1,325 1,135 716 1957 4,549 1,634 252 856 1,153 654 1958 5,045 1,775 268 1,214 1,089 699 1959 6,613 2,641 260 1,446 1,367 898 1960 5,987 2,573 232 714 1,414 1,054 Source: Federal Hobo Loan Bank Board, as reported in California Savings and Housing 1962 Data Book (Pasadena: California Savings and Loan League, “1552), p» 257“ “ '^ 196 197 TABLE 39 PROPORTION OF HOME MORTGAGES RECORDED, IN CALIFORNIA, BY TYPE OF LENDER, 1950-1960 Year Savings Associa tions Insurance Coananies Coa- aercial Banks Indi viduals Other Lenders 1950 27.896 14.496 27.396 19.896 10.796 1951 26.7 15.3 25.5 21.2 11.3 1952 32.4 10.0 24.6 22.7 10.3 1953 31.3 9.3 22.3 22.9 14.2 1954 30.4 8.6 23.1 20.9 16.9 1955 31.3 6.0 26.2 20.7 15.7 1956 30.6 5.2 26.8 23.0 14.4 1957 35.9 5.5 18.8 25.3 14.4 1958 35.2 5.3 24.0 21.6 13.8 1959 39.9 3.9 21.9 20.7 13.6 1960 43.0 3.9 11.9 23.6 17.6 Source: Federal Hoae Loan Bank Board, aa reported in California Savings and Housing 1962 Data Book (Pasadona: California Savings and Loan League, 1352), p. 23. 198 share of the total recordings declined to a low of less than 12.0 per cent of the total in I960. The peak in the connercial banks' share of nortgage lend ing activity was in 1955 and 1956 when they recorded nore than 26.0 per cent of the nortgages each year. Third ranked anong the institutional lenders was the group identified as "other lenders." This group, conprised of nortgage bankers for the nost part, advanced sufficient funds in each of the years since 1951 to outrank the insurance conpanies both as to dollar volune and share of total nonfara nortgage recordings of $20,000 or less. The other lenders group advanced amount■ each year which were larger than the previous year's total, whereas insurance coapany volune of originations declined steadily throughout the period. That is, the forner group originated $255 aillion in 1950, and in each year since 1951 noderately larger aaounts were advanced until the peak was reached in 1960 when nore than $1 billion was originated. In contrast, insurance conpanies1 originations slipped froa a peak $345 aillion in 1950 to a low $232 nillion in 1960. 199 V. TRENDS IN RESIDENTIAL CONSTRUCTION FUNDS PROVIDED BY SAVINGS AND LOAN ASSOCIATIONS Saving* Associations' Role in Financing Residential Construction in tha United States Additional insight into the significant role of savings and loan associations is afforded by an examination of the statistical series which reports the purpose of the mortgage loans recorded. In this FHLB statistical series loans made for residential construction purposes are distinguished from loans made for purposes of purchasing an existing home or other reasons. This important series reports the loan purpose of both insured and uninsured savings in the United States. The purpose of loan data for California associations, however, reflects only the insured associations' activities. The data with regard to purpose of loans made by all savings associations in the United States are presented in Table 40. Both annual loan volume and percentage of total loans are shown for the categories: construction, home purchase, and other. The latter category includes loans made for purposes of refinanc ing, modernisation, alteration, or other miscellaneous TABLE 40 MORTGAGE LOANS MADE IN THE UNITED STATES BY ALL SAVINGS ASSOCIATIONS, BY PURPOSE, 1950-1960 (DOLLAR AMOUNTS IN MILLIONS) Year Total Loans Construction Hoae Purchase Other Dollar Percentage of Total Dollar Percentage of Total Dollar Percentage of Total 1950 $ 5,237 1,767 33.7 2,246 42.9 1,224 23.4 1951 5,250 1,657 31.6 2,357 44.9 1,236 23.5 1952 6,617 2,105 31.8 2,955 44.7 1,557 23.5 1953 7,767 2,475 31.9 3,488 44.9 1,804 23.2 1954 8,969 3,076 34.3 3,846 42.9 2,047 22.8 1955 11,255 3,984 35.4 5,155 45.8 2,116 18.8 1956 10,325 3,699 35.8 4,620 44.8 2,006 19.4 1957 10,160 3,484 34.3 4,591 45.2 2,085 20.5 1958 12,182 4,050 33.2 5,172 42.5 2,960 24.3 1959 15,151 5,201 34.3 6,613 43.7 3,337 22.0 1960 14,304 4,678 32.7 6,132 42.9 3,494 24.4 Source: Federal Hoae Loan Bank Board, Savings and Hoae Financing Source Book. 1961 (Washington: Governaent Printing Office, l$6l), p. 26. 200 201 purposes. It can be seen from the data that hoae purchase loans comprised the greatest share of total loans made by savings and loan associations in the United States. Typically, this category produced approximately 43.0 per cent of total loans during the 1950-1960 period. Each year this purpose of loan category accounted for more than 42.0 per cent of total loans. The share of construction loans, in contrast, never exceeded 35.8 per cent of total loans. Its share in any one year was much lower than the lowest share of home purchase loans made during the period. Construction loans typically accounted for approximately 32.0 per cent of total loans. It appears, therefore, that loans to purchase existing homes predominated at the national level whereas the picture differed in California, as is shown in the succeeding section. Residential Construct ion Funds Provided by California Associations In the years between 1950 and 1960 California experienced many favorable changes in its economic base which led to increased employment opportunities and subsequent population growth. Among the many institutions which also experienced rapid growth during this same period were the savings and loan associations in California. That these institutions have played an important role in providing real estate credit throughout the state has been Indicated earlier. What is even more significant, however, is the fact that the major portion of total loans made by insured California associations in any one year during the 1950-1960 period was for purposes of financing the construction of nonfarm housing units. The data in Table 41 indicate that mortgage loans made by these institutions for construction purposes typically com prised more than 46.0 per cent of the total loans. At no time during the study period did construction account for less than 42.2 per cent of the total. The high point, in relative terms, occurred in 1954 when 51.9 per cent of the total loans made were for resi dential construction purposes. Mortgage loans made for purposes of home pur chases ranged from a high 36.0 per cent to a low 28.7 per cent of total loans made by insured California associations during the study period. Therefore loans made for this purpose ranked second in California, whereas home purchase loans had ranked first in the national data. Construction loans, which in contrast had ranked second at the national level, ranked first TABLE 41 MORTGAGE LOANS MADE BY INSURED CALIFORNIA ASSOCIATIONS, BY PURPOSE, 1950-1960 (DOLLAR AMOUNTS IN MILLIONS) Year Total Loans Construction Hone Purchase Other Dollar Percentage of Total Dollar Percentage of Total Dollar Percentage of Total 1950 $ 693 353 50.9 205 29.6 135 19.5 1951 653 301 46.2 209 31.9 143 21.9 1952 880 423 48.1 269 30.5 188 21.4 1953 983 451 45.9 319 32.5 212 21.6 1954 1,206 626 51.9 346 28.7 234 19.4 1955 1,575 741 47.1 558 35.4 276 17.5 1956 1,572 734 46.7 531 33.7 307 19.6 1957 1,767 776 43.9 637 36.0 354 20.1 1958 2,028 910 44.9 591 29.1 527 26.0 1959 2,961 1,289 43.5 948 32.0 724 24.5 1960 2,952 1,246 42.2 926 31.4 780 26.4 Source: Federal Hone Loan Bank Board, as reported in California Savings and Housing 1962 Data Book (Pasadena: California Savings and Loan League, 1962), p. IT. 203 204 in California. Loans made by insured California associations for other purposes ranged from a high of 26.4 per cent of total in 1960 to a low of 17.5 per cent of total in 1955. Thus in California, as in the United States, these loans produced the least stable share of total volume. Mortgage Credit Availability Impact on the Number of Housing Starts Financed by Associations Additional evidence that the availability of funds at savings and loan associations, coupled with favorable changes in the econoaic base, influenced the rate of residential construction may be found in Table 42. These data were developed to show the approxiaate annual amount of money available to savings and loan associations for real estate credit in each of the years 1950-1960, inclusive. This was accoaplished by the addition of the dollar amounts listed in each of the columns headed "mortgage repay ments, net savings, FHLB advances, and net income after dividends." The total of these annual amounts is what associations as a whole have available for loans on real estate. As shown below, the amount of money available to the associations for real estate TABD5 42 SAVINGS AND LOAN ASSOCIATION MORTGAGE CREDIT AVAILABILITT AND NUMBER CF CONSTRUCTION LOINS MADE BT THE ASSOCIATIONS IN THE UNITED STATES, 1950-1960 (DOLLAR AMOUNTS IN M3HJCNS) NTM3ERS IN THOUSANDS) Mortgage Tear Rewentg Net Serins® FHL3 Advance* Net Iheoee After Dividends Available for Landing Credit Avail ability Chinee Construction Construe— tlflD Aaount Nuaber Chance 1949 1950 8 2,351 3 >238 8 1,446 1,487 8 499 880 8 128 165 8 4,404 5,770 8 1,366 8 1,083 1,767 172 256 84 1951 3,354 2,080 884 178 6,496 726 1,657 221 (35) 1952 3,811 3,083 934 191 8,019 1,523 2,305 268 47 1953 4,226 3,642 1,014 231 9,133 1,094 2,476 295 27 1954 4,838 4,416 932 284 10,470 1,357 3,076 330 35 1955 5,989 4,891 1,522 355 12,757 2,28? 3,984 400 70 1956 5,960 5,006 1,343 387 12,696 (61) 3,699 353 (47) (38) 1957 5,930 4,765 1,373 408 32,456 (240) 3,484 315 1958 6,673 6,063 1,427 476 14,639 2,183 4,050 329 14 1959 7,639 6,604 2,344 550 17,337 2,498 5,201 401 72 I960 7,362 7,559 2,184 560 17,665 528 4,678 376 (25) () Denotes a decrease, compared to prior year. H i Source: Based on data provided by tbs Operating Analysis Division of tbs Federal Ioai Baik Board) Ubited States Savings and Loan Leagoe* 205 206 credit has grown steadily froa $5.8 billion in 1950 to aore than $17 billion in 1960— a growth of over 200.0 per cent during the study period. For purposes of this analysis data showing changes in the availability of credit are aore aean- ingful than the annual aggregates discussed above. The data which reflect the changes in the availability of savings and loan aortgage credit were derived by deducting the prior year-end aaounts available for lending froa each succeeding year-end figure. These changes ranged froa a decrease of $240 aillion in 1957 to an increase of $2.5 billion in 1959. The iapact of the savings and loan aortgage credit avail ability aay be observed by relating these changes to the changes in the dollar voluae and nuaber of resi dential construction loans financed by the associations. The nuaber of residential construction loans financed by associations during the 1950-1960 period ranged froa a low of 221,000 in 1951 to a high of 401,000 in 1959. As a result, the change in the nuaber of resi dential construction loans fluctuated froa a decrease of 47,000 to an increase of 84,000 loans. The dollar voluae of construction loans aade annually during the saae period fluctuated froa a low of $1.1 billion to a high of $5.2 billion. The change in the dollar 207 volume of construction loans ranged from a decrease of $523 million to an increase of $1.2 billion. An analysis of the data indicates that increased (over prior year) availability of money for real estate credit was accompanied by an increased (over prior year) dollar volume of loans and number of residential construction starts. In the years in which the amount of money available for real estate credit did not equal or surpass the prior year's increase, the change in the dollar volume and number of construction loans also decreased compared to the prior year change. More specifically, the data clearly show that (1) in the years 1950, 1952, 1953, 1954, 1955, 1958, and 1959 increased credit availability was accompanied by increased residential construction loans by savings and loan associations; (2) in the years 1951 and 1960 smaller increases, compared to the previous years' increases, in credit availability were accompanied by decreases (prior year basis) in residential construc tion loans financed by associations; and (3) in the years 1956 and 1957 decreased credit availability was accompanied by decreases (prior year basis) in asso ciations' construction loans. It should be noted that the associations received record advances froa the Federal Home Loan 208 banks in order to provide funds for real estate credit during the years 1959 and 1960. FHLB advances in the record amount of $2.3 billion were received by the associations in 1959. This represents a 64.2 per cent increase over the advances received in 1958. These FHLB advances demonstrate the willingness of associa tions to borrow in order to maintain their credit availability balances to finance residential construc tion loans. As expected, changes in the associations' credit availability had a more direct impact on changes in residential construction loans financed by associations than on changes in housing starts in the United States as a whole. Nevertheless, one authority plotted the pertinent data, and the resultant graphic illustration indicated that there was a positive relationship between the major shifts in nonfarm private housing starts and the major shifts in the number of construction loans made by savings and loan associations during the 1950-1960 period.1’ 0 In summary, therefore, there can be no doubt as to the significant contribution which the savings and loan associations have made in the financing of 10Federal Home Loan Bank Board, Savings and Howe Financing Chart Book. 1981 (Washington: Government Printing Of/ice, ld6l), pp. 19a-19. residential construction units in the United States during the 1950-1960 period. Furtheraore, the contribution which insured associations have made in financing residential construction in California has been even aore significant. The role of the Sacraaento savings and loan associations is discussed in the chapter which follows. CHAPTER VI THE ECONOMIC IMPACT OF SAVINGS AND LOAN ASSOCIATIONS ON RESIDENTIAL CONSTRUCTION IN THE SACRAMENTO AREA I. INTRODUCTION The two fold purpose of this chapter is to present (1) an analysis of eapirical data which relates the inpact of structural changes and growth of the savings and loan industry in the Sacraaento Metro politan Area on residential construction in the area, and (2) to report the findings of a survey which exaained the effects o£ changes in the Federal Reserve Board's nonetary policy on the aortgage lending activities of certain local financial institutions. The priaary objective of this chapter, therefore, is to test that portion of the hypothesis which states that the increased rates of residential construction in an area are dependent upon the ability and willing ness of savings and loan associations to provide funds for real estate credit. Iaaediately following the introductory section of this chapter is an analysis of the coaposition and growth of the savings and loan industry in Sacraaento. This is followed by a 211 discussion of major assets and liabilities held by the local associations and their role in financing local residential construction. The Sacramento mortgage market section presents empirical data concerning the institutional lenders' local real estate credit activities. The final section of this chapter discusses the implications of monetary policy changes at the local level. II. THE SACRAMENTO SAVINGS AND LOAN INDUSTRY Composition and Growth The number of savings and loan associations which constitute the Sacraaento savings and loan industry grew froa three in 1990 to seven in 1960, a 133.3 per cent increase, as shown in Table 43. This over-all rate of growth is extraordinary when compared with the over-all rate of growth of federally and state chartered institutions at the national level (5.5 per cent) and the state (California) level (40.2 per cent) during this same period. During the study period the nuaber of state associations located in Sacraaento increased froa one to six.1 This growth represents a 500.0 per cent ^Includes one local aain office facility of a statewide holding coapany operation} excludes one 212 TABLE 43 THE COMPOSITION AND GROWTH OF THE SACRAMENTO SAVINGS AND LOAN INDUSTRY: NUMBER OF ASSOCIATIONS BY TYPE OF CHARTER, 1950-1960 Source of Chart T a Year__________Federal*^ StaTtc___________Total 1950 2 1 3 1951 2 1 3 1952 2 1 3 1953 2 2 4 1954 1 3 4 1955 1 3 4 1956 1 3 4 1957 1 4 5 1958 1 5 6 1959 1 6 7 1960 1950-1960 Percentage Change 1 (50.0) 6 7 (Decrease) 500.0 i — u 01 . aThere are no uninsured associations in Sacraaento. bOne federal association was converted to a state charter in 1954. cIncludes one local aain office facility (charter approved, 1953) of a statewide holding coapany operation; excludes one local branch office (branch approved, 1959) of a northern California holding coapany operation. Source: Federal Hone Loan Bank of San Francisco, Roster of Meabers. annually 1950-1961; Savings and Loan Coaaissioner, Annual Reports. 1950-1961 (Sacraaento: State Printing 6rfice). 213 increase at the local level, in contrast with the 66.7 per cent increase at the national level and the 111.1 per cent increase at the state (California) level. Total assets of Sacraaento associations have also grown at rates which exceed the national and state rates of total asset growth. Table 44 indicates that the total assets of the local associations have grown aore than 1,700.0 per cent during the years between 1950 and 1960. In contrast, the rate of growth at the national level was only 323.3 per cent and at the state (California) level it was 625.0 per cent during this saae period. Furtheraore, the total assets of insured state chartered associations within the United States increased alaost 460.0 per cent, and the total assets of those within the state of California increased alaost 1,000.0 per cent. In the aeantiae the total assets of state chartered associations located in Sacraaento grew alaost 28,000.0 per cent to a level of $254 per capita, in contrast to the $392 per capita for the whole state. 1 (continued) local branch office of a northern California holding coapany operation. Unfortunately, individual data for the latter branch office have becoae available only recently. 214 TABLE 44 THE COMPOSITION AND GROWTH OF THE SACRAMENTO SAVINGS AND LOAN INDUSTRY: ASSETS OF ASSOCIATIONS BY TYPE OF CHARTER, 1950-1960 (AMOUNTS IN THOUSANDS OF DOLLARS) Year Source of Charter* Ptdcral Staie Combined Asiats 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1950-1960 Dollar Increase 1950-1960 Percentage Increase $ 6,624 6,914 7.631 9.632 7,962 9,429 10,616 10,539 12,706 15,473 19,054 $ 12,430 187.6 6 387 346 336 987 11,688 18,053 22,085 32,873 51,349 73,750 108,726 $108,339 27,994.6 $ 7,011 7,260 7,967 10,619 19,650 27,482 32,701 43,413 64,055 89,222 127,780 $120,769 1,722.5 aThere are no uninsured associations in Sacraaento. Source: Federal Hone Loan Bank of San Francisco, Roster of Members. annually 1950-1961; Savings and Loan Commissioner, Annual Reports. 1950-1961 (Sacramento: State Priniinjg 6r£ice). Another fact that reflects the intra-industry competition which has taken place locally as well as at the larger regional area levels is the shift in the share of assets held by the various types of savings and loan associations. The data in Table 45 indicate that, in terns of average assets, the federally chartered association in Sacramento has approximated the growth of the average federally chartered association in the United States. However, in terms of share of total insured assets held, the local state associations have increased from only 5.6 per cent of the total assets in 1950 to almost 86.0 per cent in I960. In contrast, at the national level the share of total assets held has remained relatively the same, and at the state (California) level the major portion of total assets held has shifted froa the federally chartered associations to the state insured associations. Assets. Liabilities, and Residential Construction Starts The different types of association charters and the managements under which the local associations operate have led to remarkable differences in the distribution of the assets and liabilities of the TABLE 45 TRENDS IN INTRA-INDUSTRY COMPETITION AS SHOWN BY SHARE OP TOTAL ASSETS AND AVERAGE ASSETS OF SAVINGS AND LOAN ASSOCIATIONS, UNITED STATES, CALIFORNIA, AND SACRAMENTO, 1950 AND 1960 Year United States Federal Charter Insured State California Federal Charter Insured State Sacraaento Federal insured Charter State 1950 1960 1950 1960 61.9% 57.4 $ 5.5 20.6 Share of Total Assets 38.1% 42.6 $ 3.8 13.0 60.0% 42.5 40.0% 57.5 Average Assets (In Millions of Dollars) $11.6 66.1 $ 7.1 36.1 94.4% 14.1 $ 3.3 19.1 5.6% 85.9 $ 0.4 18.1 Source: Derived froa Federal Hone Loan Bank Board Annual Conbined Financial Stateaents, 1950-1960; United States Savings and Loan 1961 Fact Book; Federal Hoae Loan Bank of San. Francisco, Roster of Meabers, annually 1950-1961; and Savings and Loan Coaaissioner, Annual Reports. 1950-1961 (Sacraaento: State Printing Office). 216 217 local Industry as wall as in the role that these institutions have played in the financing of local residential construction. The principal assets of the Sacramento savings and loan associations are their mortgage loans and other assets which basically represent their liquidity resources. The latter assets are their cash on hand and in banks and their holdings of United States government securities. Total mortgage loans increased by more than $102 million, from $3 million in 1950 to $105 million in 1960, as shown in Table 46. In per centage terms, the 1950-1960 total mortgage loan dollar increase represents a 3,330.0 per cent increase. In contrast, the liquidity resources of the Sacramento associations increased only 640.3 per cent, or in absolute dollars these assets grew slightly less than $15 million during the years between 1950 and 1960. Savings capital, the major liability item of the local associations, had a net gain of $102 million during the 1950-1960 period. Reserves and undivided profits, the second largest liability item on the books of the Sacramento associations, grew $7.5 million to a record high of more than $8 million. The major liability item, which was of specific interest for this study, was the third largest major TABLE 46 MAJOR ASSETS AND LIABILITIES OF INSURED SACRAMENTO SAVINGS ASSOCIATIONS, 1950-1960 (AMOUNTS IN THOUSANDS OF DOLLARS) Major Asset It— Major Liability Iteas Casb raui Advances Raaorvoa Total and U.S. and Othor and Mortgage Governaent Savings Borrowed Undivided Year Loans Obiications Accounts Moneva Profits 1950 $ 3,065 $ 2,329 $ 4,494 b $ 749 1951 3,294 1,162 4,501 $ 163 810 1952 3,614 2,127 4,776 45 856 1953 6,794 2,553 7,635 470 957 1954 15,516 3,775 16,253 642 1,330 1955 20,535 6,509 23,271 367 1,630 1956 25,321 6,610 29,082 200 2,010 1957 33,851 8,190 36,753 865 2,570 1958 51,837 10,381 51,714 2,565 3,774 1959 75,866 9,781 71,495 3,198 5,924 1960 105,131 17,241 106,211 4,885 8,199 1950-1960 Dollar Increase $102,066 $ 14,912 $101,717 $ 4,722 $ 7,450 1950-1960 Percentage Increase 3,330.0 640.3 2,263.4 2,896.9 994.7 aDollar increase and percentage increase are based on the 1951-160 period. *Not available. Source: Federal Hoae Loon Bank of San Francisco, Roster of Meabers. annually 1950-1961; Savings and Loan Coaaissioner, AnnuaT~Reports. 1950-1961 (Sacraaento: State Printing Office). liability identified as "Federal Hone Loan Bank advances and other borrowed money." In percentage terns this account increased 2,896.9 per cent during the study period, and the greatest portion of this growth occurred during the latter part of the 1950's. Relatively little use had been made of this resource during the first years of the study period; however, in order to meet the heavy demand for real estate credit during the latter half of the 1950's three local associations drew substantial advances from the FHLB. Of the three local associations which made use of the privilege of FHLB advances during the latter portion of the study period, one was a state chartered holding coapany operation. The other two were inde pendent local state chartered operations. The impact of the holding company operation was reflected in the data which indicated that it held 36.9 per cent of the total assets of all the associations in Sacramento in 1960, 40.6 per cent of the total savings and loan mortgages, and received 86.8 per cent of the FHLB o advances. This particular association financed 34.1 per cent of all the construction loans which the 2 State of California, Savings and Loan Com missioner, Sixty-Seventh Annual Report (Sacramento: State Printing Office, 1966), p. 97. 220 Sacramento savings and loans made in 1960. The test of that portion of the hypothesis which states that the increased rates of residential construction in an area are mainly dependent upon the ability and willingness of savings and loan associa tions to provide funds for real estate credit can be made on the basis of an examination of the data which relate the building permits issued and construction loans made by associations with the FHLB advances to these associations. Increases in the FHLB advances account reflect the associations' ability and willing ness to meet the demand for real estate credit. The data in Table 47 indicate that a sharp increase in the share of residential construction loans which the local associations provided occurred in 1957 and continued through I960. Beginning in 1957 local associations financed alaost 20.0 per cent of all residential construction locally and steadily increased their share to almost 33.0 per cent in 1960 (compared to 43 per cent at the statewide level). These same years were also years of greatly increased rates (over 100 per cent between 1957 and 1958) of residential construction, as well as the years in which local associations increased their borrowings from the FHLB (Table 46, page 218) in order to meet the need created 221 TABLE 47 SACRAMENTO SAVINGS AND LOAN ASSOCIATIONS' ROLE IN FINANCING LOCAL RESIDENTIAL CONSTRUCTION: 1951-1960 Year Building Permits Issued for Single-Family and Duplex Units Construction Loansa Percentage Number of Permits 1951 7,462 55 0.7 1952 3,184 31 0.9 1953 3,907 120 3.1 1954 6,281 377 6.0 1955 7,816 342 4.3 1956 5,482 623 11.3 1957 4,427 879 19.8 1958 8,439 1,887 22.2 1959 11,147 2,404 21.5 1960 7,689 2,526 32.8 a1951-1955, conventional financing only; 1956-1960, conventional, FHA, and VA financing. Source: Building permits— Sacramento City- County Chamber of Commerce, Pertinent Information about Metropolitan Sacramento. Sacramento County7 California (Sacramento: Chamber of dommerce, January, 1963); construction loans— unpublished data furnished by Sacramento savings and loan associations. 222 by the increased demand for real estate credit in the Sacramento Metropolitan Area. 111. THE STRUCTURE OF THE SACRAMENTO MORTGAGE MARKET The apparent concern' over the validity of Shaw's analysis of the California state licensed savings and loan market structures led to a new study and analysis by Grebler in 1963. The stated purposes of Grebler's study were (1) to examine the structure and performance of the California savings and mortgage markets, and (2) to investigate the structure and performance of the California savings and loan Industry which has come to play a major role in the two mar- 4 kets. A brief review of Shaw's analysis justified the apparent concern over its validity and the sound ness of its recommendations for public policy. For example, Shaw concluded that "there is restraint of ^Edward S. Shaw, Savinas and Loan Market Structure and Market Performance: A Studv~o? California Siate-Lfcensed Savings and Loani Associations (dacra- men to: State of tialiiornia Savings and Loan Commission, 1962). Lao Gr.bl.r and Eug.n. F. Brighaa, Savinna and Mortgage Markets in California: The Poaition anjj Performance of Che Savings and Loan Industry (Pasadena: California Savings and Loan League, 14&3), p. 1. 223 competition that detaches the industry from national capital markets and divides the state into local _5 market areas for savings capital and mortgages. This conclusion indicated a lack of awareness of the characteristics of real estate, the security for mortgages, which differentiate it from other economic goods. Land, which is considered to be part of the real estate, is different in degree from other economic goods in that it is durable, indestructible, and immobile. "Immobility produces a market which is g predominantly local." It is from this fixed location attribute that another differentiating characteristic develops, namely, heterogeneity. Therefore, each parcel of land, and hence the security for mortgages, is unique. Thus it is not the restraint of competition which divides the state into local market areas, it is the nature of the product. Shaw concluded, "selective merger in California is one step toward retrieving the benefits of competition, for homogeneous products [italics not in originalJ, between millions of buyers k Shaw, 0£. cit•, p. 1. g Preston Martin, Real Estate Principles and Practices (New York: The Macmillan Company, 196d), p. 3i. and sellers."7 As implied earlier, this conclusion was not necessarily a valid one. Additional evidence of the opinion that Shaw's study nay contain miscon- ceptions, defective analysis, unsupported generaliza tions and be open to serious question nay be found in the California Savings and Loan League report filed Q with the California Savings and Loan Coaaissioner. "The savings and loan league's report claias the basic error in the Shaw study is that it concerns itself exclusively with intra-industry competition and ignores a the expense of inter-industry competition.n Grebler's study supports the league's position. For exaaple, he concluded, "in the California aortgage markets [italics in originalj we found a sufficiently large nuaber of lenders in local areas as well as in the state-wide market to suggest the presence of a fairly high degree of competition."i0 The empirical data related to the Sacramento Metropolitan Area reported in the paragraphs which follow support Grebler's position. He concluded: 7 Shaw, op. cit.. p. 148. 9News item in the Wall Street Journal. Pacific Coast edition, February 1, p . 9. 9Ibid. 10Grebler and Brigham, op. cit.. p. 161. 225 . . • California savings and loan associations operate in a aortgage aarket which is relatively coapetitive by the standards of the general econoay, illustrated by the structure of other financial aarkets as well as the structure of nonfinaneial aarkets. However, the systea shows a slight degree of oligopoly resulting froa the restricted aoveaent of funds between geographic subaarkets, liaited entry, and heterogeneity [italics not in originalJ of product .1J- Institutional Lenders1 Heal Estate Credit Activities Institutions entrusted with the savings of individuals replaced private parties as the najor source of funds in the local aarket as well as other aortgage aarkets throughout the United States. Savings banks, coaaercial banks, savings and loan associations, aortgage coapanies, and life insurance coapanies are generally considered as being the so-called institu tional lenders. Life insurance coapanies have been active in both the priaary aortgage aarket, as origi nators of real estate loans, and the secondary aarket, where they buy real estate loans originated by aortgage coapanies. The Sacraaento County aortgage aarket has obtained funds froa all of the above aentioned insti tutions, with the exception of the savings banks, 1J- Ibid.. p. 162. 226 during the years between 1950 and 1960. Data avail able for the years 1955-1960 indicate that the total volume of trust deeds and mortgages recorded in Sacramento County ranged from 16,482 in 1957 to 28,389 in 1959. The dollar volume involved was alaost $172 million in 1957 and over $380 million in 1959, as reported in Table 48. Included in the data are all the trust deeds and mortgages recorded in the county. That is, all second trust deeds, loans by individuals and institu tions, as well as the single institutional loans made by lenders who do not re-enter the local aortgage aarket except after great periods of time have expired. Consequently, for analytical purposes, data which exclude the types of loans mentioned above may be more representative of the local aortgage aarket. Therefore, Table 49 was developed to report the selected recordings of institutional loans, excluding second trust deeds and individual loans by institu tions made on Sacramento County real estate during the period 1955-1960. Also, the selected institu tional recordings are reclassified into estimates of residential loans under $20,000 and nonresidential and residential loans over $20,000. The estimates are based on an inspection of the reported loan TABLE 48 (AMOUNTS IN MILLIONS OF DOLLARS; NUMBERS IN THOUSANDS) Total Institutional Miscellaneous Year Nuaber Aaount Nuaber Aaount Nuaber Aaount 1955 20.6 $214.9 12.7 $126.4 7.9 $ 88.5 1956 19.2 179.7 11.7 111.5 7.5 68.3 1957 16.5 171.8 9.2 118.4 7.3 53.5 1958 21.8 261.9 13.2 193.0 8.6 68.9 1959 28.4 380.2 16.3 242.1 12.1 138.2 1960 26.0 317.1 12.1 184.6 13.9 132.5 Source: Security Title Insurance Coenany. Bulletin of Loan Recordings, ■onthly 1955-1960. 227 TABLE 49 SELECTED RECORDINGS OF INSTITUTIONAL LOANS, EXCLUDING SECOND TRUST DEEDS AND INDIVIDUAL LOANS BY INSTITUTIONS, SACRAMENTO COUNTY, 1955-1960 (AMOUNTS IN MILLIONS OF DOLLARS) Year Selected Real Estate Loans Estimated Residential Loans Estimated Non- Res idential Loans Number Amount Number Amount Number Amount 1955 15,700 $179.4 15,500 $163.1 200 $16.3 1956 13,100 151.5 13,000 139.4 100 12.1 1957 9,300 117.9 9,200 108.7 100 9.2 1958 13,900 216.1 13,200 170.0 700 46.2 1959 18,300 302.6 16,600 244.6 1,700 58.0 1960 15,100 250.6 14,200 196.7 900 54.0 Source: Compiled fron data provided by Western Title Insurance and Guaranty Company. Loans by Institutions on Sacraaento County Real Estate, aonthly 1955-1960. 228 229 recordings for obvious nonresidential loans and a computation of average dollar amount per loan for all other recordings. Residential loans which average under $20,000 were classified as such; the remaining loans were then placed in the nonresidential and residential loans over $20,000 category. In order to better understand the institutional 12 structure of the local mortgage market and to approximate the amount of nonlocal funds attracted to the county, the data were classified by type of loan by institution, local or nonlocal. A local institu tion was defined as one which had a main or branch office in Sacramento County. Nonlocal institutions were defined as those which made real estate loans through agents located in the county without having either a main or branch office in Sacramento County. The data reported in Table 50 indicate that loanable funds have a high degree of mobility. Another useful quantitative measurement in the analysis of a local mortgage market are data which reflect the number and type of institutional partici pants who have entered the local market over a period 12 In this study "structureH refers to the pro portion of the real estate loans originated by the different types of lenders operating in the local mortgage market. TABLE 50 SELECTED SACRAMENTO COUNTY REAL ESTATE LOANS, BY LENDER, NUMBER, AND VOLUME, 1955-1960 (AMOUNTS IN THOUSANDS OF DOLLARS) Year Banks Savinas and Loan Associations Local Nonlocal Local Nonlocal Number Amount Number Amount Number Amount Number Amount 1955 9,215 $102,133 158 $ 7,196 1,680 $11,850 697 $ 5,555 1956 8,177 92,347 133 3,836 1,707 12,609 765 7,830 1957 3,643 47,706 152 2,137 2,229 18,106 335 2,967 1958 6,235 95,720 154 24,173 3,363 34,252 498 4,995 1959 7,323 151,268 249 5,426 3,778 41,283 1,129 13,973 1960 3,431 64,962 145 13,208 5,090 66,204 1,263 16,895 Selected Real Insurance Coananies Mortcane Companies Estate Loans Total Year Number Amount Nuniber Amount Number Amount 1955 958 $17,655 2,982 $34,953 15,690 $179,342 1956 859 15,597 1,499 19,302 13,120 151,520 1957 716 13,549 2,242 33,464 9,317 117,929 1958 739 15,524 2,911 41,477 13,900 216,141 1959 762 14,539 5,090 76,092 18,331 302,581 1960 707 21,584 4,486 68,652 15,122 250,614 Source: Compiled froa data provided by Western Title Insurance and Guaranty Coapany, Loans by Institutions on Sacramento County Real Estate, ■onthly 1955-1960 231 of tine. The available data, as reported in Table 51, indicate that at least 185 firms have originated real estate loans in Sacramento County during the years 1954-1960. Cnwp»Qial bank activity. One of the most volatile sectors of the local mortgage market has been that of local and nonlocal commercial banks. Residential loans under $20,000 made by local banks ranged from a low of 2,927 loans in 1960 to a high of 9,115 loans in 1955 (Table 52). In dollar amounts, the range was from $43 million in 1960 to $113 million in 1959. The activity of out-of-town banks was equally volatile. The range of nonlocal banks' number of residential loans under $20,000 was from a low of 102 in 1956 to a high of 190 in 1959. In dollar amounts the low was $905,000 in 1956 to slightly more than $2 million in 1959. Savings and loan activity. The local savings and loan associations, which are fewer in number and smaller in asset size than the local commercial banks, made fewer and smaller loans. However, their importance in the structure of the local mortgage market has grown steadily from a low of 1,680 loans in 1955 to almost 5,100 loans in 1960. This threefold growth in the TABLE 51 NUMBER AND TYPE OF INSTITUTIONS ORIGINATING REAL ESTATE LOANS IN SACRAMENTO COUNTY, 1954-1960 Coaaercial Banks Savings and Loan Associations Insurance Coaoanies Mortgage Coaoanies Local 10 8 1 1 Nonlocal 37 39 40 59 Source: Compiled froa data provided by Western Title Insurance and Guaranty Coapany, Loans by Institutions on Sacraaento County Real Estate. aonthly 1954-1960. 232 TABLE 52 COMMERCIAL BANKS, ESTIMATED REAL ESTATE LOAN ACTIVITY, BY TYPE, NUMBER, AND VOLUME, 1955-1960 (AMOUNTS IN THOUSANDS OP DOLLARS) Year Local Residential Under 120.000 fiusber Anount Nonresidential and Residential Over 120.000 Nnnber Anount N o n l o c a l _____ Nonresidential Residential Under <20.000 Nnnber Anount and Residential Over 620.000 Nnnber Anount 1955 9,115 $98,536 100 $ 3,597 150 $ 1,208 8 $ 5,988 1956 8,152 89,122 25 3,225 102 905 11 2,930 1957 3,620 45,975 23 1,732 138 1,188 14 949 1958 5,850 86,909 385 8,812 114 1,256 40 22,918 1959 6,076 112,622 1,247 38,646 190 2,323 59 3,103 1960 2,927 42,724 504 21,348 90 975 55 12,232 Source: Conpiled froa data provided by Western Title Insurance and Guaranty Coapany, Loans by Institutions on Sacraaento County Real Estate, nonthly 1955-1960. 233 niuiber of loans was more than equaled by the steady growth in the dollar volume of loans. That is, from less than $12 million in 1955 the dollar volume of loans grew steadily to almost $65 million in 1960. Savings and loan associations, which are specialists in home financing, do not offer the diversified department store type lending services available to commercial bank customers; consequently, they have shown comparatively little interest in nonresidential loans and residential loans over $20,000. Furthermore, legal conditions under which these institutions operated restricted the amount and type of loan. For example, the radius from the savings and loan associa tion main office within which loans may be made had been fifty miles. However, as the data in Table 53 indicate, the latter legal restriction did not prevent nonlocal savings and loan associations within the fifty mile radius, or whose charters exempt the legal restriction, from entering the local mortgage market. Life insurance and mortgage companies. Sacra mento was the home office location of only one life insurance company and one mortgage company during the study period. The available data were not categorized into the local and nonlocal classifications in order TABLE 53 SAVINGS AND LOAN ASSOCIATIONS ESTIMATED HEAL ESTATE LOAN ACTIVITY, BY TYPE, NUMBER, AND VOLUME, 1955-1960 (AMOUNT IN THOUSANDS OF DOLLARS) Year Local Nonlocal Residential Under $20,000 Nonresidential and Residential Over $20,000 Residential Under $20,000 Nonresidential and Residential Over $20,000 Nuaber Aaount Nuaber Aaount Nuaber Aaount Nuaber Aaount 1955 1,680 $11,850 0 $ o 697 $ 5,555 0 « 0 1956 1,707 12,850 0 0 759 7,695 6 135 1957 2,229 18,106 0 0 330 2,839 5 128 1958 3,276 29,355 87 4,897 493 4,844 5 151 1959 3,755 40,533 23 750 1,057 11,946 72 2,027 1960 5,027 64,644 63 1,560 1,241 J6,203 22 691 Source: Compiled froa data provided by Western Title Insurance and Guaranty Coapany, Loans by Institutions on Sacraaento County Real Estate. aonthly 1955-1960. 235 to preclude Identification of the nunber or dollar volume of loans made by either company. Nevertheless several interesting points may be noted in Table 54. First, the volume of residential loans under $20,000 originated by life insurance companies exclusively was in the narrow range of from 507 loans in 1960 to a high of 838 in 1955, and the trend in the number of loans has been downward. Also, there has been little change in the dollar volume— from a high of $11*7 million in 1955 to a low of $9 million in 1960. Second, mortgage companies have become increasingly active in originating loans locally. Their residential loans under $20,000 ranged in number from 1,494 in 1956 to 4,910 in 1959, and in volume from $19.1 million to $67.8 million, respectively. Thus, the increasing dependence of life insurance companies on the increas ingly important mortgage companies is reflected in the data reported in Table 54. Third, the data indicate that life insurance companies can invest in more diversified types of real estate loans, which may be located in geographically dispersed areas, than can either the commercial banks or the savings and loan associations. TABLE 54 LIFE INSURANCE AND MORTGAGE COMPANIES ESTIMATED REAL ESTATE LOANS IN SACRAMENTO BY TYPE, NUMBER, AND VOLUME, 1955-1960 (AMOUNTS IN THOUSANDS OF DOLLARS) Year Life InsuranceCoepanies Nonresidential Residential and Residential Under >20.000 Over S20.000 Nnnber Anount Nnnber Anount Mortgage Companies______ Nonresidential Residential Under *20.000 Nnnber Anount and Residential Over $20.000 Nunber Anount 1955 838 ♦11,743 120 ♦ 5,911 2,973 $34,196 9 ♦ 757 1956 789 9,901 70 5,696 1,494 19,141 5 161 1957 644 11,026 72 2,522 2,197 29,596 45 3,868 1958 624 9,123 115 6,402 2,825 38,479 86 2,998 1959 579 9,379 183 5,159 4,910 67,810 180 8,282 1960 507 9,074 200 12,510 4,432 63,035 54 5,617 Source: Conpiled fron data provided by Western Title Insurance and Guaranty Company. Loans by Institutions on Sacranento County Real Estate, nonthly 1955-1960. 237 238 Government Underwriting Activity The federal government also played an important role in the local mortgage market. The two most important agencies were the Federal Housing Administra tion and the Veterans Administration. The underwriting activities of these two agencies take the form of insurance on the part of the FHA and a guarantee of the loans which the VA has underwritten. This risk assumption activity on the part of these governmental agencies is an incentive for institutional lenders, mainly eastern insurance companies and savings banks, to invest in real estate loans. Generally these lenders enter the local mortgage market when the net yield on real estate loans is sufficiently competitive with alternative investments. The fact that, at times, the yield is not sufficiently attractive is reflected in the data which report the wide variations in government underwritten activity. Table 55 reports the estimated FHA applications received and new homes started in Sacramento County. The estimate was derived by the application of a 95 per cent * actor, which was recommended by local FHA officials, to the total eighteen-county territorial jurisdiction activity of the Sacramento FHA insuring office. The data do not reveal any significant pattern among the annual 239 TABLE 55 TREND OP FHA TITLE II SECTION 203 APPLICATIONS RECEIVED AND NEW HOMES STARTED IN THE TERRITORIAL JURISDICTION OF THE SACRAMENTO INSURING OFFICE, 1950-1960 Year Estiaated Sacraaento County New Construction 'Existing' Construction New Ekoaes i Started 1950 7,505 1,546 5,742 1951 3,958 1,653 4,436 1952 3,481 3,330 3,237 1953 4,396 3,677 4,181 1954 6,385 4,686 4,793 1955 6,865 7,116 6,447 1956 4,429 5,711 4,509 1957 3,187 2,468 2,747 1958 7,104 6,656 5,840 1959 7,829 5,703 7,189 1960 4,064 5,532 3,970 Source: Froa records furnished by the Sacraaento FHA office. 240 volume of applications or new homes started. Table 56 does not reveal any significant pattern either; however, it reports the estimated number and dollar volume of the Sacramento County cases actually insured by the Sacramento FHA office. It also was estimated by utilising the 95 per cent of total territorial volume factor recommended by the local FHA office. The Veterans Administration data are based on unpublished records furnished by the VA regional office in San Francisco. Table 57 reports, by purpose of loan, number and dollar volume of G1 loans in Sacramento County as of December 25, 1951, and the number and dollar volume of GI loans underwritten each year from 1952 until I960. Similarly, Table 58 reports the number and dollar volume of Sacramento County GI loans by type of lender. While neither of these two tables reveals any significant patterns, they do reflect the attitude of the institutional lenders who were in and out of the local mortgage market. IV. EFFECT OF FEDERAL RESERVE BOARD MONETARY POLICY ACTIONS ON LOCAL LENDERS Lender Survey: Purpose and Methodology The purpose of the lender survey was to examine Year 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 TABLE 56 ESTIMATED FHA INSURING OPERATIONS SACRAMENTO COUNTY, 1951-1960 (AMOUNTS IN MILLIONS OF DOLLARS) Nwr (instruction Existing Construction Total Casos Anount l/nitsa Casas Isoont Units* Casas Anonni Units* 2,889 $22.3 2,998 1,102 8 8.4 1,123 3,982 8 30.7 4,150 1,721 15.7 2,003 1,848 15.3 1,911 3,567 31.0 3,914 2,282 19.9 2,350 2,291 19.6 2,344 4,573 39.7 4,694 3,503 31.6 3,538 2,938 27.6 2,991 6,441 59.1 6,585 6,862 68.6 6,905 6,647 64.8 6,754 13,510 133.4 13,691 4,279 44.8 4,309 5,201 55.3 5,441 9,481 100.1 9,751 3,008 34.2 3,011 2,424 28.6 2,458 5,431 62.8 5,468 5,380 66.1 5,402 5,565 70.3 5,626 10,943 136.4 11,028 4,700 63.3 4,726 4,297 53.8 4,367 8,998 117.1 9,093 6,267 85.0 6,308 5,173 72.8 5,288 11,440 157.7 11,596 ^Includes duplexes. Source: Froa records furnished by the Sacraaento FHA office. (0 * ■ H TABLE 57 SCHEDULE OF "GI" LOANS MADE IN SACRAMENTO COUNTY BY PURPOSE OF LOAN, 1951-1960 (AMOUNTS IN THOUSANDS OF DOLLARS) As of Deceaber 25. 1951 Putposs_________________________________________________Nunber Aaount Hobs: Existing Structure (1) Proposed Construction (2) New Existing Structure (3) Alterations, Iaproveaents, or Others (5) Repairs (4) 2,307 3,193 6,260 2 73 * 14,563 24,870 25,265 14 88 Fara (6) Business (7) 1952 1953 25 181 130 599 1954 Purpose Nuaber Anount Nuaber Aaount Nuaber Aaount (1) 9 $ 95 6 I 65 i 65 | 687 (2) 1,510 115,265 457 A,867 1 15 (3) 37 355 590 6i,377 1,677 18,368 (4) 0 0 0 0 0 0 (5) 0 0 1 2 0 0 (6) 0 0 0 0 0 0 (7) 36 77 11 (Continued) 82 5 49 242 TABLE 57 (CONTINUED) 1955 1956 1957 Pumose Nuaber Aaount Nuaber Aaount Nuaber Aaount (1) 263 S 2,789 126 $ 1,528 55 * 644 (2) lla 195 39 518 7 105 (3) 4,263 49,563 2,667 33,248 2,212 29,561 (4) 0 0 0 0 ' 0 0 (5) oa oa 0 0 0 0 (6) 0 0 0 0 0 0 (7) 8 66 1958 4 26 1959 34 436 1960 Pumose Nuaber Aaount Nuaber Aaount Nuaber Aaount (1) 6 82 23 290 34 436 (2) 0 0 0 0 0 0 (3) 816 11,994 1,572 23,941 1,191 18,620 (4) 0 0 0 0 0 0 (5) 0 0 0 0 0 0 (6) 0 0 0 0 0 0 (7) 10 90 3 33 3 25 aIncoaplete or not availablo because of revision of data. Source: Based on unpublished records furnished by the VA Regional Office in San Francisco. 243 TABLE 58 SCHEDULE OF "GI" LOANS MADE IN SACRAMENTO COUNTY BY TYPE OF LENDER, 1951-1960 (AMOUNTS IN THOUSANDS OF DOLLARS) As of Decenber 25. 1951 Lender Nunber Anount Mutual Savings Banks (1) 1 $ 7 Coaaercial Banks, including National Trust Companies (2) Banks and 8,372 37,952 Building and Loan or Savings and Loan Associations (3) 47 333 Insurance Companies (4) 1,919 17,411 Titlo, Mortgage, and Real Corporations (5) Estate Firns, Companies, or 1,702 9,826 Suppliers (6) 0 0 1952 1953 1954 Lender Nunber Anount Nunber Anount Nunber Anount (1) o « 0 0 S 0 0 S 0 (2) 68 339 23 195 61 737 (3) 15 145 17 154 80 780 (4) 521 5,732 70 882 78 1,026 (5) 997 9,575 955 10,161 1,529 16,576 (6) 0 0 0 (Continued) 0 0 0 TABLE 58 (CONTINUED) Lender 1955 1956 1957 Nunber Anount Nunber Anount Nunber Anount (1) 0 s o 0 $ 0 0 * 0 (2) 413 5,214 633 7,299 253 3,144 (3) 95 967 60 703 30 360 (4) 237 2,931 127 1,788 73 1,009 (5) 3,798 43,481 2,019 25,530 1,929 25,894 (6) 2 20 0 0 0 0 1958 1959 1960 Lender Nunber Anount Nunber Anount Nunber Anount (1) 0 $ o 0 S 0 0 $ 0 (2) 25 325 5 66 4 42 (3) 38 684 122 1,847 112 1,787 (4) 3 40 2 27 0 0 (5) 766 11,117 1,469 22,323 1,112 17,269 (6) 0 0 0 0 0 0 Source: Based on records furnished by the VA Regional Office in San Francisco. 245 246 the attitudes, opinions, and policies of mortgage lenders with respect to the effect of changes in the Federal Reserve Board's monetary policy on the financ ing of new residential construction. In order to gauge these effects, a lender questionnaire was pre pared and tested. Personal interviews utilizing the pretested questionnaire were used. All of the 13 lenders in the survey were asked eight questions 13 For an excellent discussion of the borrowers' attitudes see Jack M. Guttentag, "Some Studies of the Post-World War II Residential Construction and Mortgage Markets" (unpublished Ph.D. dissertation, Columbia University, New York, 1958), pp. 425-448. This dis sertation, which is a collection of essays written by the author for the Monthly Review of the Federal Reserve Bank of New York,presents a survey of the evidence which suggests that borrowers are responsive to changes in credit terns. He states that "evidence of various types suggests that the demand for mortgage credit was responsive to the modest changes in mortgage terns (particularly to changes in down payments) that occurred during the period covered by this study" (p. 426). It was concluded that "the interest rate elasticity of mortgage borrowers is appreciable, and certainly it is higher than for business borrowers. The point is that because of the various restrictions and constraints on the magnitude of mortgage interest rate changes, such changes are not adequate to equili brate the market when the balance of demand and supply changes; hence, changes in credit availability, of which the down payment is the most important dimension, are generally required to complete the adjustment" (p. 448). He also pointed out that a limited discussion of the problem of the elasticity of consumer demand for housing and mortgage credit will be found in the follow ing: E. M. Fisher, Urban Real Estate Markets: Character istics and Financing Mow York: National bureau of Economic Research, 1951), pp. 83-88; Leo Grebler, The Role of Federal Credit Aids in Residential Construction (New York: National bureau oT~Economic Research, 1953), 247 which were related to their own lending policies and procedures. The rest of the questions were asked only of the bank and savings and loan respondents since these two types of institutions are more directly affected by the actions of the Federal Reserve Board. Of the 107 names of local individuals and institutions listed under the "Real Estate Loans" heading in the December 1961 Sacramento Telephone Directory, only forty-five were institutional-type lenders. These forty-five institutional-type lenders were contacted in October 1962, and usable interviews were obtained froa this entire institutional lender population. The distribution of the usable interviews was as follows: eight commercial banks (two other local commercial banks had not been active in real estate lending activities during the study period), eight savings and loan associations, seven life insurance companies, and twenty-two mortgage companies. It nay be concluded, therefore, that the total 13 (continued) p. 25; Housing and Home Finance Agency, "Home Financing 1949-51— Changes Under Credit Controls," Housing Research. Winter, 1951-1952; Housing and Home Finance Agency, "New Hone Price Shifts, 1951- 1952: Under Credit Controls as Amended in 1951," Housing Research. March, 1953, p. 425. 248 institutional-type lender population responded to the questionnaire. A review of recent deed recording lists and the data reported in Table 52, page 233, led to the conclusion that reliable, representative interviews were obtained from the total institutional lender population which had been active in the field of real estate lending during the study period. Anticipated Impact of Monetary Policy Action: A General Summary In general, monetary policy changes are insti tuted to effect anticipated changes in the supply, availability, and cost of credit and money. For pur poses of this study it was thought that answers to several general questions related to monetary policy changes might be obtained through the use of a questionnaire survey of local lenders. These general questions, upon which the several series of specific questions are based, are previewed here in the approximate order of the reporting of the specific questions and responses in the latter part of this chapter. The first general question, related to years of restrictive lending policies, is, "Does a restric tive monetary policy cause a decided shift in invest- 249 ment policy on the part of local lenders?” Stated in other words, "Do tight monqy conditions lead to changes in types of assets held by local lenders? Will long-term investments give way to more liquid or near-money types of investments and more selective mortgage lending policies during periods of restric tive monetary policy?" A general question involving interest and dividend rates paid to savers is, "Can monetary policy encourage savings?" That is, "Will savers' propensity to save be stimulated by allowing higher rates of return to be offered by passbook type savings institutions?" Obviously, saving and dissaving affect the flow of funds and the availability of credit at these financial institutions. Other major sources of funds available to the above institutions are mortgage resales, open market borrowing, and for savings and loan associations advances from the Federal Home Loan Bank. Another general question for which answers were elicited is, "What is the impact of restrictive monetary policy actions on the sale of mortgages, open market borrowings, and advances from the FHLB?" Finally, a general question related to lending activities is, "Do monetary restraint policies exer- 250 else a restrictive influence upon mortgage lending activities?" It was thought that evidence of a restrictive influence might be found in the applica tion of more rigorous lending standards, higher interest rates and fees charged for loans, more restrictive lending terms, and less use of advance commitments. The responses reported in the sections which follow-provide some insights into this and the other general questions posed earlier. Summary of Responses One of the questions asked of all respondents was, "Has there been a tendency on the part of your company to restrict loans generally during the last six months to a year?" This question was followed by a request for an explanation for their actions. The responses of the forty-five lenders are reported in Tables 59 and 60. More than two-thirds of the respondents indicated that they had not generally restricted loans recently. More interesting than this over-all statistic, however, is the correlation of the category of lender with the type of action. That is, almost half the number of respondents who reported a tendency to restrict loans generally in the last twelve month period were savings and loan TABLE 59 DISTRIBUTION OF RESPONSES, BY TYPE OF LENDER, TO THE QUESTION, "HAS THERE BEEN A TENDENCY ON THE PART OF YOUR COMPANY TO RESTRICT LOANS GENERALLY DURING THE LAST SIX MONTHS TO A YEAR?” Response Savings and Loans Banks Mortgage Companies Life Insurance Companies Total Yes 6 1 6 - 13 No 2 7 16 7 32 Source: Lender questionnaire responses. TABLE 60 SEASONS GIVEN FOR RESTRICTIVE OR NONRESTRICTIVE POLICY, DURING THE PAST TWELVE MONTHS, BY TYPE OF LENDER AND NUMBER OF TIMES MENTIONED ____ Policy_____ Non- Restrict restrict Reason Nnnber of Tines Mentioned b: Savings and Loans Banka Mortgage Conpanies Insurance Conpanies X X X X X Overbuilt Areas Lover Denand Trend Restrictions upon Speculative Loans Recent Bad Investnents High Foreclosure Rate X Abundant Money Supply X Desire to Increase FHA- VA Portfolio X Desire to Increase Conven tional Loan Portfolio X Conpetition for Mortgage Loans X Want All Business Can Get X Absence of Regulations upon Coapetitors X More Applications, More Acceptances X Denand Is Still Here, We Must Stinulate It X Want to Expand Connercial and Industrial Loans 3 1 3 1 2 5 2 6 5 1 Source: Lender questionnaire responses. 253 associations. More specifically, 75 per cent of the savings and loan associations responded that they tended to restrict loans during the last twelve months. In contract to the savings and loan affirmative replies, only 12.5 per cent of the banks, 27.3 per cent of the mortgage companies, and none of the life insurance companies reported that there was a tendency on their part to restrict loans generally during the past twelve months. The explanations given for the lenders' actions have been summarized in Table 60. In general, the responses indicate that the lenders have not followed a policy of restrictive action because of the abundant supply of investment funds. The underlying reasons stated for adopting a restrictive policy involved basic supply of and demand for homes considerations, foreclosures, and recent investment experiences. The reasons motivating the lending institutions to engage in a restrictive lending policy include: (1) the number of completed unsold houses in certain areas, (2) the apparent weakening of demand for new homes, (3) restrictions which the lenders were placing on "spec builders," (4) the increased number of fore closures, and (5) investments which had become some what less desirable chan when they were made. The 254 data should not, however, be Interpreted to lead to the conclusion that these institutions have stopped ■aking real estate loans. The foregoing statements, considered with the increasing number of deeds and loans recorded, clearly indicate a policy that emphasizes more selective lending. Additional evidence of continued lending by banks and savings and loan associations, and an indication of the real estate loans which have become somewhat less desirable, may be seen in Table 61. The summary data reported in this table partially reflect the result of general overbuilding. In Sacramento County, as well as in certain other California areas, " there has been a significant pause in new construction to permit demand to offset the supply.The data indicate that defaults are beginning to develop on real estate loans, "most of which originated as loans in 1960 and early in 1961 15 and are not the result of current lending activity." The reference cited above was not meant to suggest that overbuilding and mortgage defaults are 14 Great Western Financial Corporation, Annual Report. 1962 (Los Angeles: The Corporation, l9S3j»~ p. 7. 15Ibid. 255 TABLE 61 NUMBER OF LOANS AND DEFAULTS IN SACRAMENTO COUNTY, BY SELECTED LENDERS, AND SELECTED DATES First Tsn Months 1961 of 1962 Lenders___________Loans Defaults Loans Defaults Savings and Loans 4,093 735 2,869 791 Banks 3,723 148 4,425 60 Source: Loans— The Data Sheet, California Tax Agency; defaults— suaunary of extractions froa public records for a private unpublished report. 256 exclusively California problems. A recent issue of | g a national publication contained repeated references and warnings about the deterioration in mortgage credit standards throughout the United States. The statistical evidence suggests that these warnings should not go unheeded. For example, during 1962 there were 86,440 foreclosures in the United States, which represents an 18 per cent increase over the 17 73,074 foreclosures in 1961. An authoritative source has indicated: The rate of foreclosures also rose during 1962, the rate being 4.1 per 1,000 home mortgages outstanding compared with 3.6 per 1,000 in 1961. While the 1962 rate was the highest recorded in the 13 years for which such rates have been calculated, it was undoubtedly far below the level of rates prevailing in the period from 1928 through 1938, when foreclosures ranged between 116,000 and 252,000 per year with a much smaller inventory of mortgaged homes.^ 16 The Mortgage Banker. March 1963 issue, con tains the followingarticles: "Concerning the Quality of Mortgage Credit" by Leo GTebler; "Two Projections: For Investment Funds and for Mortgage Banking" by Saul B. Klaman; "Who Will Want to Borrow and How Much Will They Want" by Roger F. Murray; and "FHA Experience with Mortgage Foreclosures and Property Acquisitions." 17 Housing and Home Finance Agency, Sixteenth Annual Report. 1962 (Washington: Government Printing OTHce, 19537, pTT2. 18Ibid. 257 A factual study of other data by another authority led to the conclusion that "the pressures on financial institutions to employ their funds in high-yield investments such as mortgages are probably 19 greater than at any time covered by the statistics.” He gave the following reason for this: The volume of savings placed in the institu tions is very large, actual returns to savers and returns expected by savers are higher, and outlets for investment in corporate securities are limited by the slackened pace -of expansion in industrial plant and equipment. Therefore: A high rate of mortgage lending appears to offer a way out of this dilemma, and the compe tition for mortgage loans has become so fierce that lending standards are in danger of being further compromised. There is real question whether such a large volume of lending activity can continue without increasing deterioration of the quality of mortgage credit, more severe and costly loan servicing problems, and still higher rates of defaults and foreclosures.21 Another relevant study concluded: The number of mortgage loans passing through the three stages of delinquency, default and foreclosure has risen at an increasing rate. However, the number of such loans in relation to the total outstanding still is small, in 19 Grebler, loc. cit., p. 43. 20Ibid. 21Ibid. 258 fact, almost insignificant. The concern over the quality of credit stems from the fact that the number of loans in difficulty is rising steadily.22 Many of the increased defaults and foreclosures on FHA-insured loans have "been traced, in part, to the relatively easy terms granted during the 1958-1959 23 housing boom." Also, "the FHA study shows that foreclosures and property acquisitions prior to 1958 were abnormally low and, thus, some increase could 24 have been expected." The widespread public discussion of the Increase in mortgage foreclosures has led the United States Savings and Loan League to be "concerned about the confusion in public utterances between liberalization of credit and deterioration of lending 25 standards." A thorough review of the savings and loan business resulted in the league's position that hone mortgage lending standards of credit have been 22 Oscar F. Litterer, "New Developments in the Residential Mortgage Market," Federal Reserve Bank of Minneapolis Monthly Review« November, T5557™P*™77™* 23Ibid., p. 8. 24Ibid., p. 10. 25 United States Savings and Loan League, Board of Directors, League Statement of Policy. Adopted at the League's 71st Annual (JonveniTon (San Francisco: The League, November 7, 1963). 259 maintained in a highly satisfactory manner and that "today's applicants for horns loans CarsJ comparable 26 as personal risks to those of the 1940's and 1950's.H Lenders' reported years of most restraint. It is frequently stated that certain financial inter mediaries are not directly affected by the monetary policy actions of the Federal Reserve Board. This, it is further stated, allows individual institutions to pursue their own interests and policies, which may be contrary to the over-all objectives of the monetary authorities. Therefore, an attempt was made to relate various financial institutions' local real estate lending activities with changes in the Federal Reserve Board's policies regarding general credit conditions of ease or restraint. Open-end questions were utilized in the attempt to examine the causal relationship, if any, of these two variables. One of the many questions asked of all the lenders was, "In what years have you 27 been most strict?" This question was followed by the request for an explanation as to why this action 26Ibid. 27 In this study the word "strict" is used synonymously with the term "exercise a greater degree of restraint." 260 was necessary. Their responses are tabulated in Tables 62 and 63. It is significant to note that 17.8 per cent of the lenders reported that at no tine during the 1935-1962 period had they been as strict as at present in processing.real estate loans. This is important 28 because two recessions, as identified by the National Bureau of Economic Research, occurred during this tine period. These recessionary periods were followed by periods of expansion. Also, throughout this tine period the Federal Reserve had pursued a shifting course of varying degrees of easy noney and "ease" to tight noney and "restraint" conditions. This would inply that during tight noney periods lenders would be more strict than during easy money periods. This hypothesis, however, was not conclusively supported by the data. Almost 18 per cent of the respondents indicated that they had not been more strict in any one of the years considered. The detailed analysis of the types of lenders who were not affected during the varying periods of 28 The National Bureau of Economic Research has identified the two recessionary periods as being July 1957 through April 1958 and May 1960 through February 1961. TABLE 62 DISTRIBUTION OF RESPONSES, BY TYPE OF LENDER, TO THE QUESTION, "IN WHAT YEARS HAVE YOU BEEN MOST STRICT?" Single Year Double Year NBER Recession Periods Nuaber of Tines Mentioned by Mortgage Connanies Life Insurance Coananies Savings and Loans Banks Total 1955 1 1 '55-'56 1956 1 1 *56-'57 1 1 1957 1 1 1 3 P-July '57 * 57-158 T-Apr. '58 1 1 2 1958 4 4 158-* 59 1959 2 1 1 4 •59-*60 2 1 1 4 1960 1 3 1 5 P-May '60 '60-'61 T-Feb. *61 1 2 3 1961 1 2 3 «61-'62 3 1 4 1962 1 4 1 6 None of the Years 3 2 1 ( 2 8 “Not includod in the tabulations are the four responses: two, 1960-1962 period; one, pre-1960 period; and one, 1955-1957 period. P - Peak T - Trough Source: Lender questionnaire responses. TABLE 63 DISTRIBUTION OF RESPONSES, BY TYPE OF LENDER, EXPLAINING THE STRICT REAL ESTATE LENDING YEARS OR THE LACK OF STRICT YEARS Tyne of Action Exnlanation Number of Times Mentioned bv Restric tive Non- - Restric tive Mortgage Companies Life Insurance Companies Savings and Loans Banks X Tight Money 6 6 — 4 X General Overbuilding 3 - 7 1 X General Business Conditions 3 - - 1 X Lack of Qualified Borrowers 1 - 1 - X Vacancy Factors - 1 - - X Greater Denand 1 I 1 - X High Level of Eaploynent 1 - - - X Lack of Competition 1 Source: Lender questionnaire responses. 262 263 easy or tight money is even more revealing. The data in Table 62 indicate that 25.0 per cent of the respond ing banks did not report any years when they exercised a greater degree of restraint on their real estate loans. That is to say, only three-fourths of the banks, which are the key institutions in the money market and the institutions which are most directly bound by Federal Reserve Board regulations and policies, reported that there were certain years during the period when they were more strict. Only the life insurance company respondents reported relatively less restraint than the bank respondents. Almost 29.0 per cent of the life insurance company respondents reported that there were no years when they were more strict in their real estate loans. However, less than 14 per cent of the mortgage company respondents reported that they had not had more strict years. It might be well to point out that the mortgage companies originate most of their loans for resale to life insurance companies, and in this fact might lie the explanation as to why life insurance company respond ents were less affected by periods of restraint. That is, it is possible that the life insurance companies continued to originate their own volume of real estate loans throughout all of the study period. However, 264 in periods of ample money, or when real estate loans were competitive with other investments, the life insurance companies resorted to mortgage company originations. This procedure has been found to be more efficient than expanding their own staffs during periods of ample money and then reducing them in periods of tight money. In this manner it is possible for the life insurance companies' staffs to produce a stable volume of real estate loans without being directly influenced by the Federal Reserve Board's actions. Only one of the savings and loan respondents replied "none" to the question, "In what years have you been most strict?" In other words, 87.5 per cent of these responding financial intermediaries who were engaged in local real estate lending activities *replied in the affirmative to this question. Only 75.0 per cent of the bank respondents, 71.4 per cent of the life insurance company respondents, and 86.4 per cent of the mortgage company respondents gave similar responses to this question. The mortgage company respondents' reasons for not having any restrictions centered around greater demands for loans and high levels of employment. The responding mortgage companies generally had restricted loans in most of the years during the 1955-1962 period. The heaviest concentration, however, occurred during the contraction stages of general business activity, which was more commonly known as the 1957-1958 business recession. A condition of general credit ease or easy money prevailed, yet these respondents reported a more strict real estate lending policy during this period. On the other hand, two mortgage company respondents reported strict lending policies in 1959 and two reported similar action in the 1959-1960 period, which was the expansion or upswing period of general business activity prior to the peak in May of 1960. A larger number of them reported strict policies since the February 1961 trough in general business activity during which a plentiful supply of funds was available for real estate loans. The explanations offered for the fore going actions centered on the general overbuilt situation and the tight money condition in the earlier periods. Life insurance company respondents indicated that their most strict real estate lending years occurred during the upswings of general business conditions, thereby reflecting the competition from 266 alternative investments. Their heaviest concentration occurred during the build-up in 1959 to the May 1960 peak of general business activity. Fewer life insur ance company respondents reported similar action during the 1957-1958 recession. It might be implied from their responses that their actions to restrict loans were consistent with Federal Reserve policy objectives. Savings and loan association respondents reported that not until 1961 did they experience restrictive loan policies. Two respondents reported restriction in 1961, and five reported restrictions since then. Several explanations for this phenomenon have been extended. One reason for the lack of restriction in the earlier years was that the number of these financial intermediaries in Sacramento has been small and their numerical growth has been rela tively slow. Another reason is that these institutions are specialists in home financing and as such maintain the largest portion of their assets in home loans. The explanation for the current restrictions lies in the general overbuilt situation and the lack of qualified borrowers, as indicated in Table 63, page 262. Commercial bank respondents indicated that they had restricted real estate loans during most of the 267 years in the study period. Their pattern of more strict years, however, approximates that of the life insurance companies. That is, restrictions occur during the upswing stages of the business cycle and the heaviest concentration occurred during the build up to the May 1960 peak in general business activity. Thus, the responsive nature of commercial banks to Federal Reserve Board monetary policy is reaffirmed. Questions Asked Only of Savings and Loan and Bank Respondents All the responding lenders were asked the first eight questions in the questionnaire. These questions were considered to be general in nature and applicable to all the institutional real estate lenders. Those questions which followed the first eight were con sidered to be applicable only to savings and loan associations or commercial bank operations as related to the monetary policy actions of the Federal Reserve Bank. Therefore, mortgage company and life insurance company respondents were asked the first eight ques tions only. The balance of the questions were equally divided as related to savings and loan operations or bank operations and were asked of those respective institutions only. The findings of these specific-type 268 questions are reported separately in the two sections which follow. Savings and Loan Respondents1 Replies Inter- and intra-industry price and nonprice competition. The savings and loan respondents were asked, "What was the major reason for the increase in the dividend rate recently?" Their replies to the question were unanimous: the underlying reason was competition. All of the association respondents mentioned competition in one form or other. More specifically, three simply reported "followed competi tors' actions,” one responded "followed large competi tor," one stated "competition between savings and loans," and another feared the "competition from southern California savings and loan associations." Two of these respondents indicated the basic need for additional funds to meet the increased demand for real estate loans, thereby implying that in order to attract more savings the incentive to save would have to be made greater. In order to determine the relative degree of competition from local banks and nonlocal savings and loan associations, the association respondents were 269 asked two related questions. First, "What differential in the rate paid to savers is necessary for you to attract savings away from local commercial banks?" The reply to this question which was received the greatest number of times was that 1 per cent more than that offered by local banks was necessary. Two respondents felt that their associations could attract savings away from banks with a one-half of 1 per cent differential only. One respondent indicated that a 1.5 per cent difference between savings and loan rates and bank rates was required to attract savings to the savings and loans. Two respondents did not reply to this question, and one indicated that the present rates were adequate. Along with the question relating to bank competition, the respondents were asked to identify what services would be necessary in order to compete successfully with the banks. Four respondents identified nonprice competition factors such as loca tion and convenience, better quality services and reputation, safety of accounts, and public confidence and sise of association. Two respondents felt that in order to compete successfully it would be necessary to offer those services which their competitors offered. Another simply stated "what the law allows." The second question related to competition was, ''Compared to other savings and loans, what differential in the rate paid to savers is necessary for your association to attract savings out of other areas of the United States?" The responses to this question were not indicative of a general agreement among the respondents other than the necessity that some differential is needed. That is, except for the two respondents who indicated that "any differ ential would be adequate," each of the other respond ents simply stated the following: one-half per cent; one-eighth to one-quarter of 1 per cent; at least 1 per cent; pay the going rate; none, we use only our own money; and all follow the moves of the San Francisco and Los Angeles money centers. From the responses indicated above it might be concluded that local savings and loan associations (1) generally assume a policy of "follow the industry leader" for establishing the price that they will pay for savings, (2) need a 1 per cent differential in the rate paid to savers in order to attract savings away from local banks, and (3) need a rate which is at least one-eighth of 1 per cent more than that paid by savings and loans in geographic areas other than Sacramento in order to attract savings from those 271 other areas. Sources of savings increases. Public sources have published reports which indicate a general rise in the gross savings receipts of financial institu tions. On the assumption that the local savings and loan associations have also experienced a general rise in their savings, the local respondents were asked a series of opinion questions related to their savings increases. The first question in this series was, "Between new accounts opened and increases in existing accounts, what was the relative importance of each in contributing to gross savings receipts?" Seventy-five per cent of the savings and loan respondents replied that new accounts were a more important factor than existing account expansions in their associations' increases in gross savings receipts. Three of them replied simply "new accounts." Two others reported that new accounts were responsible for 80 per cent of the increase in gross savings. One other respondent indicated 75 per cent; whereas only one felt that existing accounts and new accounts contributed equally, and one of the respondents did not reply to this question. Where did these new accounts originate? The respondents were specifically asked, "What per- centage of new accounts came from the Sacramento area?" The percentages reported ranged from 75 to 99 per cent. Two respondents reported that 95 per cent of their respective associations new accounts cane from the Sacramento area, two others reported 75 per cent each, and the responses 99, 98, 90 per cent and "impossible to determine" were recorded once each. The respondents were then asked, "What were the major sources of new local accounts?" This was asked in order to distinguish between individual savers and institutional savers. The responses indicated that the individual saver was responsible for the growth in the local savings and loan associa tions' savings accounts. These new accounts were the result of the increase in the local population. The individual savers selected their specific associa tion on the basis of customer referral, newspaper and radio advertising, and the reputation of the savings and loan, and some were "walk-ins." The responses were supported to a great extent by the replies to another question, "What percentage of outside new accounts came from broker noney?" The replies received in response to this question ranged from none (three respondents) to 4.8 per cent (one respondent). Two respondents failed to reply to the question. One other 273 stated less than 1 per cent, and the eighth reported that his association had reached its broker noney quota (5.0 per cent) early in 1961. Comnercial bank price competition. In order to examine the effect of price competition in the form of the interest rate increase by the commercial bank, the savings and loan respondents were asked, "What were the major reasons for withdrawals?M Their replies indicated that the increase in the interest rate on savings accounts by the banks to 4.0 per cent had little effect on the local savings and loan asso ciations. In fact* while they frequently mentioned investment purposes as the reason for withdrawals from their associations, none of the savings and loan respondents indicated that the investment might have been a commercial bank savings account. Their responses to this question included the following: "for investment purposes; stock market investments, taxes, and emergencies; stock market and consumer spending— used to think that the stock market was the major reason, however, there wasn't much going into the stock market in July— must be consumer buying of personal goods; don't know— don't ask; and no withdrawals." Only two respondents failed to 274 reply to this question. Sale of mortgages as a source of funds. In addition to net savings inflows and the mortgage repayments on the amortized real estate loans mortgage lenders can obtain funds through the sale of mortgages to other institutional investors. The respondents were questioned regarding this aspect in order to evaluate the relative importance of the sale of mortgages in the over-all operation of obtaining funds to meet the demand for credit in a dynamic capital-deficit urban area. The question asked was, "Have you ever sold mortgages?n Seventy-five per cent of the savings and loan respondents reported that they had not sold mortgages. Their responses indicated that the majority of Sacramento savings and loans have been able to meet the demands placed on them for mortgage funds through means other than the sale of packages of mortgages. Both of the association respondents who indicated that they had sold mortgage packages reported that those sales were made during I960. Neither of these two respondents replied in the affirmative to the subsequent question, "Did you curtail sales during 1955-1957 or 1960-1961 due to the drop in market price?1 * It should be noted that 275 (1) the National Bureau of Economic Research designated peak in business activity occurred in May 1960, (2) the first reported years of strict real estate lending activity by savings and loans was 1961, (3) only two savings and loans reported ever having restricted real estate lending activity (Table 61, page 255), (4) five other associations reported restricting real estate loans during either 1961-1962 or 1962 only, and (5) all seven of the associations which reported restric tions did so because of the general overbuilding conditions rather than because of tight noney situa tions. Bond portfolio. The respondents were asked, nDid the dollar amount of your bond portfolio increase during 1955-1957 or 1960-1961?" This question was followed by a request for an explanation of the increases, if any. The reason for these two questions was to explore other investment activities of the associations and their reason for investing in bonds during those periods. The responses to these ques tions indicated that two of the four local associations in existence during 1955-1957, and four of the eight operating in the 1960-1961 period, did expand their bond portfolios during these years. The reason 276 mentioned most frequently for this action was to maintain the liquidity requirement ratios. Two respondents did not reply to this question. One other simply stated that his association had no need to expand its bond portfolio. The eighth respondent replied negatively to the bond portfolio expansion question. His explanation was that they could not profitably employ their funds in investments other them real estate loans. The bond-investing respondents were asked, "What was the maturity of the bonds purchased?" This was done in order to relate their stated reasons to their investment actions. Their responses did support their actions; all of then mentioned the short-term government bonds of four years or less. Specifically, one each reported investments of less them one year, one month to seven years, three to four years, four years maximum, and one to four years and some twenty year municipals. Open market borrowing. Another source of funds available to the savings and loan associations is to borrow funds in the open market. The association respondents, therefore, were asked the question, "Did you borrow in the open market?" Two respondents did 277 not reply to this question, and the rest of the responses were equally divided. There were three "yes" responses and three "no" responses. Two of the three ope 3 market borrowers did so in order to maintain their liquidity ratios, and the third did so in order to meet loan demands. On the other hand, two of the three association respondents who did not resort to the open market reported that it was not necessary for them to do so since their association had ample money available. The third "no" respondent reported this was not done "because savings and loan borrowing in the open market is restricted in California.n Federal Home Loan Bank advances. Recently 29 certain monetary policy theorists have expressed concern over the availability of funds which savings and loan associations have as members of the Federal Home Loan Bank. In addition to the several sources of funds mentioned earlier, association members of the 29 See, for example, Roland 1. Robinson, "The Impact of Monetary and Fiscal Policy on Residential Financing," in United States Savings and Loan League, Proceedings of the Conference on Savings and Residential Financing. 1 9 5 8 (Chicago: The League, 1958), pp. 130- 134. Federal Home Loan Bank (FHLB) are permitted to borrow funds, under certain conditions, from the FHLB. These monetary policy theorists have pointed out that the Federal Reserve Board of Governors do not have control over the operations of the FHLB. This lack of control at times can result in actions which are contrary to the Federal Reserve Board's current policy objectives. That is, it is possible that during a period in which the Federal Reserve has adopted a policy of general credit restraint, i.e., a tight money policy, money may be made available to savings and loan associations from the FHLB. This, in effect, nullifies the effectiveness of the Federal Reserve Board's actions. It is partly for this reason that some writers would have the FHLB brought under the direction of the Federal Reserve Board. The respondents who represented savings and loan associations were asked a series of questions which were specifically designed to examine their associations1 relationship with the FHLB. The respondents were asked, "Did you borrow from the FHLB during 1955-1957 or 1960-19617" Three of the respondents indicated that they had in the 1955-1957 period. That is, three-fourths of the local associa tions in existence at that time did so. This same 279 ratio (six out of eight) held for those who borrowed from the FHLB during 1960-1961. Another question asked was, "To what degree did you use FHLB borrow ing?" The replies were divided equally among those who did so to a great extent and those who made very little use of FHLB borrowings. All the respondents were then asked, "Did the restrictions placed on borrowing by the FHLB or Federal Reserve have a significant impact on your operations?" Six replied that the FHLB or Federal Reserve restrictions did not have a significant impact on their operations. Two respondents failed to reply. Then the respondents were asked, "Compared with 1955 and 1959, were borrowers' credit requirements more selective, less selective, or about the same in 1957 and 1961?" Four of them indicated greater borrower credit selectivity in the latter periods, two replied that there was less selectivity, and the remaining two replied that there was no change in the degree of borrower credit selectivity. The next question asked was, "During the 1955-1957 and 1960-1961 periods did legitimate loan requests or applications exceed available funds?" The responses indicated that half of the associations' legitimate loan demands exceeded their available funds. The other four association respondents said "no." 280 The respondents were then asked about property require ments in the question, "Compared with 195S and 1959, did your property selection requirements in 1957 and 1961 become more selective, less selective, or stay about the same?" Five of the respondents reported greater selectivity in their property requirements, only one reported less, and the other two stated that there had been no change in their property selection requirements. These data tend to confirm the findings reported earlier. Interest rates and fees charmed. Table 64 reports the responses to two questions which were asked of the respondents. These questions simply asked for their most usual interest rates charged in 1955 and 1959, and 1957 and 1961. The tabulated responses indicate that not only were interest rates higher in 1959 (6.6 per cent) than in 1955 (6.0 per cent), which were both years of generally high business activity, and higher in 1961 (6.25 to 7.0 per cent) than in 1957 (6.0 per cent), which were both periods of general business contraction, but also that the trend was of a gradual rise in interest rates during the entire study period. From 1955 to 1961 fees charged on loans by the savings and loans also developed a trend toward 281 TABLE 64 MOST USUAL INTEREST RATES CHARGED BY SACRAMENTO SAVINGS AND LOAN ASSOCIATIONS IN SELECTED YEARS, BY NUMBER OP TIMES MENTIONED Year 6,094 6.25% 6.696 6.75% 6.994 7.0* 1955 2 1957 2 1959 3 2 1961 1 1 1 1 3a aIncludee one 6*6 to 7.0 per cent response. Source: Lender questionnaire responses. 282 gradually higher levels. The respondents were asked to indicate the fees charged on their loans in 1955 and 1959, and 1957 and 1961. Their responses are tabulated in Table 65* The data reflect a distinction between loans made (1) on existing houses, and (2) for the construction of houses. Borrowers had to pay higher fees for construction loans in alnost all instances. Also reflected in the data was the shift away from the flat $10.00 fee and 1.0 per cent charge on loans to the currently higher per cent of loan fee. Loan-to-value ratios. Savings and loan associa tions increased their fees during recent yearsj how ever, they also have assuned relatively and absolutely greater risks. That is, the average loan Maturity has gradually increased fron ten years to twenty-four years. Furthermore, the ratio of loan-to-sales price also increased as indicated in Table 66. The data indicate that greater amounts were being loaned on individual Sacramento properties since the loan-to-sales price ratio increased fron 60.0 per cent in 1955 to 75.0 per cent in 1961. The latter ratio was also that which was mentioned most often. It was mentioned by four of the eight Sacramento savings and loan associations. TABLE 65 FEES CHARGED BY SACRAMENTO SAVINGS AND LOANS FOR LOANS MADE IN SELECTED YEARS, BY NUMBER OF TIMES MENTIONED Loan Foes on Existinc Houses Construction Loan Fees Year S10.00 1.0% 1.9% 2.0% 2.8% 3.0% 1.0% 2.6% 2.5% 3.0% 4.0% 1955 1 1 1 2a 1957 1 1 1 1 1959 2 2 1 4b 1961 1 2 1 1 2 2 2C aIncludes one rate range of 2.t to 3*0 per cent and one rate range of 2.0 to 4.0 per cent, ^Includes one rate range of 2.0 to 5.0 per cent, one rate range of 2,5 to 3.0 per cent, and one rate range of 2.0 to 4,0 per cent. cIncludes one rate range of 2.0 to 5.0 per cent and one rate range of 3.0 to 4.0 per cent. Source: Lender questionnaire responses. TABLE 66 SACRAMENTO SAVINGS AND LOANS' MOST USUAL TERM FOR REPAYMENT, AND LOAN-TO-SALES PRICE RATIO IN SELECTED YEARS, BY NUMBER OF TIMES MENTIONED Year Most Usual Tern for Repayment (Years) Number of Tines Various Loan-to Price Ratios Mentioned -Sales £0.0*; 66,0% 66.6* 70.0* 71.0* 75.0* 6o.o* 1955 10 2 1 1957 10 1 1959 15 1 1 1 1 1 1961 20-24 1 1 4 l Source: Lender questionnaire response*. 284 285 Government underwritten loans. Another facet of the real estate lending operations which was intended to be examined through the use of the ques tionnaire was the percentage of lending activity which was channeled into conventional loans, FHA loans, and VA loans during the 1955-1957 and 1960-1961 periods. This question was asked of the respondents, and their responses are tabulated in Table 67. These data indicate that Sacramento savings and loan associations definitely favored the conventional loans over both FHA and VA loans. The trend was fron 95.0 per cent conventional loans by one respondent association in the 1955-1957 period to almost 100.0 per cent by seven of the respondent associations in the 1960-1961 period. Impact of credit restraint. Finally, two other questions which relate to the impact of credit restraint on the local savings and loan were asked. It has been pointed out in the literature that the use of the advance commitment technique often inter feres with the effectiveness of Federal Reserve actions. With this thought in mind, an attempt was 30 See, for example, James J. O'Leary, "The Effects of Monetary Policies on the Mortgage Market," The Journal of Finance. XII (May, 1958), 185-187. 286 TABLE 67 DISTRIBUTION OF PERCENTAGE OF SACRAMENTO SAVINGS AND LOANS* CONVENTIONAL, FHA, AND VA LENDING ACTIVITY IN SELECTED YEARS Tvne of Loan 1955-1957 I960--1961 Percent- age Loans Nuaber of Tines Mentioned Perceni- age Loans Nuaber of Tines Mentioned Conventional 95.0 1 100.0 6 FHA 5.0 1 98.0 1 VA 0.0 1 2.0 1 Source: Lender questionnaire responses. 287 made to determine the extent of use made of this technique during the study period. The question asked was, "Compared to 1955 and 1959, did you make more, less, or about the same use of the advance commitment technique in 1957 and 1961?" Four of the savings and loan respondents replied that they had made no use of this technique; three others reported they had made very little use of advance commitments. Therefore the indications are, based on this limited sample, that savings and loans made little or no use of this technique. Consequently, their actions, or lack of actions, in utilizing this device could not have had much, if any, effect on the effectiveness of Federal Reserve policy. The final question asked was, "What do you consider is the major impact of credit restraint on your operations?" The responses received are rather significant. Only two of the respondents felt that general credit restraint policies led to fewer real estate loans on their part; four others replied that the major impact of credit restraints would have little or no effect on their operations. One other savings and loan respondent felt that "it helps the demand for money from us. People who don*t use savings and loans come to us when credit is tight." 288 The eighth savings and loan respondent made the follow ing statement: "Not too important because credit restraint deals in the short-term. It does lead to an increase in savings deposits, however. The average loan is paid off in seven or eight years. Therefore, [the impactJ is not tied too tightly to [credit restraintJ policy because our money doesn't come from outside sources." Commercial Bank Respondents1 Replies The commercial bank respondents unfortunately appeared to be less familiar with the total scope of their operations than were the savings and loan respondents. This apparent lack of total familiarity nay be, in part, the result of the "branch office" character of the commercial banks serving the local area. In comparison with the savings and loan asso ciations, the commercial bank's main office is located, for the most part, outside the Sacramento area. Furthermore, the commercial banks effectively have practiced a greater division of labor in carrying out their functions, and hence have been termed department stores of finance. That is, being large institutions, they have been able to employ a larger nuaber of 289 specialists to carry out the activities of the various departments. This may be the reason why there appeared to be less familiarity on the part of respondents with the total scope of their operations. In contrast, the savings and loans, being specialised, smaller local institutions, rely on a few well-rounded executives. Inter- and intra-industry price and nonprice competition. The first specific-type series question asked was, "What was the major reason for the increase in the interest rate recently?" All six bank respond ents who replied indicated that it was the savings and loan competition and the increased demand for long-term real estate loans which influenced the recent interest rate increase. This question was followed by the inquiry, "What differential in the rate paid to savers is necessary for you to attract savings away from local savings and loan associations?" Also, "What services?" Two of the respondents replied that rate differentials were not important. In their opinion the bank services offered were more influential than rate differentials. The other six bank respond ents' replies ranged from one-half of 1 per cent to 2 per cent rate differentials. One of them simply stated "half of 1 per cent." A third responded 290 "three-fourths of 1 per cent to 1 per cent," and another eaid nl to 1.25 per cent." The fifth respond ent replied "2 per cent." These responses, therefore, indicate that as long as the spread or rate of differ ential between cowsercial banks and savings and loans does not becone greater than indicated above, the banks can successfully conpete with the savings and loans for savings funds. These respondents felt that all the various bank services, including bank credit card service, an active pronotion progran, and the safety-1iquidity features of connercial banks, are decided advantages in their conpetltion for savings. The bank respondents were asked, "Conpared to other banks, what differential in the rate paid to savers is necessary for your bank to attract savings out of other areas?" Seven responses indicated that this is not an inportant factor. This, therefore, reflects another distinguishing feature between banks and savings and loans. The latter institutions sponsor heavy advertising campaigns in order to attract savings funds fron outside areas as well as locally. Sources of savings increases. The bank respond ents were asked the question, "Between new accounts 291 opened and increases in existing accounts, what was the relative importance of each in contributing to gross savings receipts?” They replied in the follow ing manner: three stated that it was evenly divided, two did not respond, two felt that increases in existing accounts were relatively more important, and only one felt that new accounts were more important. In contrast, as reported earlier, savings and loan respondents had indicated that new accounts were rela tively more important to their gross savings receipts. Another significant contrast was found in the replies to the question, ”What percentage of new accounts came from the Sacramento area?” The bank respondents' replies ranged from 15.0 to 100.0 per cent. More specifically, one response in each of the following ratios was received: 15.0 per cent, 50.0 per cent, 75.0 per cent, 80.0 per cent, 99.0 per cent, 100.0 per cent, unknown, and no response. That is, only two of the bank respondents indicated more than 90.0 per cent of their new accounts came from the Sacramento area, whereas five savings and loan respondents had so indicated. The inference might be made that the savings and loan associations' aggressive advertising policies, coupled with higher rate paid to savers, have been more effective. The next question asked 292 was, "What were the major sources of new local accounts?" The replies received in response to this question were definitely similar, in most respects, to the savings and loan respondents' Responses. Thus, the major sources of new local accounts were individuals already in the existing population who were attracted by the advertising promotions. The major differences in the responses were that com mercial banks were able to (1) attract business accounts, and (2) attract savers away from savings and loan associations. Two other questions asked in this series were, "What percentage of outside new accounts came from broker money?" "What were the major reasons for withdrawals?" Responses to the first of these two questions indicated very little, if any, broker money activity. Six respondents replied "nil," and one replied "15.0 per cent.” The responses to the with drawal question reflected the savings and loan competition, which two respondents specifically mentioned. Other reasons involved population mobility; investments in homes, cars, or stocks and bonds; and one reason involved nonprice competition from other banks. One bank respondent replied "convenience of other banks and their too numerous 293 branches." Sale of Mortgages as a source of funds. The next series of questions dealt with the banks' Mortgage activities. The respondents were asked, "Have you ever sold Mortgages?" There were four "yes" and four "no" responses. The latter tabulation included one statenent, "not at this branch." The inplication was that while the local office had not sold any Mortgages, their hOMe office, located outside of SacraMento, had sold Mortgages on local properties. The respondents were then asked, "Did you sell Mortgages during the period 1955-1957 or 1960-1961?" Their replies May again be contrasted with those of the savings and loan respondents. Two bank respond ents had sold Mortgages during the 1955-1957 period, three had not. None of the savings and loan respond ents had sold Mortgages in that period. Regarding the 1960-1961 period, four of the bank respondents had sold Mortgages and two had not, whereas none of the savings and loan respondents had sold Mortgages in this period. The last question asked in this series was, "Did you curtail sales during 1955-1957 or 1960-1961 due to a drop in Market price?" Of the two bank respondents who had sold Mortgages in 1955- 1957 only one indicated that his bank curtailed 294 sales later In that period "due to large money supply, sales decreased." None of those who sold mortgages during the 1960-1961 period indicated that they cur tailed sales due to the drop in market price in that period. Bond portfolio, open market. and Federal Reserve loans. Another series of questions, related to their bond investments, was presented to the bank respond ents. They were first asked, "Did the dollar amount of bond portfolio increase during 1955-1957 or 1960- 1961?" This was followed by a request for an explana tion of their action. The responses which were received regarding these questions reflected a lack of total familiarity of the scope of operations. Six of the bank respondents failed to reply. One indicated that their bond portfolio increased in 1960-1961. The other bank respondent reported no change in their bond portfolio during either of those periods. The bank respondent who reported an increase in his bond portfolio explained that this was a result of the growth of the bank. When asked, 'fWhat was the maturity of bonds purchased?" he replied, "Short-term six months, one year, and five year maturities." The respondents were asked, "Did you borrow in the open 295 market?” Five did not reply to this question, two indicated that they had borrowed in the open market, and the eighth respondent reported that his bank had not. The two respondents who had borrowed explained that they had done so to make more money available in 1955, and because of the growth in loan demands at their banks. The subject of the next question was Federal Reserve and bank borrowing activity. The specific question was, "Did you borrow from the Federal Reserve during 1955-1957 or 1960-1961?” The responses received indicated that two banks borrowed from the Federal Reserve during 1955-1957, and one had not. Also, three banks borrowed from the Federal Reserve in 1960-1961, one did not, five gave no response to this question. The borrowing bank respondents were asked, "To what degree did you use Federal Reserve borrowing?” One replied "not much” to this question, the others gave no response. The bank respondents were then asked, "Did the restric tions placed on borrowing by the Federal Reserve have a significant impact on your operations?" Only two of the respondents replied to this question. One of the two indicated that Federal Reserve restrictions on borrowing had a significant impact on his bank's operations; the other indicated that 296 the restrictions had not had any significant inpact on his bank's operations. Credit selectivity. "Compared with 1955 and 1959, were borrowers' credit requirenents nore selective, less selective, or about the sane in 1957 and 19617" In response to the foregoing ques tion, three of the bank respondents replied "less selective." Two others felt that the borrowers' credit requirenents were about the sane. The other three bank respondents did not reply. All of the respondents were then asked, "During the 1955-1957 and 1960-1961 periods did legitinate loan requests or applications exceed available funds?" Two respondents did not reply; two others reported legitinate loan requests exceeded available funds during the indicated periods. Three stated "ne." The renaining bank respondent renarked that they had plenty of noney during 1955-1957; however, noney was tight during 1960-1961. The next question was, "Compared with 1955 and 1959, did your property selection requirenents in 1957 and 1961 becone nore selective, less selective, or stay about the sane?" Fifty per cent of the respondents reported their property selection requirenents stayed about the 297 sane; one respondent reported nore selective property selection requirenents in 1957 and 1961 than in 1955 and 1959; none of the three others responded. Interest rate and fees charged. As was the case with the savings and loan respondents, the bank respondents were asked their nost usual rate of interest charged in 1955 and 1959, and 1957 and 1961. The bank respondents' replies are tabulated in Table 68. The trend to higher interest rates charged by savings and loans, which was evident in the data in Table 64, page 281, is also evident in the banks' ■ost usual interest data in Table 68. By 1961 six of the banks were charging up to 6.5 per cent as their nost usual interest rate. Only one bank reported a straight 6 per cent as the aost usual interest rate. In contrast, in 1959 only four banks reported as high as 6.5 per cent, and alnost an equal nuaber (three) reported a 6.0 per cent aost usual interest rate charged. A coapariaon between the data in Tables 64 and 68 indicates that the banks were charging generally lower interest rates than the savings and loans. Sacraaento banks coapared favorably in one other respect. That is, the data in Table 69 indicate that relatively lower fees were charged by banks them 298 TABLE 68 MOST USUAL INTEREST RATE CHARGED BY SACRAMENTO BANKS IN SELECTED YEARS, BY NUMBER OF TIMES MENTIONED 6.596 4C aIncludes one 5.5 to 6.5 per cent response. ^Includes one 5.75 to 6.5 per cent and one 6.0 to 6.5 per cent response. °Includes one 5.75 to 6.5 per cent response. dIncludes one 5.75 to 6.5 per cent and one 6.0 to 6.5 per cent response. Source: Lender questionnaire respouses. Year_________ 5.596_______ 5.7596_______ 6.096 1955 1 1 1957 1 1959 3 1961 1 299 TABLE 69 FEES CHARGED BY SACRAMENTO BANKS FOR LOANS MADE IN SELECTED YEARS, BY NUMBER OF TIMES MENTIONED Year $15-25 $25-30 $50-75 0.596 1.096 1,596 2.096 1955 1 1 1 1957 1 1 1959 1 1 la 1961 lb lc la la 1 aNew construction loan fee. ^Response stated, "Minimum $35, maxinun $55." cResponse stated, "From $15-25 to half of 1 per cent of loan." Source: Lender questionnaire responses. 300 by savings and loans (Table 65, page 283) during the selected years. The fee charged aost often by the reporting banks was 1 per cent, as coapared to the 2 per cent charged by aost savings and loans. Loan-to-value ratios. The bank respondents' replies suanarized in Table 70 were aade to the ques tion, "What was the loan-to-sales price ratio followed by you during 1955, 1957, 1959, and 19617" These data indicate that there was little, if any, change in the banks' loan-to-sales price ratio during the selected years. This reflects the fact that there has been little change in national and state bank regulation regarding real estate lending activity. Connercial banks must, by regulation, restrict their loans rela tively nore than the savings and loans. That is, the two-thirds of sales price maxima bank regulation, which has only recently been relaxed, as coapared to the three-fourths or more savings and loan aaxiaua. These ratios do not apply to FHA or VA loans. The data reflect the fact that the majority of the banks which aade real estate loans did so at the aaxiaua ratio (66.6 per cent) throughout each of the selected years. 301 TABLE 70 DISTRIBUTION OF BANKS' LOAN-TO-SALES PRICE RATIOS IN SELECTED YEARS, BY NUMBER OF TIMES MENTIONED Perctntagt of Loan-to-Sales Prieo, Y.ar SS16K " W . ' flK 1955 1 1 3a 1957 1 1 3a 1959 2 5a 1961 1 1 4 1 aIncludes one 60.0 to 66*6 per cent response. Source: Lender questionnaire responses. 302 Government underwritten loans. Another question asked of all the bank respondents was, "Approximately what percentage of lending activity was channeled into conventional, FHA, and VA loans during 1955-1957 and 1960-1961?" The responses to this question are shown in Table 71. These data reflect the shift away from government insured (FHA) or guaranteed (VA) loans to the conventional loans. Five of the eight bank respondents reported that 95.0 per cent or more of their bank loans were conventional loans during the 1960-1961 period. One respondent did not reply; one other reported an equal distribution between the con ventional loans and the government underwritten type. The eighth bank respondent reported that 95.0 per cent of his bank's loans during both periods were FHA and VA loans combined. The advance cnaugtment t an offset to Federal Reserve policy. The advance commitment technique was not used to any great extent by the local savings and loan associations. In contrast, bank respondents indicated relatively greater use of this technique than the savings and loans. The question asked was, "Compared to 1955 and 1959, did you make more, less, or about the same use of the advance commitment tech- 303 TABLE 71 DISTRIBUTION, BY PERCENTAGE COMBINATIONS, OF SACRAMENTO BANKS CONVENTIONAL, FHA, AND VA LOANS DURING SELECTED YEARS, BY NUMBER OF TIMES MENTIONED Percentage Combinations of Types of Loans Conventional FhA-VA______ 1960-1961_______1955-1957 5.0 95.0 1 40.0 60.0 50.0 50.0 1 95.0 5.0 1 98.0 2.0 1 100.0 3 Source: Lender questionnaire responses. 304 nique in 1957 and 1961?" Two bank respondents failed to respond; however, three others indicated they aade about the sane use of the technique during the latter years. Two othersreported greater use of the advance coaaitaent technique. It is rather significant to note that the eighth bank respondent replied, "It was used very auch in 1960-1961, however, in 1962 it has been used very little." Thus, three banks reported aore use of the technique and three banks reported at least the saae usage of the advance coaaitaent tech nique during the selected years. Therefore, in the coaparison of the usage of this technique by the two types of lending institutions, the banks and the savings and loans, it appears that the greater use by the banks and the relative lack of use by the savings and loans would place the weight of the charge of nullifying Federal Reserve policy on the coaaercial banks, ironically, the institution which is directly regulated by the Federal Reserve. Ianact of credit restraint. It is noteworthy that the bank respondents' replies reflected a concern regarding the availability of funds when asked the question, "What do you consider is the aajor iapact of credit restraint on your operations?" Four of the 305 bank respondents related the competition between long-term real estate loan funds and alternative investments with the availability of money. Three respondents failed to reply to the question. One other simply stated "none." More revealing, however, was the response, "The federal authorities' actions have created an excess of loanable funds, thus driving down the profitability of commercial banks." Another particularly revealing response was, "Top management picks this period up quickly, looks at the portfolio to see what funds are available. May curtail long term real estate lending." Summary This chapter discussed the impact of the Sacramento savings and loan industry on residential construction in the area. It provided evidence that the local savings and loan industry, and one holding company facility, in particular, aided by timely advances from the Federal Home Loan Bank of San Francisco, financed the major portion of the resi dential construction starts in Sacramento during the latter part of the 1950-1960 decade. Another section of the chapter utilised empirical data to describe the role of institutional lenders in providing credit 306 for local real estate transactions. The findings of a survey which sought to examine the attitudes and opinions of lenders with respect to the effect of changes in the Federal Reserve Board's monetary policy on mortgage lending activities con cluded the chapter. While the limited population involved in the survey does not provide conclusive answers to the underlying general questions, it does, however, provide some helpful insights into the impact of restrictive monetary policy actions. For example, the survey responses did not conclusively support the hypothesis that during periods of tight money and restraint conditions lenders exercise a greater degree of restraint than during easy money periods. Restated in other words, a restrictive monetary policy dees not necessarily cause a decided shift in investment policy on the part of local lenders. The survey responses imply, with respect to increased rates of return to savers and the propensity to save, that interest and dividend rate increases were primarily due to intra- and inter-industry competition and not to the impact of monetary restraint; and savers' propensity to save received little stimulation from monetary policy during the period of the study. PART D GENERAL SUMMARY AND CONCLUSIONS CHAPTER VII SUMMARY AND CONCLUSIONS Research of the literature and interviews with economists and knowledgeable practitioners led to the development of the hypothesis underlying this study. Briefly stated, the central thesis is that increased rates of nonfarm residential construction are influ enced by (1) favorable changes in the economic base of an area, (2) availability of funds for real estate credit at savings and loan associations, and (3) the Federal Reserve System's monetary policies as they affect the availability of funds of nonbank financial intermediaries. It was the purpose of this investiga tion to determine if the hypothesis was reasonable and consistent with the experiences noted at the national, regional, state, and local levels. A statistical analysis of (1) empirical data developed at the local level, and (2) published data related to the larger geographic and economic areas was employed, in addition to other methods, to test the validity of the hypothesis. It should be noted, however, that "statistical evidence never provides positive proof of the truth of a hypothesis. The essence of sta tistical testing is that the facts are given a chance 309 to disprove hypotheses."* Therefore, "if we are to have confidence in a hypothesis it aust have support beyond the statistical evidence. It aust have a rational basis."2 As this investigation suggests, the hypothesis does, of course, have a rational basis. I. SUMMARY The deaand for real estate credit is a derived deaand. The iaportant econoaic factors which affect the deaand for real estate and hence the deaand for real estate credit are: (1) eaployaent levels, (2) incone levels, (3) the quantity and quality of avail able housing, and (4) the availability of capital to finance the construction of new housing units or to assist in the purchase of existing housing units. The effect of changes in eaployaent and incone levels, both being key sectors in econoaic baee analysis, on new nonfarn residential construction starts was exaained early in the study. The influence of the availability of capital to finance new residential construction was exaained in a later chapter. *Frederick C. Mills, Statistical Methods (third edition; New York: Henry floit and Coapany, 1955), p. 243. 2Ibid., p. 244. 310 Econoaic Base Theory By definition, the econoaic base of an area is the sua total of all econoaic activity in that area which produces eaployaent and incoae. The underlying concept of econoaic base analysis is that as basic econoaic activity expands in an area, as opposed to stability or decline in other areas, workers are attracted froa the areas of lesser or declining oppor tunity. Urbanisation of these workers in the dynaaic area leads to additional econoaic opportunities for other workers, particularly in the nonbaaic or service industries. The expansion of such opportunities leads to inaigration and population growth. Under such conditions the existing supply of housing would not be sufficient to aeet the increased deaand for housing units. The resultant bidding for the existing units would then raise prices and builders would be encour aged to build houses. Econoaic theory has, in the past, paid priaary attention to long-run national aggregative data, and as a result housing policy has been established with little consideration for local forces which influence local residential construction activity. This investi gation utilised changes in eaployaent and per faaily incoae at the regional, state, and aetropolitan area 311 levels as indicators of changes in their respective econoaic bases. These eaployaent and incoae changes were then related to changes in the rate of new non- fara residential construction at the various area levels in order to help bring this sector of econoaic theory into its proper perspective. Rap1 ny.nt. Level Chanae s Coanared with Changes in Nonfara Residential Housing Starts Because of the high degree of specialisation which is characteristic of aodern econoaic activity the various econoaic areas of the United States have becoae highly interdependent. Consequently, national developaents influence all areas of the nation in varying degrees over tiae. It is unlikely that any one area can insulate itself coapletely froa the effects of aajor national cyclical swings in general business activity. Evidence of the inability of any one area to coapletely insulate itself was shown by an analysis of the regional eaployaent data. These pertinent data were used to tost the econoaic baso theory relating changes in regional eaployaent with changes in regional nonfara residential housing starts 312 during two recessionary periods which weakened the econoaic bases of aany areas in the United States during the 1954-1960 period. In general, the two regions which reported the greatest percentage fluctuations in eaployaent also reported the greatest percentage changes in housing starts during the study period. The two regions which reported less deviation in their respective eaployaent levels reported less volatility in their respective housing starts levels. It appears, therefore, that a strong degree of correlation nay exist between changes in regional eaployaent levels and regional nonfara residential construction starts. An exaaination of the relevant data at the state level indicated that a close relationship also exists between changes in eaployaent and nonfara residential construction starts in each of the selected states observed in this study. The discernible pattern of fluctuations in annual eaployaent levels, correlated with changes in annual residential construction starts at the regional and selected states levels, was also apparent at the selected aetropolitan area levels. However, it was to be expected that, consistent with econoaic base theory, greater fluctuations aight occur at the local level than at the state or regional level. 313 Consequently, changes in eaployaent levels in any one aetropolitan area will have a greater effect on the annual housing starts within that area. This aay lead to considerable changes in housing starts during the short cycle of business activity. In brief, annual housing starts in the selected econoaic areas reflected diversified aagnitudes of reactions to changes in their econoaic bases, as aeasured in teras of changes in the level of annual eaployaent in the area. It aay be concluded, there fore, that variations in the econoaic base have played a definite role in the deteraination of the level of annual housing starts during the short-tera cycle. Incoae and Housing Starts Correlation analysis is one of the statistical tools available for econoaists to coapare the degree of relationship between a pair of variables. This technique was used in the current study to test the relationship between two variables in the four rogions of tho Unitod Statos, eleven selected states, and twelve selected aetropolitan areas. The results of the correlation analysis failed to develop any dis cernible pattern of relationship between the two variables— per faaily incoae and nonfara housing 314 units authorized— in any of the various areas. These findings support the literature which stated that the analysis night not indicate a direct cause-and-effect relationship but that the relationship night be nor# of an indirect nature. In ether words, changes in per faaily incones produce certain secondary factors such as doubling and denand filtering which influence hous ing denand in the short-run. The absence ef statis tical evidence which relates changes in incone levels to changes in nonfarn residential starts does not invalidate the hypothesis. Greater reliance aust be given to changes in eaployaent levels. This was shown te be consistent with accepted econoaic base theory. Sacranento Econoaic Base Changes Related to Local Housing Starts: An Eaniricai Analysis Industrial, coaaercial, and governneat activ ities in the Sacranento Metropolitan Area have expanded, as in other areas of the country, during the 1950-1960 period. As a result of the investaent which has taken place in a great variety of activities within the Sacranento area, eaployaent and incoae opportunities have been created in the econoaic base. An analysis of Sacranento's basic and nonbasic industries' enploy- 315 aent data provided further insight into the labor force and the econoaic base. Basic industries, by definition, are those eaployaents which hring incoae into an econoaic area. The expansion of these basic industries was, therefore, the priaary cause of the area's econoaic growth. Priaarily as a result of expansion of basic industry in Sacranento, total eaployaent increased 192 per cent during the years between 1940 and I960. More inportant eaployaent nultipliers, which reflect the iapact of changes in basic and nonbasic eaployaent as a result of business fluctuations throughout the national, regional, and local econoaies, show that for each local basic industry worker there were 1.5 local nonbasic industry workers eaployed in the 1950-1960 period. Furtheraore, there have been significant structural changes which have taken place in the local econoay during the sane period; however, federal, state, and local govemaent activities continue to be aajor influences in the local econoay* The defense-oriented industries have assuaed an inportant role in the growth of the Sacra nento Metropolitan Area. In general, there has been a trend away froa specialisation toward relative diversification. 316 Defense-oriented Industry was cltsd as being responsible for the increase in the average wage in the manufacturing sector of the local economy. Most of this incroaso aay bo related to the location of ono now firm in the Sacramento Metropolitan Area. Each new job at the firm's plant in Sacranento has (1) had a multiplier effect of 1.5 jobs in the non basic sector of the local economy, and (2) had an effect on the rate of now residential construction starts in the area. The impact of favorable changes in the local econoaic base on the rate of residential construction starts was analysed and the results nay be interpreted to show that a causal relationship exists between the increases and decreases of the total annual additional employment and increases and decreases in the number of building permits issued. Xh£ IttHVtBgf SL Selected Demographic Factors on Nonfarm Residential Construction Starts Although the basic influence on the new housing sector of the over-all demand for housing is popula tion growth, it is the mobility aspect of population growth which is the key factor in the deaand for new housing. That is, population growth, principally 317 through Migration, ia the underlying force in the residential construction Market. In general, the data in Chapter III offer conclusive evidence that a good correlation exists between the year-end share of regional population and the share of total annual housing starts which occur in the region. A cooperative analysis of population changes and housing starts changes in fourteen selected states was utilized to show that (1) above average increases in population were acconpanied with above average increases in housing units; (2) tho seven states which had the largest population incroasos, on a poreentage basis, also incroasod thoir respective shares of total Unitod Statos housing units; (3) tho seven states which had the lowest population incroasos, on a porcentago basis, also decroased thoir rospoctive shares of total United States housing units, and (4) the foregoing are consistent with the conclusion that a good correlation exists between the year-end share of regional population and the share of total annual housing starts which occur in tho rogion. It is also shown that in six of the fourteen selected states where net Migration was a Major factor in tho popula tion growth of that state increases in their respective shares of total United States housing units were 318 reported during the 1950 to 1960 period. In the fire states where net sigration did not equal the United States net Migration rate losses in their respective share of total United States housing units were recorded. Additional evidence to support the thesis that there is a relationship between Migration, eaployaent, and housing deaand was presented in the final section of Chapter III. The fourteen selected states are ranked above and below the United States rate of population change attributable to Migration during the 1950-1960 period. Consequently, there were nine states above and only five states below the United States rate. Also, the percentage increases in total nonfara eaployaent for these selected states were shown along with the United States percentage increase for the saae period. It is iaportant to note that six of the nine states which were ranked above the United States Migration figure reported auch greater eaployaent increases on a percentage basis than the United States percentage increase in total nonfara eaployaent during the years between 1950 and 1960. Only one of the five states which had lost population as a result of Migration reported a greater percentage increase in total nonfara eaployaent than the cea- 319 parable percentage increase in United States employ- ment. It is also significant to note the aggregate need for housing will increase more from a migration of 20,000 people in 6,150 households than from a natural increase in population of the same number of people. The Influence of Credit Availability and Cost of Money on Housing Starts The relationships between new nonfarm resi dential construction activity and the influence of the money rate and credit availability were discussed in Chapter IV. Long-term treasury bond prices and "free reserves" available for lending and investing, which are representative of the independent variables (money rate and credit availability, respectively), were utilized to test that portion of the hypothesis which relates cost of money and credit availability to increased rates of residential construction, A cursory review of the published data for the 1950-1960 decade appears to indicate that the rate of housing starts generally moved in close relationship to activity in the money market. Periods of easy money were associated with accelerated housing-starts rates. Periods of tight money were associated with 320 decreased housing-starts rates, and the honebuilding induatry considered tight aoney periods aa being the aajor restrictive eleaent to an ever-expanding rata of now raaidantial construction. The available data suggests the hypothesis that there ia a direct causal relationship between bond price increases and decreases and housing starts increases and decreases* However, this hypothesis was not validated when subjected te statistical testing. It nay, therefore, be concluded that for the period under review the influence of the cost of aoney on the nuaber of housing starts has not been a factor which can be shewn to be statistically significant by the usual tests* The statistical record, however, dees support the hoaebuilding industry opinion, contrary to that which econoaic theory has thus far denied, that the availability of aortgago aoney credit aay stiaulate an increase in housing starts* An analysis of the evidence suggests that there is a high degree of correlation between aoney availability and increased rates of residential construction since the return to a flexible aonetary policy in 1951* More specifically, aajor shifts (increases) in free reserves during the years since the accord were acconpanied by aajor shifts (increases) in the nuaber of housing units 321 started. Also, major downward shifts in free reserves paralleled major downward shifts in the number of housing units started. It may therefore be concluded that a positive relationship existed between increases and decreases in the availability of money (free reserves) and increases and decreases in the volume of housing units started at the national, regional, selected states, and aetropolitan area levels during the 1952-1961 period. amvimmm and Loan Associations: £ Malor Source of Residential Real Estate Credit A brief analysis of the growth of the savings and loan industry at the national level revealed that these institutions had grown steadily during the 1950- 1960 period. The growth in the nuaber of associations during this period represents a 5.5 per cent increase. However, federally chartered associations increased 22.7 per cent, insured state chartered associations increased 66.7 per cent, and noninsured state chartered associations decreased 29.1 per cent during this period. The growth in the total assets of all the associations represented a 323.3 per cent increase. Asset growth, in dollar terms, was greatest in the federally chartered 322 associations; however, in percentage terns, insured state chartered associations reported the greatest growth. A review of the growth of the savings and loan industry in California revealed that (1) the insured state chartered associations have grown, in terns of both nuaber and total assets, nore than the other two types of chartered associations; (2) the nunerically fewer federally chartered associations are relatively larger in terns of average assets per association; and (3) the noninsured state chartered associations are relatively uninportant. An exanination of the conpetitive shifts of the sources of nortgage credit revealed that savings and loan associations continue to supply the greatest anount of residential nortgage financing in terns of dollar anounts, share of total dollar anount, and the nuaber of loans recorded. During the 1950-1960 period savings and loan associations have increased, alaost without interruption, their share of the total nonfara nortgage recordings as well as the aggregate dollar voluae at both the national and state (California) levels. 323 Impact of Credit Availability on Reaidential Construction Financing by Savings and Loana At the national level construction loana typically accounted for approximately 32.0 per cent of the total loana made by associations during the 1950-1960 period. In contrast, construction loans financed by associations in California typically accounted for more than 46.0 per cent of their total loans during the same period. Additional evidence that it is the availability of funds for real estate credit, and more specifically the ability of savings and lean associations to provide funds, coupled with favorable changes in the economic base, which accompany increased rates of residential construction is found in Chapter V. An analysis of the evidence indicates that Increased credit availability at savings and loan associations was accompanied by an increased number of residential construction starts. During the years in which the amount of money available for real estate credit did not equal the prior year's increase, the change in the number of construction loans also decreased. More specifically, the evidence shows that (1) in seven of the eleven years, 1950-1960, increased 324 credit availability was accompanied by increased residential construction loans financed by savings and loan associations; (2) smaller increases (prior year basis) in credit availability in two of the eleven years were accompanied by decreases (prior year basis) in residential construction loans financed by associations; and (3) in the other two years decreased credit availability was accompanied by decreases (prior year basis) in associations' con struction loans. It was also noted that the associa tions received record advances from the Federal Home Loan banks during the latter part of the 1950's in order to provido funds for roal estate credit during this period. These FHLB advances demonstrate the willingness of associations to borrow in order to maintain their credit availability balances and hence finance residential construction loans. This chapter presented evidence which indicates that there is a positive relationship between major shifts in nonfarm private housing starts and the major shifts in the number of construction loans made by savings and loan associations during the 1950-1960 period. 325 Inpact of Saving! and Loan Associations on Residential Construction in tho Sacranento Aroa Both tho nuaber and total asooto of Sacranento association* have grown at rates which exceed the national and state rates of growth during the 1950- 1960 period. In terns of nunber of associations, Sacranento associations increased 133.3 per cent; in contrast, the rate of increase at the national level was 5.5 per cent, and 40.2 per cent at the state (California) level during the sane period. Total assets at the local (Sacranento) level increased nore than 1,700.0 per cent; whereas total assets at the national level increased only 323.3 per cent and at the state (California) level the increase in total assets was 625.0 per cent during the 1950-1960 period. It was reported that intra-industry conpetition has taken place locally as well as at the larger area levels. Evidence of this fact was the shift in the share of assets held by the various types of savings and loan associations. In terns of average assets the federally chartered association in Sacranento has approxinated the growth of the average federally chartered association in the United States. However, 326 in terns of share of total insured assets held, the local state associations have increased fron only 5.6 per cent of total assets in 1950 to alnost 86.0 per cent in I960. In contrast, at the national level the share of total assets held has renained relatively the sane, and at the state (California) level the najor portion of total assets held has shifted fron the federally chartered associations to the state insured associations. The inpact of a savings and loan holding conpany operation which located in the Sacranento area in nid- year 1952 was reflected in the data which reported that by 1960 it (1) held 36.9 per cent of the total assets of all Sacranento associations, (2) held 40.6 per cent of the total nortgages, (3) had received 86.8 per cent of the Federal Hone Loan Bank advances nade to local associations during the year, and (4) had financed 34.1 per cent of all the construction loans which Sacranento savings and loans nade in 1960. Increases in the FHLB advances account reflect the associations' ability and willingness to neet the denand for real estate credit. The enpirical data developed fron unpublished data indicate that a sharp increase in the share of residential construction loans which the local associations provided occurred 327 in 1957 and continued through 1960. Beginning in 1957 local associations financed alnost 20.0 per cent of all residential construction locally and steadily increased their share to alnost 33.0 per cent in 1960. These sane years (1957 through I960, inclusive) were also the years during which (1) increased rates of residential construction occurred in the Sacranento area, (2) local associations increased their borrow ings fron the FHLB in order to neet the need created by the increased denand for real estate credit in the Sacranento Metropolitan Area, and (3) the area was sonewhat overbuilt. Lender Survey Findings The findings of a lender survey, which had as its purpose the exanination of attitudes, opinions, and policies of nortgage lenders with respect to the effect of changes in the Federal Reserve Board's nonetary policy on the financing of new residential construction, were reported in the study. A partial sunnary of the lender survey findings is as follows: 1. Seventy-five per cent of the savings and loan respondents tended to restrict loans during the 1961-1962 period. In contrast, only 12.5 per cent of the bank respondents, 27.3 per cent of the nortgage 328 companies, and none of the life insurance companies reported a tendency to restrict loans generally during the sane period. 2. The underlying reasons stated for adopting c restrictive policy involved basic considerations of the supply of and deaand for hones. Responses of tho lenders who did not follow a restrictive policy indi cate that, in general, their actions were influenced by an abundant supply of available funds for lending and investment purposes. 3. The thesis that lenders would exercise greater restraint during tight aoney periods than during easy aoney periods is not conclusively sup ported by the data. Almost 18.0 per cent of tho respondents reported that at no tiae during the 1955- 1962 period had they exercised a greater degree of restraint in processing real estate loans. 4. Commercial bank and life insurance eoapany respondents indicated that they had restricted real estate loans during most of the years in the study period (1950-1960). That is, restrictions occurred during the upswing stages of the business cycle and the heaviest concentration of responses to that effect occurred during the build-up to the May 1960 peak in general business activity. Thus, the responsive 329 nature of commercial banks to Federal Reserve Board nonetary policy action was reaffirmed. 5. The underlying reason for increases in the dividend rate paid by savings and loan associations early in 1962 was coapetition froa other associations. 6. Nonprice coapetition factors, such as location and convenience, better quality services and reputation, safety of accounts, sise of associa tion, and public confidence, were identified as being necessary services in order to coapete successfully with banks. 7. Local savings and loan associations generally assumed a policy of "follow the industry leader" for establishing the price they pay for savings. 8. Local associations stated that they need a 1 per cent rate differential in order to attract savings away from local banks. 9. Inter-industry price competition influenced the coaaercial banks to raise, with the supervisory authorities' approval, the interest rate paid to their savers. Another reason for the raising of the inter est rate was to attract funds in order to fulfill the increased deaand for long-tern real estate loans. 330 10. The saving* and loan associations' aggressive advertising prograas, coupled with higher rates paid to savers, have been aore effective than those of the coamercial banks; however, the latter have been able to attract business accounts aore effectively than savings and loans. 11. Both banks and savings and loan associa tions indicated a trend to high interest rates and charges; however, banks generally charge lower rates of interest and loan fees than the savings and loans. 12. Little use of the advance conaitnent technique is nade by the local savings and loan associations. Bank respondents, however, indicated greater use of this technique than did the savings and loan respondents. It appears, therefore, that Federal Reserve policy can, at tines, be offset by the conaercial bank, ironically, the institution which is directly regulated by the Federal Reserve. II. CONCLUSIONS The following fundanental conclusions energe froa the present study of the econonic deteminents which effect changes in the rate of residential con struction at the national, regional, selected states, and selected netropolitan area levels. 331 1. In this investigation the available evidence has, in general, supported the central thesis advanced early in the study. That is, the underlying hypothesis has been tested and found to be reasonable and con sistent with the experiences noted at the national, regional, state, and local levels. 2. Expansion of economic opportunities leads to Migration and population growth. The aggregate need for housing in a dynamic area will increase aore froa population increases caused by Migration than froa natural increases. Builders will be encouraged to build houses to meet the increased deaand stimulated by Migrants who are attracted by the favorable changes in the economic base of the dynamic area. 3. The influence of the cost of money on the number of housing starts is not a factor which can be shown to be statistically significant by the usual tests. 4. The availability of funds for real estate credit, coupled with expanding economic opportunities, may stimulate an increase in housing starts. 5* Financial institutions, in general, and savings and loan associations, in particular, are filling a basic need in the local economy— promoting savings and providing the necessary real estate credit 332 for residential construction and hone financing. 6. Both inter- and intra-industry coapetition have taken place within the local financial coaaunity as well as at the larger econoaic area levels. Savings and loan associations have grown at the relative expense of other institutional lenders. 7. The need for careful and detailed research on the iapact of aonetary policies on the local aortgage market continues to exist. 111. SUGGESTED AREAS OF NEEDED RESEARCH The need for quantitative, analytical, and conceptual research continues to exist. The hypothesis underlying this study should be subjected to additional tests in other regional and aetropolitan areas. Systeaatic investigations of the econoaic base of the various regions and other local areas throughout the United States should be undertaken. Data aade avail able by the current study nay then be coapared and analyzed with other relevant data. The suggested conparative analyses nay produce aore conclusive support for the underlying hypothesis. This would enhance the developaent of a aore adequate econoaic base theory. Systematic investigations of the nature and structure of the local mortgage markets, and analyses of factors influencing regional and local financial institutions and the inter-regional flow of funds, should provide additional insights into the behavior of real estate markets and the forces determining behavior, including new residential construction. The recommended investigations may further analyze the relationship between high levels of available funds for mortgage lending activities, high levels of residential construction, and the burden and quality of mortgage debt. Related investigations of the effects of monetary policy would produce aore conclusive evidence and corroborative answers to the general questions posed in the study. This should result in a more precise theoretical framework relating changes in the economic base of an area and the availability of funds to new residential construction and the sound growth of the urban complex. BIBLIOGRAPHY BIBLIOGRAPHY A. BOOKS Behrens, C. F. Coaaorcial Bank Activities in Urban Mortgage Financing. New York: National~T5ureau of Economic Research, 1952. Boehnler, Erwin W., et al. Financial Institutions. Honewood, IllinoTs: ""Richard D. Irwin, Incorporated, 1956. Bryant, Willis R. Mortgage Lending. New York: McGraw- Hill Book Conpany, 1956. Burns, Arthur F., and Wesley C. Mitchell. Measuring Business Cycles. New York: National Bureau or Econonic Research, 1946. Case, Frederick E. Real Estate. Boston: Allyn and Bacon, Incorporated, 15527” Colean, Miles. The Impact of Governaent on Real Estate Finance in the United Siates. New Yorlc: National Bureau oT~Econoaic Research, 1950. Cover, John H., et al. Residential Mortgage Financing: Hagerstown. HarvTandT CollegePark, Maryland: University of Maryland Press, 1951. Federal Reserve Board of Governors. The Federal Reserve System. Purposes and Functions, frhird edition.Washington: Federal Reserve, 1954. Fisher, E. M. Urban Real Estate Markets: Character istics and financing, New York: National Bureau of Econoaic Research, 1951. __ and Robert M. Fisher. Urban Real Estate. New York: Henry Holt and Coapany, ld&4. Gillies, Jaaes, and Clayton Curtis. Institutional Residential Mortgage Lending in Los Angeles Countv. 1545-1951. Los Angeles: universliy ofCalifornia, Bureau of Business and Econoaic Research, 1956. 336 Goldsmith, Raymond W. Financial Intermediaries in the American Economy Since I960. Princeton, Mew Jersey: Princeton University Press, 1958. Gordon, Robert A. Business Fluctuations. New York: Harper and Brothers, ldsST Grebler, Leo. The Role of Federal Credit Aids in Residential Construction! New Vork: National Bureau oi Economic Research, 1953. David M. Blank, and Louis Vinnick. Capital Formation in Residential Real Estate. Princeton, New Jersey: Princeton University Press, 1956. Grebler, Leo, and Eugene F. Prigham. Savinas and Mortgage Markets in California: The Position and performance ot ihe Savings and Loan industry. Pasadena, California: California Savings ana Loan League, 1963. Gurley, John G., and Edward S. Shaw. Money in a Theory of Finance. Washington: The Brookings iMilirfionV'im. Honnold, Junia, Richard U. Ratcliff, and Daniel B. Rathbun. Residential Finance. 1950. New York: John Wiley and Sons, Incorporated,1957. Hoyt, Homer, and Arthur Weiner. Principles of Real Estate. Fourth edition. New York: The Ronald ’ Press Company, 1960. Lewis, John P. Business Conditions Analysis. New York: McGraw-Hill Book Company, Incorporated, 1959. Martin, Preston. Real Estate Principles and Practices. New York: The Macmillan Company, 1959. Mills, Frederick C. Statistical Methods. Third edition. New York: Henry Holt and Company, 1955. Moore, Geoffrey H. (editor). Business Cycle Indi cators. Volume I. Princeton, New Jersey: Princeton University Press, 1961. 337 Morton, J. E. Urban Mortgage Lending— Comparative Markets and Experience.Princeton, New Jersey: Princeton University Press, X956. Phillips, E. Bryant. Consumer Econoaic Problems. New York: Henry Holt and Company, T9571 Prochnow. Herbert V. (editor). Aaerican Financial Institutions. New York: Prentice-HalI"J Incorporated, 1951. Ratcliff, Richard U. Real Estate Analysis. New York: McGraw-Hill Book Coapany, Incorporated, 1961. . Urban Land Econoaics. New York: McGraw- Mill Book Company, Incorporated, 1947. Robinson, Roland I. (editor). Financial Institutions. Third edition. Homewood, Illinois:Richard b. Irwin, Incorporated, 1960. . The Management of Bank Funds. New York: McGraw-Hill Book Company, Incorporated, 1951. Russel, Horace. Savings and Loan Associations. Albany, New York: Matthew lender and Coapany, 1956. Saulnier, Raymond Joseph. Federal Lending and Loan Insurance. Princeton, New Jersey: Princeton University Press, 1958. . Urban Mortgage Lending by Life Insurance Companies. New York: National Bureau of Economic Research, 1950. Unger. Maurice A. Real Estate Principles and Prac tices. Cincinnati: South-Western Publishing Coapany, 1959. Wickens, David L. Residential Real Estate. New York: National Bureau of Econoaic Research, 1941. Winnick, Louis. American Housing and Its Use. The Deaand for Shelter Space! New York:John Wiley and Sons, Incorporated, 1957. 338 B. PUBLICATIONS OF THE GOVERNMENT, LEARNED SOCIETIES, AND OTHER ORGANIZATIONS California Savings and Loan League. California Savings and Housing 1962 Data Book. Pasadena: California Savings ana Loan League, 1962. California Tax Agency. The Data Sheet. Northern California Loan Recording Statistical Monthly. San-FrancTsco: California Tax Agency, 1960-1962. Federal Hone Loan Bank Board. Annual Report. 1959. Washington: Governnent Printing Office,19557” Financial Statements. Annual. Washington: Governnent Printing Office, 1950-1960. . Rules and Regulations for Insurance of Accounts. Washington: Government £rinting”5ffice, 1555^1550. . Savings and Hone Financing Chart Book. 1961. Washington: Governnent Minting Office, 1961. . Savings and Hone Financing Source Book. Annual' Washington: GovernnentPrinting Office, 1950-1960. Federal Hone Loan Bank of San Francisco. Roster of Members. Annual. San Francisco: Federal done Loan Bank, 1950-1961. Federal Reserve Board of Governors. Federal Reserve Bulletin. Monthly. Washington: Governnent Printing Office, 1950-1961. _______• Forty-Fifth Annual Report. Washington: GovernmentPrinting 6ffice, 1958. Federal Savings and Loan Insurance Corporation. Annual Financial Report. Washington: The Corporation, 1945-1§59. Great Western Financial Corporation. Annual Report. 1962. Los Angeles: The Corporation, 1963. 339 Housing and Home Finance Agency. Annual Report. Washington: Government Printing bfflce, 1948-1963. Housing Statistics. Monthly. Washington: Government Printing Office, 1948-1963. Hurff, George B., et al. Residential Mortgage Financing. Jacksonville. Florida: First Six Months or 1950. Washington: Government Printing Office,"19327" Rapkin, C., L. Winnick, and O. Blank. Housing Market Analysis. Housing and Home Finance Agency. Washington: Government Printing Office, 1953. Sacramento City-County Chamber of Commerce. Pertinent Information about Metropolitan Sacramento. Sacra mento County. CaXiforn£e7 Sacramento: Chamber of Commerce, January, 1963. Security Title Insurance Company. Bulletin of Loan Recordings. Monthly. San Francisco: The Company, 1955-1960. State of California, Economic Development Agency. California Statistical Abstract. Annual. Sacramento: State Printing Office, 1961-1962. State of California, Savings and Loan Commissioner. Annual Reports. Sacramento: State Printing lOTce,TMfcl961. State of California. California Labor Statistics Bulletin Area Supplement• Sacramento: State Printing WOTcm, United States Bureau of the Census. Business Cycle Deyelopments. Washington: Government Priniing Office, December, 1962. . "Preliminary Estimates of the Components of Population Change, by States: 1950 to I960," Current Population Reports. Series P-25, No. 227 and No. 2429. Washington: Government Printing Office, April 26, 1961 and May 22, 1961. . Statistical Abstract of the United States. Washington: Government Printing 6ffice,l5S5I 340 . United States Census of Population: 1960, Volume I. Washington: Government Printing Office, 1961. United States Congress, Joint Committee on the Economic Report, Subcommittee on General Credit Control and Debt Management. Hearings, Monetary Policy and the Management of the Public Debt. 82nd Congress, 2nd Session. Washington:“Govern ment Printing Office, 1952. United States Congress, Joint Economic Committee, Committee Staff and Office of Statistical Standards, Bureau of the Budget. 1960 Supplement to Economic Indicators. 86th Congress, 2nd Session. Washington: Government Printing Office, 1960. United States Congress, Senate, Committee on Banking and Currency. Study of Mortgage Credit. Hearings before Subcommittee onHousing, 86th Congress, 1st and 2nd Sessions. Washington: Government Printing Office, May, 1959; December 22, 1958, Revised July 11, 1960. United States Department of Commerce. Construction. Monthly. Washington: Governnent Printing Office, 1946-1954. Construction Review. Monthly. Washington: Government Printing 6fiice, 1955-1960. . County Business Patterns. Washington: Government Printing 6flice, 1053, 1956, 1959. United States Department of Labor. Construction Volume and Costs. 1915-1956. Washington: Government Printing 6ifice, 1958. . Bureau of Labor Statistics. Employment and Earnings. Annual Supplement Issue. Washington: Government Printing Office, November, 1961 United States Savings and Loan League. Annals. Chicago: The League, 1947-1960. 341 . Proceedings of the Conference on Savings and Residential Financing. Annual. ""Chicago: the League, 19^8-1962. Articles by the following authors have been cited: Quarterly Letter on Savings and Mortgage Lending! Chicago: the League, 1956-1960. . Savings and Loan Fact Book. Annual. Chicago: TheLeague, 1956-1962. . Statement of Policy. Adopted at the League's TTst Annual Convention. San Francisco: The League, November 7, 1963. Wendt, Paul F., and Daniel B. Rathbun. The San Francisco Bay Area Residential Mortgage Market. Housing Research Paper ^o. 20. Washington: Government Printing Office, May, 1952. Western Title Insurance and Guaranty Company. Loans by Institutions on Sacramento County Real Estate. Monthly. San Francisco: TheCompany,”15^4^15517 Bogen, Jules I. Friedman, Milton Goldsmith, Raymond W. Grebler, Leo Gurley, John G. Kendall, Leon T., and Frank H. Gane Lerner, Abba P. O'Leary, James J. Parker, George Robinson, Roland I. Rogg, Nathaniel H. Samuelson, Paul A. Solomon, Ezra Woolworth, G. IV. 1958 1962 1958 1959 1959 1959 1962 1958 1961 1958 1960 1960 1959 1960 C. PERIODICALS Alberts, William W. "Business Cycles, Residential Construction Cycles, and the Mortgage Market," Journal of Political Economy. LXX (June, 1962). 342 Alhadeff, D. A. "Credit Controls and Financial Interaediaries," American Econoaic Review. L (September, I960). Burck, Gilbert, and Sanford S. Parker. "The Danger in Mortgage Debt," Fortune, LV (April, 1956). "Business Outlook," Business Week. June 21, 1961. Case, Fred E. "The Use of Junior Mortgages in Real Estate Financing," Journal of Finance. X (March, 1955). . "Why Is Credit Tight?" Savings and Loan Journal. December, 1955. Dailey, Don. "Current Trends," Savings and Loan News, LXXIX-LXXXI (Monthly Issues 1935^13377. _______• "Facts and Figures," Savings and Loan News. LXVII-LXXVTII (Monthly Issues 1946-1957). Daniel, Coldwell, III. "Effects of General Credit Controls on Nonfarm Residential Construction," The Journal of Finance, XV (March, I960). Edwards, Edward E. "Real Estate Economics, A Return to Fundamentals," The Appraisal Journal. April, 1949. Federal Reserve Bank of Kansas City. "Technical Appendix to 'The Export-Local Employment Rela tionship in Metropolitan Areas,'" Monthly Review, March, I960. "FHA Experience with Mortgage Foreclosures and Property Acquisitions," The Mortgage Banker, March, 1963. First National City Bank. Monthly Econonic Letter. March, 1959. Gillies, James. "Industry's Role in Metropolitan Growth: A Public Management Problem." California Management Review. II (Winter, I960). . and Clayton Curtis. "A Note on the Small Mortgage Market," The Journal of Finance. XLV (September, 1959)• 343 Grebler, Leo. "Concerning the Quality of Mortgage CreditThe Mortgage Banker. March, 1963. Gregory, Paul M. "Inperfect Coapetition in the Mortgage Market," Southern Econoaic Journal, X (April, 1944). Guttentag, Jack. "Credit Availability, Interest Rates, and Monetary Policy," The Southern Econonic Journal, XXVI (January, 1960). . "The Short Cycle in Residential Construc- iion, 1946-59," The Anerican Econonic Review, LI (June, 1961). Housing and Hone Finance Agency. "Hone Financing 1949-51— Changes Under Credit Controls," Housing Research, Winter, 1951-1952. "New Hone Price Shifts, 1951-1952; Under Credit Controls as Anended in 1951," Housing Research, March, 1953. Klanan, Saul B. "Two Projections: For Investnent Funds and for Mortgage Banking," The Mertgage Banker. March, 1963. Litterer. Oscar F. "New Developnents in the Resi dential Mortgage Market," Federal Reserve Bank of Minneapolis Monthly Review. Novenber. 15357 Mao, Janes C. T. "The Inpact of Federal Mortgage Insurance Prograns on Ann Arbor's Hone Mortgage Market, 1956," The Journal of Finance. XXII (Septenber, 1958). Martin, Willian McChesney. "Monetary Policy and the Real Estate Market," Connercial and Financial Chronicle, CLXXXII (Decenber 15,“I5537T Murray, Roger F. "Who Will Want to Borrow and How Much Will They Want," The Mortgage Banker. March, 1963. Muth, Richard F. "Interest Rates, Contract Terns, and the Allocation of Mortgage Funds," The Journal of Finance. XVII (March, 1962). 344 O'Leary, James J. "The Effects of Monetary Policies on the Mortgage Market," The Journal of Finance. XII (May, 1958). Saulnier, Raymond Joseph. "Federal Monetary Policy and Its Effect on Home Financing," The Mortgage Banker. June, 1956. Torrance, Charles M. "Gross Flows of Funds through Savings and Loan Associations," The Journal of Finance, XV (May, I960). Wall Street Journal. Pacific Coast edition, February Wendt, Paul F., and Daniel R. Rathbun. "The Role of Government in the San Francisco Bay Area Mortgage Market," The Journal of Finance, December, 1951. Wojnilower, Albert M., and Richard E. Speagle. "The Prime Rate,” Federal Reserve Bank of New York Monthly Review. April. 1&62. Woodlief, Thomas. "Federal Reserve Policies," Commercial and Financial Chronicle. CLXXXIV (July 19, 15) 56) . D. ESSAYS AND ARTICLES IN COLLECTIONS Burns, Arthur F. "Long Cycles in Residential Con struction, " Economic Essays in Honor of Wesley C. Mitchell, ftew York: dolumFla University Press, 15) 35. Distelhorst, Carl F. "Savings and Loan Associations and Mutual Savings Banks," AnaiHFinancial Institutions. Herbert V. Prochnow,editor. New York: Prentice-Hall, Incorporated, 1951. Shaw, Edward S. "Savings and Loan Market Structure and Market Performance," A Study of California State-Licensed Savings an3 Loan Associations. Sacramento: state of California Savings andLoan Commission, 1962 345 Weiner, Arthur M. "Real Estate and the Business Outlook," Proceedings, The Sixteenth Stanford Business Conference, July 22-<£3. 19jj7~ Stanford, California; Stanford University, 1939. E. UNPUBLISHED MATERIALS Balderston, Frederick E. "Revision of the Standards of Applications for New Charters and New Branches in the Savings and Loan Industry of California," February 8, 1963. (Mimeographed.) Bobo, Janes Robert. "An Analysis of the Growth of Savings and Loan Associations in the Ninth Federal Hone Loan Bank District, 1945-1959." Unpublished Ph.D. dissertation, Louisiana State University, 1961. Eastburn, David P. "Real Estate Credit Controls as a Selective Instrunent of Federal Reserve Policy." Unpublished Ph.D. dissertation, University of Pennsylvania, Philadelphia, 1957. Gregory, Paul M. "The Worchester Mortgage Market." Unpublished Ph.D. dissertation, Clark University, Worchester, Massachusetts, 1942. Guttentag, Jack M. "Some Studies of the Post-World War II Residential Construction and Mortgage Markets." Unpublished Ph.D. dissertation, Columbia University, New York, 1958. Johnson, Mathew M. "The Philadelphia Mortgage Market: With Special Emphasis on Residential Mortgages." Unpublished Ph.D. dissertation, University of Pennsylvania, Philadelphia, 1951. Kinnard, William N. "The Mortgage Market in Middle town, Connecticut." Unpublished Ph.D. disserta tion, University of Pennsylvania, Philadelphia, 1956. Phillips, Doris G. "Economic Dualism in Classical Economic Thoughts." A paper read at the Western Economic Association Annual Meeting, San Francisco, August 22, 1963. 346 Saunders, Jerry A. "Impact of a Restrictive Monetary Policy on Los Angeles Metropolitan Savings and Loan Associations.H Unpublished Master's thesis, University of Southern California, Los Angeles, June, 1960. Vernon, Jack Raymond. "Savings and Loan Association Response to Monetary Policy." Unpublished Ph.D. dissertation, Northwestern University, Illinois, 1961. APPENDIX DETAILED ANALYSIS OF SACRAMENTO ECONOMIC BASE DIVERSIFICATION The desirability of a more detailed analysis of changes in the government and manufacturing sectors of the Sacramento economy led to the development of the following paragraphs. Changes in government employment. In 1953 almost 28.0 per cent of total employment in Sacramento was accounted for in the government sector (Appendix Table 1). By 1959 government employment at the local level (city and county) had dropped to 25.0 per cent of the local labor force. Nationally, government employment increased from 11.0 per cent in 1953 to more than 12,0 per cent in 1959. It is interesting to note that, even though the Sacramento Metropolitan Area contains the capital city of the state, a larger number of the labor force were employed at the federal level (22,000) than at the state level (14,943) in 1953. However, by 1959 federal employment had added only 300 employees, whereas employment at the state level had increased by more than 5,200 employees or an increase of almost 35 per cent. Government employment as a percentage of the labor force at the local level in Sacramento had remained relatively the same in 1959 as in 1953. That APPENDIX TABLE 1 FEDERAL, STATE, AND LOCAL GOVERNMENT EMPLOYEES AS A PER CENT OF TOTAL LABOR FORCE, UNITED STATES AND THE SACRAMENTO METROPOLITAN AREA, 1953 AND 1959 United States Sacramento Numbera Per Cent Percentage Increaseb Number Per Cent Percentage Increaseb 1953 Total All Employees 63,815 100.0 144,000 100.0 Government 7,047 11.0 40,081 27.8 Federal 2,385 3.7 22,000 15.3 State 1,129 1.8 14,943 10.4 Local 3,533 5.5 3,138 2.1 1959 Total All Employees 69,394 100.0 8.7 189,000 100.0 31.3 Government 8,487 12.2 20.4 47,191 25.0 17.7 Federal 2,399 3.5 5.9 22,300 11.8 1.4 State 1,518 2.2 34.5 20,147 10.7 34.8 Local 4,570 6.6 29.4 4,744 2.5 51.2 aLabor force data in thousands. b1953-1959. Source: United States Bureau of the Census, Statistical Abstract of the United States (Washington: Government Printing Office, 1958); United States Department of Commerce, County Business Patterns (Washington: Government Printing Office, 1953); State of California, Economic Development Agency, California Statistical Abstract (Sacramento: State Printing Office, 1961- 1962); State of California Department of Employment. 349 350 is, local government employment increased slightly from 2.1 to 2.5 per cent (3,138 to 4,744 employees) of the local labor force. The data reported in Appendix Table 1 indicate that 6.6 per cent of the nation's labor force was employed in local government, whereas only 2.5 per cent of Sacramento's labor force was employed in local government. Employment in industry covered by unemployment insurance. Data reported in the United States Depart ment of Commerce publication, County Business Patterns.1 indicate that employment growth trends have been reported in six out of eight major areas of employment in the nation and in seven out of eight major areas of employment in Sacramento. The major areas of employment are defined as mining; contract construction; manufacturing; transportation and utilities; wholesale trade; retail trade, finance, insurance, and real estate; and services. As reported in Appendix Table 2, increases in Sacramento employment were made in all major areas except transportation and utilities. Similarly, in the ^United States Department of Commerce, County Business Patterns (Washington: Government Printing Office, 19&i, 19S6, 1959). 355 Sacramento was highly specialized as to federal and state government, contract construction, transporta tion and utilities, wholesale and retail trade, finance, insurance and real estate, and services. Local government, mining, and manufacturing were substantially lacking in representation. None of the major areas indicated average or "normal" repre sentation. Consequently, all industries were either highly represented or lacking in representation. As Appendix Table 3 indicates, although industry special ization remained relatively high in 1959, all industries, 2 (continued) greater employment relative to the national norm can be considered as specialized to the locality; on the other hand, those local industries with less employ ment relative to the national norm may be considered to be lacking in representation. For example, the location quotient for the mining industry would be found by substituting the data in the formula: local employment in the mining industry » Q f gross local employment ,, * _, national emnlovmeni in tne mining industry " the mining -------- groa. national' e.ployi.nt------- inta.try For excellent discussions of the diversification and the location quotient concepts see: Federal Reserve Bank of Kansas City, "Technical Appendix to 'The Export-Local Employment Relationship in Metropolitan Areas,"1 Monthly Review, March, I960; James Gillies, "Industry's Role in Meiropolitan Growth: A Public Management Problem," California Management Review. II (Winter, 1960), 38^! APPENDIX TABLE 2 EMPLOYMENT GROWTH IN SELECTED INDUSTRIES, UNITED STATES AND THE SACRAMENTO METROPOLITAN AREA, 1953-1959 United StatesB Sacramento 1953 1959 Percentage Chance 1953 1959 Percentage Chance Totalb 39,866 41,404 3.9 59,541 85,843 44.2 Mining 854 705 (17.3)C 600 634 5.6 Contract Construction 2,380 2,499 5.0 7,291 10,268 40.8 Manufacturing 17,297 16,207 (6.3)C 9,520 20,584 116.2 Transportation and Utilities 2,843 2,921 2.7 6.621 6,392 (3.5)c Wholesale Trade 2,835 3,092 9.1 6,734 7,608 13.0 Retail Trade 7,470 7,744 3.7 16,434 20,188 22.8 Finance, Insurance, and Real Estate 2,060 2,505 21.6 4,034 5,789 43.5 Services 3,965 5,796 56.9 7,579 13,222 74.5 aEmployment data in thousands. bData do not equal totals because the latter include a small number of employees not classified elsewhere. A Negative change. Source: County Business Patterns: based upon workers covered by unemployment insurance. 352 nation gains were also made in all areas other than mining and manufacturing. The total employed in nonagricultural industry on the basis of unemployment insurance coverage in the United States was 39,865,800 in 1953. This non agricultural employment total increased only 3.9 per cent by 1959. Sacramento, in contrast, increased its total nonagricultural employment from 59,541 to 85,843, or 44.2 per cent. By far the largest employ ment growth in Sacramento was in the manufacturing industries. The data indicate an increase of 116.2 per cent in manufacturing employment between 1953 and 1959 in the local area, whereas manufacturing employ ment in the nation declined 6.3 per cent during the same period. In addition to the decline in manufactur ing employment, the number of workers engaged in mining in the nation also showed a relatively large decline of 17.3 per cent. The greatest relative increase in employment in the nation was apparent in service industries. Between 1953 and 1959 employment in the nation's service industries increased almost 57 per cent. Sacramento, on the other hand, showed relatively large increases in all areas except transportation, where a 3.5 per cent decline took place. The sub stantial increase in Sacramento manufacturing employment 353 was accompanied by an increase of 74.5 per cent in service industries and a 43.5 per cent increase in the finance, insurance, and real estate category. Specialization of industry. The structural changes which have taken place in the Sacramento Metropolitan Area's economy between 1950 and 1960 have been significant; however, federal, state, and local government activities continue to be major influences upon the local economy. The defense- oriented industries have assumed an important role in the growth of the Sacramento Metropolitan Area. The data in Appendix Table 3 indicate that all industries, except mining and manufacturing, which were represented in the area in 1953 were highly specialized within the structure of the local economy. The indicated degree of specialization in major industry in the Sacramento Metropolitan Area is shown relative to the nation in 1953 and 1959. o In 1953, on the basis of location quotients, o Location quotients is the term used to define the relationship between location of industry on the basis of local employment relative to the nation. Since the United States is mainly a closed economy on the basis of industrial output, it is assumed that employment in any given industry would be normal for that industry. Those local industries which have APPENDIX TABLE 3 RELATIVE SPECIALIZATION OF INDUSTRIES BY MAJOR CLASSIFICATION, AS INDICATED BY LOCATION QUOTIENTS, SACRAMENTO METROPOLITAN AREA, 1953 AND 1959 Location Quotients Suecialisation 1953 1959 1953 1959 Governaent 2.53 2.04 High High Federal 4.48 3.44 High High State 5.89 4.89 High High Local .39 .38 Very Low Very Low Miningb .47 .44 Very Low Very Low Contract Construction 2.04 2.01 High High Manufacturing .37 • 62 Very Low Low Transportation and Utilities 1.56 1.07 High Average Wholesale Trade 1.59 1.20 High High Retail Trade 1.47 1.27 High High Finance, Insurance, and Real Estate 1.31 1.13 High Average-High Servicep 1.37 1.11 High Average-High aBased upon all employees. bBased upon uneaployaent insurance covered industries. Source: Derived froa data in Statistical Abstract of the United States; County Business Patterns. 1953 and 19$9: California Statistical Abstract: State of California Departaent of Eaployaent. 354 356 except manufacturing, showed relative decline in their weighted representation in the local economy. Trans portation and utilities, finance, insurance, real estate, and services were highly specialized to the area in 1953; however, by 1959 these had declined in relative weight to slightly above average specializa tion. Where manufacturing industry was substantially lacking in representation in 1953, it had increased to within 40 per cent of average representation by 1959. Government activity, although highly specialized to Sacramento in 1959, had shown considerable decrease in its weighted representation— declining from a quotient of 2.53 in 1953 to 2.04 in 1959, or almost a 20 per cent decline. The relative specialization of industry for both 1953 and 1959 in Sacramento, by major industry classification as developed in the Standard Industrial Code Classification System,** is reported in Appendix The Standard Industrial Code Classification System, devised by the United States Bureau of the Budget for the purpose of promoting uniformity and comparability in the presentation of industrial statistical data, consists of twenty-one major groups, 142 industrial groups, and more than 450 detailed industries in the manufacturing sector of our economy. To clarify the definition of industries, the term refers to a group of establishments primarily engaged in the same line or similar lines of economic activity. In the manufacturing industry the classification is 357 Table 4. Within the 1953 manufacturing industry classification the food and kindred products industry was highly represented in the local economy, as indicated by the 2.18 location quotient. By 1959 its degree of specialization (location quotient), indicative of the over-all trend, had declined to 3 (continued) usually based on products made or the processes of manufacture used. Therefore, the principal or primary product made is generally the basis for the classification of an industry. The major industry groups are identified by two-digit numbers, and these groups consist of at least four industrial groups. The industrial groups contain one or more of the four-digit industries and are identified by the first three digits of the four-digit industries of which the group is composed. An example of this is the classification of industries which produces food and kindred products. The numerical code for all major industries which manufacture food and kindred products is SIC 20. The major groups consist of nine industrial groups in this case. One of these groups is the three- digit group which consists of all industries that produce dairy products. Their numerical code is SIC 202. This industrial group consists of about three detailed industries, one of which is the creamery butter industry. Its numerical code is SIC 2021. Another example of this standard method of classifying industries is the major group of Industries which manufactures chemicals and allied products, whose numerical code is SIC 28. The nine industrial groups within the chemical and allied products group, which manufacture soap and related products, have the numerical code SIC 284. And finally, the numerical code of the detailed four digit industries which manufacture cleaning and polishing preparations is SIC 2842. 358 APPENDIX TABLE 4 RELATIVE SPECIALIZATION OF SELECTED INDUSTRIES, AS INDICATED BY LOCATION QUOTIENTS, SACRAMENTO METROPOLITAN AREA, 1953 AND 1959 Location SIC No. Industry Title Quotient i m IMS Total Mining .47 .44 14 Nonmetallic Mining and Quarrying .50 2.19 Total Contract Construction 2.04 2.01 15 General Contractors, Building 2.42 2.42 16 General Contractors, Others 2.21 1.62 17 Special Trade Contractors 1.76 1.94 Total Manufacturing .37 .62 20 Food and Kindred Products 2.18 1.57 23 Apparel and Other Fabric Products .04 .05 24 Lumber and Wood Products .94 .65 25 Furniture and Fixtures .14 .24 27 Printing, Publishing, and Allied Industry .87 .72 28 Chemical and Allied Products .23 .29 32 Stone, Clay, and Glass Products .23 .28 33 Primary Metal Industries .51 .08 34 Fabricated Metal Products .49 .32 35 Machinery (except Electrical) .15 .15 37 Transportation Equipment .21 NA 39 Miscellaneous Manufacturing .24 .26 Total Public Utilities 1.56 1.07 42 Trucking and Warehousing 1.59 1.04 48 Telecommunication 2.61 .08 49 Utilities and Sanitary Services 1.48 .82 Total Wholesale Trade 1.59 1.20 Total Retail Trade 1.47 1,27 52 Building Materials and Farm Equipment 1.40 1.15 53 General Merchandise 1.04 1.19 54 Food 1.46 1.07 55 Automobile Dealers and Service Stations 1.61 1.36 56 Apparel and Accessories 1.30 1.10 57 Furniture, Home Furnishings, and Equipment 2.37 1.99 (Continued) APPENDIX TABLE 4 (CONTINUED) 359 SIC No. Industry Title Location Quotient 58 Eating and Drinking Places 1.87 1.49 59 Miscellaneous Retail Stores 1.16 1.18 Total Finance, Insurance, and Real Estate 1.31 1.13 60 Banking 1.43 1.13 61 Credit Agencies, Other Than Banks .68 .92 62 Security and Commodity Brokers, etc. NA .48 63 Insurance Carriers 1.43 .98 64 Insurance Agents, Brokers, and Service 1.68 1.50 65 Real Estate .88 1.31 66 Combination of Real Estate, Loans, and Law 4.19 2.81 Total Services 1,37 1.11 70 Hotels, Rooming Houses, Camps 1.43 1.11 72 Personnel Services 1.27 1.20 73 Miscellaneous Business Services 1.27 1.19 75 Auto Repair Services and Garages 1.83 1.67 76 Miscellaneous Repair Services 1.49 1.19 78 Motion Pictures .88 .73 79 Amusement and Recreation Services 1.65 1.50 80 Medical and Other Health Services 1.74 1.22 81 Legal Services 1.18 1.00 83 Educational Service, Museums, etc. • 33 .25 86 Nonprofit Membership Organizations 1.33 .93 89 Miscellaneous Services 1.25 1.23 NA— not available. Source: Primary data from County Business Patterns. 1953 and 1959. 1,57, which amounted to almost a 28.0 per cent decline in representation. It is apparent from the data that of all the industries in Sacramento only the non metallic mining and quarrying industry, the special trade contractors in the construction industry, transportation equipment sector of the defense industry general merchandisers in retail trade, miscellaneous retail stores, credit agencies other than banks, and real estate showed minor increases in representation. Only these seven, of the forty-nine industries reported reflected increases in representation. The other forty-two declined in their weighted representation in Sacramento between 1953 and 1959.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Some Effects Of Monetary And Fiscal Policy On The Distribution Of Wealth

PDF

The Role Of The Manager Of The Federal Reserve System Open Market Account: 1951-1961

PDF

A Money Supply Model: Jordan

PDF

Economic Aspects In The Evolution Of The Great Lakes Freighter

PDF

An Econometric Analysis Of The Influence Of Money Supply And And Money Demand Relations In The Determination Of National Income

PDF

Economic Analysis Of Factors Underlying Pricing In The Southern California Tuna Canning Industry

PDF

An Economic Appraisal Of Public Financial Agencies In The Housing Sector Of The United States Economy: Their Origin, Operations, And Contributionssince 1930

PDF

Financing Of Medical Care For The Aged: A Comparative Evaluation Of Compulsory Versus Voluntary Health Insurance

PDF

Competition And Concentration In The Brewing Industry

PDF

Concepts Of Entrepreneurship In Recent Economic Thought

PDF

A Positive Approach To Advertising And Publicity As A Function Of Consumer Preference

PDF

Effects Of Swedish Fiscal Policy And Planning On Economic Growth And Stability, 1930-1964

PDF

An Economic Analysis Of The Factor Market For Pharmacists

PDF

The Industrial Relations System In Sweden With Suggestions Concerning Itsrelevance To An African Nation, Ghana

PDF

Institutional Mortgage Lending In The Los Angeles Metropolitan Area, 1953-1954 And 1957-1958

PDF

An Examination Of Elasticity As An Econometric Guide To Fiscal Policy

PDF

The Effect Of Wealth On Consumption: Empirical Tests For Significance Andlags

PDF

Forecasting Selected Statewide Recreation Requirements

PDF

The Interaction Among Financial Intermediaries In The Money And Capital Markets: A Theoretical And Empirical Study

PDF

The Church Missionary Society, The Imperial Factor, And Yoruba Politics, 1842-1873

Asset Metadata

Creator Tsagris, Basilios Estathios (author)

Core Title The Economic Impact Of Savings And Loan Associations On Residential Construction

Degree Doctor of Philosophy

Degree Program Economics

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

Tag Economics, Finance,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Phillips, E. Bryant (committee chair), DePrano, Michael E. (committee member), Martin, Preston (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-335909

Unique identifier UC11359060

Identifier 6409627.pdf (filename),usctheses-c18-335909 (legacy record id)

Legacy Identifier 6409627.pdf

Dmrecord 335909

Document Type Dissertation

Rights Tsagris, Basilios Estathios

Type texts

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

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

Repository Name University of Southern California Digital Library

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