This is to certify that the thesis entitled THE IMPACT OF LOCAL GOVERNMENT POLICIES ON LAND VALUES AND APPRECIATION presented by George McCl el Ian Johnston has been accepted towards fulfillment of the requirements for Ph.D. Agr. Economics degree in 42$wa [7% Major professor Date May 16, 1980 0-7 639 [Q L *1... ‘ 5 I .1... ‘ mull” (am; ”:1 ' .Sajb 91w: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation record THE IMPACT OF LOCAL GOVERNMENT POLICIES ON LAND VALUES AND APPRECIATION I By George McClellan Johnston A DISSERTATION \ / Submitted\to Michigan State University in partial fulfillment of the requirement for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics 1980 L CHOYZISI' © Copyright by GEORGE MCCLELLAN JOHNSTON 1980 men rel me] [a ABSTRACT THE IMPACT OF LOCAL GOVERNMENT POLICIES ON LAND VALUES AND APPRECIATION By George McClellan Johnston The land conversion process, which creates the shape and environs ment of urban, suburban, and rural areas, raises specific concerns related to agricultural land retention, the cost of public services, and environmental issues associated with urban sprawl. Key performance variables in this process are residential site prices and appreciation of land values over agricultural opportunity costs and site develop- ment costs. There are significant differences in appreciation across metropolitan areas. Appreciation can be considered a measure of economic rent and profit. Local government policies can create economic profit fkn: landowners by restricting land supply for certain uses. Specifically the question is whether differences in site prices and appreciation across metropolitan areas can be explained by zoning, sewer provision and pricing, and property tax policies. These policies, in the aggregate, can restrict land supply and change the pattern of land use. Furthermore, depending upon the variation in policies from one jurisdiction to another, greater supply restricting policies, such as low density zoning, can increase the appreciation and economic profit throughout a metrOpolitan area, without changing the relative prices across jurisdictions. Also, competition for appreciation not only raises housing costs, but also puts great pressure on land use plans. 'lhe IOI the by a; ci (/2 rn _ v m.“ George McClellan Johnston The economic and institutional interdependence of local government jurisdictions is an integral part of examining the hypotheses. The zoning hypothesis stated that the greater the percentage of low density residentially zoned land in the land conversion market, the greater would be appreciation and site prices. The sewer provision hypothesis stated that the greater the percentage of land where sewer provision is controlled or restricted,the greater would be site price and appreciation. It was also noted that under-supply should increase I appreciation,while over-supply would increase price but lower appre- ciation. Other policies such as septic tank regulations could also ameliorate the hypothesis. Furthermore, the greater the percentage of subsidization of sewer services, the greater would be site price and appreciation. Property taxation effects on holding costs and prOperty values were examined but no Specific hypotheses were develOped because of the complexity of the variable. The theoretical model was examined in a cross-sectional regres— sion model, a pooled cross-sectional time series regression model and a comparative case study of Lansing, Kalamazoo, and Jackson, Michigan. Site price and appreciation data from the National Association of Home Builders and the Federal Housing Administration weretbe dependent variables. The independent variables in the econometric analyses included the demand variables analyzed in earlier research, site characteristic variables, and instrumental variables (such as percent all or new homes sewered and the property tax range, a proxy for variation in property taxes across jurisdictions. The comparative case study, which included developer and planner interviews, examined or thl SB ic George McClellan Johnston Operational difficulties with zoning as well as the applicability of the econometric results to detailed metropolitan situations. The weight of the evidence supports the conclusion that zoning, sewer, and tax policies can increase site prices and appreciation. The econometric results demonstrated a consistent statistical signif- icance for agricultural opportunity costs, percent all homes with public sewer, and to a lesser degree, the property tax range. These results varied between data sets and were less stable over time, as tested in the pooled regressions. The comparative case study results 'supported the general hypotheses by noting developer and landowner behavior, but raised questions about theoperational definitions of the variables used in the econometric analyses. Policy implications suggest that preferential agricultural tax policies lead to increased appreciation, as well as, in the aggregate, zoning and sewer supply policies which restrict certain kinds of development. Further quantitative analysis requires better data for both the dependent and instrumental variables. DEDICATION To my parents, George and Mildred Johnston, for their understanding and encouragement. iii gradu super tual help comm ment McBI tuai yea Pet Cla thi Wh: St ACKNOWLEDGMENTS I would like to express my appreciation to those who made my graduate studies rewarding. Allan Schmid as major professor, thesis supervisor, and friend provided the patience, guidance, and intellec- tual rigor to keep me going. My debt Lsprofound. lee Manderscheid and Larry Libby completed the thesis committee and significantly helped in its realization. Les Manderscheid should be on every thesis committee, if he isn't already. Larry Libby's humanity and encourage- ment helped when I reached the walls throughout the process. Glynn McBride's support and willingness to listen is also appreciated. Fellow graduate students contributed to an incredible intellec- tual environment for graduate school. Phil Wandschneider gave me five years of insight while suffering mein the same office. R. Neal Peterson, Preston Pattie, Phil Favero, Compton Chase-Lansdale, and Claude Falgon each in his own way contributed to my growth during this period. I value their friendship and their patience. Pam Christopherson maintained an amazingly positive attitude while typing the difficult final draft. Bruce Mackey provided the graphics. I sincerely thank all of those who made my time at Michigan State one of significant growth and change. iv LIST OF LIST OF Chapter 1.2 II. ‘ III . TABLE OF CONTENTS Page LIS T OF TABLES I I I I I I I l l I I I I I I I ix LIST OF FIGURES . . . . . . . . . . . . . . xiii Chapter I. INTRODUCTION . . . . . . . . . . . . . . 1 II. THE ECONOMICS OF THE RENT AND PROFIT SEEKING SOCIETY . 6 Introduction . . . . . . . 6 Overview of the Situation-Institutions-Behavior- Performance Model . . . . . . . . . . . . 6 Economic Rent and Profit . . . . . . . . . . 9 Instrumental Variables . . . . . . . . . . ll Introduction I I Q I I I I I I I I I O 1 Z Oning O I I I I I I O I ' I I I I I 12 Sewer Provision . . . . . o . . . . . o 16 Property Taxation . . . . . . . . . . . . 22 Slim-Imam, I I I I I I I I I I O I I I I l 25 III. SURVEY AND CRITIQUE OF EMPIRICAL STUDIES OF INTER- URBAN LAND VW I I I I I I I I I I I I I 27 Introduction . . . . . . . . . . . . . . 27 Dependent Variables . . . . . . . . . . . . 29 Geographic Definitions . . . . . . . . 20 Comparison of Dependent Variables ’ . . . . . . 36 Independent Variables and Regression Results of Previous Research . . . . . Issues of Functional Form in the Dependent and Independent Variables . . . . . . . . “2 Comparison of studies by Witte and Ottensmann . . . 9) Elasticity Issues Raised by Witte . . . . . Summary . . . . Chap' IV Chapter IV. SITE PRICE AND APPRECIATION . . . . Introduction . . Site Prices . Introduction . . Data Sources . . Site Size . . Development Costs . . . Agricultural Opportunity Costs . O o o C I n I o I u o a Appreciation . . . . Summary of the Land ConversiOn Process <§ CROSS-SECTIONAL REGRESSION ANALYSIS Introduction . . . Independent Variables . . Site Price Model . Introduction . . . . . Description of the Method Interpreting the Results . . . Appreciation Model . . . . . . . Description of the Method . . . Interpreting the Results . . . . Methods of Sewer Financing and Raw Land Conclusions . . I u I 0 0- l I Prices VI. POOLED CROSS-SECTION AND TIME SERIES ANAIXSIS Introduction . . Test of the Covariance Model Description of the Method Interpreting the Results . . . Analysis of Covariance Model . . Site Price Regression Results Description of the Method . Interpreting the Results . vi Page 100 102 104 106 106 109 114 114 11% 125 ChaI VI: VII Chapter VII. VIII. IX. Appreciation Model . . . . . o . . . . . Description of the Method . . . . . . . Interpreting the Results . . o . . . . . Speculative Analyses of City and Time Dummy Arrays Brief Summary and Conclusion of Pooled Regression AIlatlysj-S I l I I I I I I I I I I I I CONCLUSIONS AND IMPLICATIONS OF THE ECONOMETRIC ANALYSIS . . . . . . . . . . . Introduction I I I I I I I I I I I I I I Geographic Definition Issues . . Preliminary Interpretation of Econometric Results . COMPARATIVE CASE STUDIES: LANSING, KALAMAZOO, AND JACKSON, MICHIGAN . . . . . . . . . . . Introduction I I I I I I I I I I I I I I Developer Behavior Literature . . . . . . . . Land Values and Land Uses . . . . . . . . . Preliminary Analyses and Case Selection . . . Development Costs, Site Size, and Agricultural Opportunity Cost . . . . . . . . . . Towards Defining the Land Conversion Market . . . Instrumental Variables . . . . . . . . . . Zonirlg I I I I I 0 I I I I I O I I I Introduction . . . . . . . . . . . Zoning and Other Maps . . . . . . . . . Developer and Planner Comments on Zoning . . SeWer Provision . . . . . . . . . . o . Property Taxation . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . CONCLUSION I I I I I I I I I O O I I I Introduction I I I I I I I I I I I I I I Summary . . . . . . . . Policy and Research Implications . . . . . . . Page 130 130 131 135 136 139 139 142 146 146 148 149 149 153 160 163 163 163 167 171 173 177 182 182 182 190 Chaptei Append A. B. C. D. BIBLI( Chapter Page APPENDICES Appendix A. Sewer and Water Use Tap Fees For Selected Jurisdictions in Metropolitan Detroit, Michigan, 1969-1974 . . . . 195 B. Estimated Subdivision Development Costs for Selected Site Sizes, Detroit, Michigan, 1963 and 1965—1976 . . 200 C. Calculated Agricultural Opportunity Costs . . . . . 217 D. Comparative Case Study Interview Questions . . . . . 220 E. Example of Development Costs in Lansing, Michigan . . 223 . .Wnrr'mn— — BIBLI WRAPI'IY I I I I I I I I I I I I I I I I I 226 viii Table 10. ll. 12. 13 14 Table 10. 11. 12. 130 14. LIST OF TABLES Percent Allocation Between Property Tax and Sewer User Charges for Sewage Utilities Selected MetrOpolitan Areas or Utility Districts in the United States . . . . . Summary of Empirical Models of Previous Research . o . Site Price and Related Statistics, 1960-1977 . a . . Average Size of Lot, Single Family Homes, FHA 1966-1977, U.S. High Average State and Low Average State, 1966-1977 Average Square Feet of Finished Lot, NAHB . . . . . Estimated Subdivision Development Costs (Dollars), 1963 and 1965-1976. For Lot Size 90' x 193' (1200 sq. feet) and Average 2.8 Lots Per Acre . . . . . . . . . Selected Municipal Development, Utility and Building Fee Ranges, and Average Fees for Associated SMSA, 1978 . . DeveIOpment Cost Estimates Based on 1969 National Association of Home Builders Data . . . . . . . Site Price Appreciation Above Agricultural Opportunity Cost and Development Costs, NAHB Data, 1969 Thirty-Six Cities I I I I I I I I I I I I I I I I Site Price Appreciation Above Agricultural Opportunity Cost and Development Costs, FHA Data, 1969 Thirty-Six Cities 0 . o o o o o o o o o o o o o 0 Comparison of Land Value Appreciation and Market Price of Site for Thirty-Six Cities Common to Both National Association of Home Builders and Federal Housing Administration Data 1969 . . . . . . . . . . Description and Sources of Independent Variables Used in Site Price and Appreciation Models . . . . . . . Site Price Regressions, NAHB, 1969 . . . . . . . Site Price Regressions, FHA, 1969 . . . . . . . Page 37 5o 53 56 58 6O 65 66 67 73 85 88 Table 15. ° 11' 16. 17. 18. 19. . 20. ; 21. 22. 23. 211. 25. 26. 27. 28. 29. 30. 31. 32. Table 15. 16. 17. 18. 1.9- 20. 21. 22. 23. 2L1 25 26 27 28. 29 30. 31. 32. Site Price Regressions For Thirty-Six Cities Common to NW ald FHA For 1969I I I I I I I I I I l I Appreciation Regressions, NAHB, FHA, 1969 . . . . . Appreciation Regressions for Thirty-Six Cities Common to NW and FHA, 1969 I I I I I I I I I I O I Effect of Sewer Charges on Raw Land Prices, 1960 for Seventeen Cities . . . . . . . . . . . . . Test of Covariance Model for Site Price, 1967-1970, Ten Cities: 40 Cases . . . . . . . . . . Site Price Pooled Regressions, 1967—1970, 17 Cities, 4 Years: 68 Cases . . . . . . . . . . . . Site Price Pooled Regressions, 1969-1973, 21 Cities, 5 Years: 105 Cases . . . . . . . . . . . . . Site Price Pooled Regressions, 1964-1974, 10 Cities for 11 Years : 110 Cases I . I . I I I . I I I I Site Price Pooled Regressions, 1967-1974, 43 Cities, 8 Years: 344 Cases . . . . . . . . . . . . . Summary of R2 and R2 for Tables 19, 20, 21, 22 for OLS- Pooled and ANCOVA-XE Models 0 a o a o u o o a Appreciation Covariance Model, 1967—1970, 25 Cities,‘ Years: 100 Cases . . . . . . . . a . . . . Selected Statistics on Selected Michigan Cities, 1970 . Selected Statistics for Lansing, Kalamazoo, and Jackson, Mi ch igan I I I I I I I I I I I I I I I I Site Price and Raw Land Data for Lansing, Kalamazoo, and Jackson, Michigan for Available Years Between 1960 and 199 o n o o o o u c a o I a o a o n . Agricultural Data for Counties in the Lansing, Kalamazoo, and Jackson, Michigan Area 1969, 1974 . . . . . . Land Use in Counties in the Lansing, Kalamazoo, and Jackson, Michigan Areas . . . . . . . . . . . Total Population and Percent Population Change for Lansing, Kalamazoo, and Jackson, Michigan Jurisdictions, 1960’ 1970 o l a a a c n I I a u I o u 0 Selected Statistics for Lansing, Kalmazoo, and Jackson, Michigan, Urbanized Areas 1960 and 1970 . . . . . . X Page 90 95 98 101 110 115 117 119 122 126 132 150 152 154 157 161 162 Table 33. 35- A3. Bl. B2. B3. 35. B6. Table 33. 35. B1. B2. B3. B5. 136. Property Tax Rates and Taxes by Type of Jurisdiction and State Equalized Value for Lansing, Michigan Metropolitan Area Jurisdictions, 1977 . . . . . Property Tax Rates and Taxes by Type of Jurisdiction and State Equalized Value for the Kalamazoo, Michigan Area. ’ 1977 I I I I I I I I I I I I I I Property Tax Rates and Taxes by Type of Jurisdiction and State Equalized Value for the Jackson, Michigan Metropolitan Area, 1977 . . . . . . . Sewer and Water Use and Tap Fees, 1969, For Selected Jurisdictions in Metropolitan Detroit, Michigan . . Sewer and Water Use and Tap Fees, 1972, For Selected Jurisdictions in Metropolitan Detroit, Michigan . . Sewer and Water Use and Tap Fees, 1973, For Selected Jurisdictions in Metropolitan Detroit, Michigan . . Sewer and Water Use and Tap Fees, 1974, For Selected Jurisdictions in Metropolitan Detroit, Michigan . . Estimated Subdivision Development Costs, 1963, For Selected Site Sizes With Assumed Land Cost of $4,000 Per Acre, Detroit, Michigan . . . . . . . . Estimated Subdivision Development Costs, 1963, For Selected Site Sizes With Assumed Land Cost of $5,500 Per Acre, Detroit, Michigan .» . . . . . . . Estimated Subdivision Development Costs, 1963, For Selected Site Sizes With Assumed Land Cost of $7,000 Per Acre, Detroit, Michigan . . o . . a . . Estimated Subdivision Development Costs, 1965, For Selected Site Sizes With Assumed Land Cost of $4, 000 Per Acre, Detroit, Michigan . . . . . . . Estimated Subdivision Development Costs, 1966, For Selected Site Sizes With Assumed Land Cost of $4,000 Per Acre, Detroit, Michigan . . . . . . . . Estimated Subdivision Development Costs, 1967, For Selected Site Sizes With Assumed Land Cost of $4,000 Per Acre, Detroit, Michigan . . . . . . . . xi Page 174 175 176 196 197 198 199 201 202 203 204 205 206 Tabl B7. B8. 39 B11 B12 B1: B11 11} B11 Table Page B7. Estimated Subdivision Development Costs, 1968, For Selected Site Sizes With Assumed Land Cost of $4,000 Per Acre, Detroit, Michigan . . . . . . . . . 207 B8. Estimated Subdivision Development Costs, 1969, For Selected Site Sizes With Assumed Land Cost of $5,000 Per Acre, Detroit, Michigan . . . . . . . . . 208 B9. Estimated Subdivision Development Costs, 1970, For Selected Site Sizes With Assumed Land Cost of $5,000 Per Acre, Detroit, Michigan . . . . . . . . . 209 310. Estimated Subdivision Development Costs, 1970, For Selected Site Sizes With Assumed Land Cost of $5,000 Per Acre, Detroit, Michigan . . . . . . . . . 210 B11. Estimated Subdivision Development Costs, 1971, For Selected Site Sizes With Assumed Land Cost of $5,000 Per Acre, Detroit, Michigan . . . . . . . . . 211 312. Estimated Subdivision Development Costs, 1972, For Selected Site Sizes With Assumed Land Cost of $5,000 Per Acre, Detroit, Michigan . . . . . . . . . 212 B13. Estimated Subdivision Development Costs, 1973, For Selected Site Sizes With Assumed Land Cost of $6,000 Per Acre, Detroit, Michigan . . . . . . . . . 213 314. Estimated Subdivision Development Costs, 1974, For Selected Site Sizes With Assumed Land Cost of $6,000 Per Acre, Detroit, Michigan. . .. . . . . . . . 214 B15. Estimated Subdivision Development Costs, 1975, For Selected Site Sizes With Assumed Land Cost of $6,000 Per Acre, Detroit, Michigan . . . . . . . . . 215 B16. Estimated Subdivision Development Costs, 1976, For Selected Site Sizes With Assumed Land Cost of $6,000 Per Acre, Detroit, Michigan . . . . . . . . . 216 C1. Calculated Agricultural Opportunity Costs and Value of Agricultural Land and Buildings, 1969 . . . . . . 218 El. Categorization of Typical Development Cost Breakdown for Selected Subdivisions in Lansing, Michigan . . . 224 Figure LIST OF FIGURES Figure Page 1. Conceptual Framework of the Situation—Institutions— Behavior-Perforlrlance Model o o o o I o o a I a n 7 2. Development Costs With the Supply of Sites Fixed . . . . 77 3. Development Costs as Determining Site Supply . . . . . 77 4. Comparison of Time Dummy Regression Coefficients and New Home Purchase Mortgage Rates . . . . . . . . . . 137 xiii envi: Eggp andl not cost resi usin is w bee visi suck and higl CHAPTER I INTRODUCTION There are many facets to the process which creates the shape and environment of urban and surrounding suburban and rural areas. Schmid's Converting Land From Rural to Urban Uses noted that, "There is a large and growing residual land value contributing to high lot prices which is not explained by agricultural opportunity costs, lot size, improvement costs, or general inflation." (1968, p.12). This thesis reexamines this residual land value, or land value appreciation, as well as site prices, using the economic theory of rent and profit. Specifically the question is whether differences in site prices and appreciation across cities can be explained by such local government policies as zoning, sewer pro- vision and property taxation. Understanding the role of these policies in the land conversion process should also enable a better grasp of such issues as agricultural land retention, the cost of public services, and other environmental issues associated with urban sprawl. A recent Department of Housing and Urban Development report highlights the timeliness of the issues examined in this research. Much of the increase in housing costs is directly attributable to a steady rise in the cost of the serviced site. A survey by the Urban Land Institute of developer members in seven metropolitan areas found an average increase in urban land prices between 1970 and the spring of 1974 of 100 percent. This is an average annual rate of increase of 20-30 percent for the period, compared to an increase of 8-10 percent between 1958 and 1970. The Department of Agriculture found that the average value per acre of farm land--a prime source of developable lots--had almost tripled between 1967 and 1977. Nationally, the developed lot now accounts for about 20 percent of the cost of a typical single-family house with FHA mortgage insurance, compared to about 15 percent in 1960. In areas with stringent land use regulations, ratios of 30 percent are not uncommon for conventionally financed development. Discounting inflation, consumers are getting less housing for their dollar because they are paying proportionately more for the site. There are three major reasons for this increase in the cost of sites: (1) Constraints in the supply of developable lands; (2) High site development costs; and (3) Procedural delays. In many areas the supply of developable land has been constrained in part by limitations in the capacity of public facilities--especia11y water and sewer--and by restrictions on the use of land through zoning and related controls. Rapid increases in site development costs have been caused by higher governmental standards and fees. Procedural delays have resulted from the proliferation of governmental regulations affecting land development. (HUD, May,l978, p.13) The interdependencies between local government jurisdictions on one part of a land conversion market with decisions taken by other local government jurisdictions can affect the land conversion process and site price and appreciation, in particular. As Clawson noted, "the use and value of any tract or parcel of land Within a metropolis is affected more by the use and value of other tracts or parcels of land than it is by what takes place on the tract itself."(197l, p.174) Much of the focus of research to date has been on the implications of various local government policies on a particular piece of land or category of use. The unit of observation of this research is the land conversion market of non-residential to residential uses, in the aggregate, across cities. to t incl the use. diff- and, am: ecom follc urbar. sewer of 1a site. Press uThan conve; inter. the be Varial In add recent The land conversion process tracks land from active farm value to the price paid for a residential lot. The steps in this process include the speculative price paid to the farmer, the price paid by the subdivider, and the costs of developing the lot for residential use. The land conversion market would, therefore, reflect land in different stages of development. The land use pattern would be mixed and, generally, on the fringe of urban areas. The role of local government policies in explaining land value appreciation and site price across cities is explored in a model using economic rent and profit theory. This is presented in Chapter II. This model specifically accounts for interdependencies between policies followed by local governments which affect the supply of land for urban residential uses. It is argued that local government zoning, sewer provision, and property tax policies which restrict the supply of land for residential uses can create economic profit and increase land value appreciation and, hence, the prices paid for a residential site. The competition for this economic profit can also create pressure on local development plans which also affect the shape of urban areas and the monetary and non monetary cost of the land conversion process. Chapter III surveys and critiques earlier empirical research on interdurban land values. These econometric studies are reviewed on the basis of the form and geographic definition of the dependent variables and the kinds and definition of the independent variables. In addition, the issues raised and techniques used by the two most recent studies are compared. resear< site p in ear price Associ apprec develc compaa the t1 land 1 Price Profi regre alizi indej and : The- mode ores SEWe Sim fon sta Chapter IV develops the dependent variables used in this research; site price and appreciation. In the initial discussion of site prices across cities, the most commonly used dependent variable in earlier research, a comparison is made of the two sources of site price data, the Federal Housing Administration and the National Association of Home Builders. Next, the data and calculation of appreciation are reviewed. This includes data on site size, development costs, and agricultural opportunity costs. After a brief comparison of site price and appreciation data and calculations using the two data sources, the chapter concludes with a summary of the land conversion process which reestablishes the importance of site price and appreciation as operational measures of economic rent and profit. Chapter V is the first of three chapters on the econometric ‘regression research of this thesis. This chapter begins by operation- alizing the independent variables. There are three groups of independent variables: demand variables, site characteristic variables, and instrumental variables designed to test the theoretical model. The first section concludes with a formal statement of the operational models for site price and appreciation. The rest of the chapter uses cross—sectional regression analysis of these models. A short test of sewer financing data is also included. Chapter VI is a pooled cross—sectional and time series regres- sion analysis. This model adds a time dimension and applies various forms of a covariance model. This chapter is designed to examine the stability of cross-sectional relationships over time using different inde avai sect and anal dev. inc. per the Pro ec< gel £01 811 Re re independent variables over various time periods, depending upon data availability. Chapter VII summarizes and analyzes both the cross- sectional and pooled cross-sectional, time series regression chapters. Chapter VIII is a comparative case study of Lansing, Kalamazoo, and Jackson, Michigan. Data and interpretations of the econometric analysis are compared to primary and secondary data and information developed on zoning, sewer provision, and property taxation. This included interviews with planners and developers in these cities. In particular, zoning is analyzed in much more detail than possible in the econometric chapters. Furthermore this chapter deals with the problems which arise from the use of national data when applied to case studies. Chapter IX, the conclusion, will analyze the results of both the econometric and comparative case study chapters with respect to the general model proposed and the specific hypotheses presented. The focus of the chapter is to respond to the issue of what people should support in terms of instrumental government policies if they want to keep prices and appreciation down. This includes both policy and research suggestions. To summarize, the next chapter will develop a theoretical model and will be sequentially followed by a literature review, development of the dependent variables and further clarification of the problem, definition of the independent variables and a cross-sectional regres- sion model, a pooled cross-sectional, time series regression model, :onclusions drawn from the regression results, a comparative case study, and, finally, the summary conclusions of the thesis. in thi only h framew theore econon model taxatf All 81 ment . I? Perfc Cones dism CHAPTER II THE ECONOMICS OF THE RENT AND PROFIT SEEKING SOCIETY Introduction This chapter conceptualizes the problems and issues of interest in this research. The conceptual framework, per se, will be described only briefly. The nature of economic rent and profit within this framework will then be described. The next step presents the theoretical linkage between each of the instrumental variables and economic rent and profit. Effectively;this chapter presents a causal nodel of the relationship between zoning, sewer provision, and taxation, the instrumental variables, and economic rent and profit. All subsequent chapters relate to the operationalization and measure- nent of this model.1 Overview of the Situation—Institutions~Behavior-Performance Model This research will apply the situation-institutions-behavior- performance model for the analysis of community issues.2 1The data or information system model, which includes the :onceptualization, operationalization, and measurement steps, is .iscussed in Bonnen (1975). 2This model is elaborated in Schmid (1979). FIGU si‘ 1c: as: si BI ac FIGURE 1: CONCEPTUAL FRAMEWORK OF THE SITUATION-INSTITUTIONS- EEHAVIOR-EERFORMANCE MODEL Situation I InstitutionsI Behavior1 Performance/ IBehavior lPerformance1 Institutions Political H Instrumental <—-—-): Performance Institutions ‘ Institutions 1' e.g: form of e.g: zoning, e.g: land value government and sewer provision, appreciation, site decision rules property taxes price, density This figure can be described briefly as follows. Defining the situation entails describing the varieties of interdependence, histor- ical setting, and degree of conflict of harmony. If scarcity is assumed, there is interdependence. Property rights define whose interests are to count. Particular relationships are defined via the situation and constrained by property rights. These property rights are the institutions which were defined by John R. Commons as "collective action in control, liberation and expansion of individual action" (1950, p.21) . For example, rents and opportunities for gain are distributed among different groups according to the rules of the game. Inelastic supply of land means that market competition does not prevent returns Or profits above costs of production (opportunity cost). If the Performance desired by the individuals wanting to get the land value appreciation is achieved, then others must bear some costs in terms of land use patterns in addition to the cost of housing. ind 8.1‘6 thc wh< the Z0] of st 9X po wh Because scarcity and interdependence are assumed there are individuals and groups who want different performances than what they are now getting. These interests may be frustrated by the ability of those wishing to manipulate the process to get profits. Regardless of who wants what, this indeed seems to be the result. The capture of these profits is a type of pecuniary externality which in the case of zoning, sewer and taxes are politically created. The political institutions consist of the form of government and the decision rules which affect who gets to choose policy and use of resources. The behavior of the individuals or groups within this structure affects the performance of the local government; for example, the specifics of zoning, sewer provision and taxes. These policies then consist of the instrumental institutions or structure which affect the behavior of the participants in the land conversion market and what will be considered here as the categories of perform- ance of interest; land value appreciation and site price. The presentation of the model will begin with defining the situation and final performance, site price and appreciation. From final performance the instrumental institutions and then political institutions will be discussed. The rationale for this approach is that while the political rules which define who has access are important in determining performance, that importance is reflected in the choice and administration of the instrumental institutions. If he instrumental institutions cannot be shown to influence the hosen performance variables, then the role of political institutions n performance will be speculative. of Mu C3. re f1 Economic Rent and Profit Rents are defined as returns above costs of production resulting from natural limitations in supply. The supply of a particular variety of land is inelastic. By way of contrast, returns arising from non-natural limitations are termed excess profits or quasi rents. Normal profits, on the other hand, include the payments necessary to draw forth the required entrepreneurial and capital resources. These definitions are not without argument in economics and the literature is rather huge (Keiper, et.al., 1961, Gaffney, 1961; Currie, Murphy and Schmitz, 1971, Lackman, 1976; Lackman, 1977; Edel, 1978). One criticism is that this definition may ignore some of the nonpecuniary rewards the factor owner may receive (Currie, Murphy and Schmitz, 1971). ‘ The implication of economic rent is that, unlike profits, rent cannot be competed away. Factor ownership controls who gets the ents. Monopoly conditions which lead to profits such as limiting firm entry can be dealt with by increasing competition. Land on the urban-rural fringe appreciates without any produc— tive effort on the part of owners of the land. There are, therefore, anentives for landOWners to compete by favoring increased develop- ment, in general, and development on their land specifically. This outs pressure on any development control plans which may exist. The results may be associated with urban environmental characteristics mder the rubric of sprawl: expensive public services, mixed uncomplementary" land uses and a leapfrogging land use pattern. pre: be ind is in‘ CO re ci 10 Land value appreciation may represent something more than the present value of future rents. Monopoly returns or profits could also be a portion of this appreciation. Whether monopoly profits are indeed a component of appreciation is the central issue of this dissertation. The profit component of land value appreciation could arise from either local government or private supply restrictions. This research is primarily interested in public supply restrictions. Examples include zoning land for large lots, controlling (limiting) sewer provision and public tax and pricing policies. Overt private supply restrictions,which could lead to monopoly conditions such as ownership of land by a few people, are not readily evident. However, there do appear to be variations between cities in the degree of concentration of ownership (Markusen and Scheffman, 1977). On the other hand, private supply restrictions exist if the owner of a resource naturally limited in supply, acting independently, withholds land from the market in eXpectation of increasing prices. This reservation price functions as a monopoly but, as Breimyer (1978) noted, it is a monopoly without monopolists. This reservation price could vary across cities because of local government policy effects on landowner expectation. Operationalizing economic rent and profit in this research will involve using site price and appreciation. Site price captures more than economic rent and profit. It also includes development costs, Lgricultural Opportunity costs and some variation in site size. [uch of the earlier research used this variable and its use in this pe1 of re: 1‘63 (ii: an. se re ti im 19 Ce is se or me 11 research allows some continuity with that research, as well as permitting a check and a contrast with appreciation. The calculation of appreciation, while rough, is designed to eliminate factors not related to economic rent and profit. Differentiation between economic rent and profit will be attempted. Instrumental Variables Introduction Instrumental variables are variables which are subject to some form of political decision and this can be changed by the instrumental variabkrh Zoning was selected because it has been considered the most direct policy instrument in controlling land use. Sewer provision and pricing were selected as examples of the effect of various public service policies. Also increasing water pollution concern and regulation indicatesean enlarged role for sewer policy. TranSporta- tion was not investigated because it was believed that much of the important part of the transportation system is in place (Tabors, et a1, 1976). This may not be the case if mass transit systems expand. Zertainly the energy component will be different. Finally, taxation ts explored because of its interactions with both zoning and public service provision rather than any strong belief that either the theory >r research would give much light to the conceptual, operational, or easurement problems inherent in taxation. §9_i§g within permit requir into n agricu olassi such a distri zones design jurisd pressu aSSoci and St Pensat market into E Such a Smut and SE ecOnon P.64: of in 0f in1 done 1 12 20 ing Zoning involves the designation of specific land use districts within which various regulations and restrictions apply, such as permitted uses, proportion, and size of lot, maximum height and bulk ' requirements and population density limits. Districts can be classified into numerous categories: residential, business, industrial, agricultural, recreational, unrestricted, etc. The residential classification can be further broken down into various categories such as single family, multiple family, and apartment building districts. Zoning could be exclusive, allowing one use, or cumulative zones which allow the previously defined uses in addition to its own designation. Implementation and form of these powers can vary between jurisdictions within a state and within a metropolitan area. It is argued that the competition for economic rent places great pressure on the development plans of communities. Zoning has long been *associated with corruption arising from rent competition (Clawson, Held, and Stoddard, 1960; McCahill, 1973).:3 Economists, however, have :3Externality is conceived by some economists narrowly as uncom- pensated benefits or costs not taken care of by the operation of the market or an effect of one economic agent on another which is not taken into account by the first agent when he decides on his actions. Others such as Samuels defined it as follows: "Externalities comprise the substance of coercion, namely, the injuries and benefits, the costs and gains, visited upon others through the exercise of choice by each economic actor and by the total number of economic actors."(Samuels,l972 p.64) This definition makes externality synonymous with the concept of interdependence. Since this research attempts to Specify the nature of interdependencies more carefully, the notion of externality will be used only as a reference to the usage of the term in the review of work done by some economists. general framewc effect Jacksor fiscal when 01 zoning by ext low pr zoning commur defini eithei Which The i Acoor the h the: cons: for- gain the 13 generally been concerned with zoning from a fiscal or externality framework. Research in this area has often shown zoning to have no effect on the externalities being tracked (Crecine, Davis, and Jackson, 1967; Maser, Riker, and Rosett, 1977). Externality and fiscal zoning are segregated as follows. Externality zoning arises when one person's land use has an effect on neighboring land. Fiscal zoning usually implies not solely the separation of land uses implied by externality zoning, but an orientation to different goals such as low property taxes or high property taxpayer residents. Exolusionary V zoning designed supposedly to keep certain land uses out of a ° community could arise, given the common understanding of the first two definitions, from either of the above OI“be the basis of motivation for either. Zoning has been considered most effective in inhibiting changes which have adverse effects on other users in well established areas. The initiator of change is opposed by those who stand to suffer losses. According to some, zoning has also provided the device for protecting the homogeneous, single family suburbs (Babcock, 1966). Zoning, however, has not been considered to be successfully administered to control the speed, direction and final character of the land conversion process (Clawson, 1971). It has not been considered effective in keeping out land uses incompatible with plans for the development of new suburban areas. Those who compete for the gains from land uses other than those permitted will attempt to change the zoning. Nevertheless, there are suburbs where low density resi offi land of ] the tim any inc wil of on So th 14 residential zoning is strongly supported by residents and public officials (Babcock, 1966). The relationship of zoning to land values and land uses in the land conversion market is affected by the degree to which the zoning of political jurisdictions in the market, in the aggregate, affect the supply of land for different uses. For example, if one jurisdic- tion allows diversity of uses while nine jurisdictions try to inhibit any development other than low density residential uses then the increased competition for the areas available other than low density will drive those prices up while the price of the zoned low density land, which is over-supplied, will depend upon the price elasticity of demand for that particular use, expectations, etc. Minimum lot size requirements can serve as an example of what can be expected from variables designed to affect the size of lots. Some jurisdictions in a land conversion market might contribute to the withholding of land from the market. This reduces supply and raises prices above costs of production. For example, some communities purposely zone available sites only for large lots, hoping thus to reduce government costs in relation to tax revenues. (Mills and Dates, 1975). The process puts a premium on areas that are open to small lot evelopment or for multi-family units. If too little land is zoned or small lot development, there could be leapfrogging developments nd a leapfrogging pattern of land acquisition. Thus large lot zoning ould not only use up more land and at a lower density, but it would lso contribute to supply restrictions for other residential construc— ion. This would increase the appreciation on high density zoned land and con] couJ pric onI lan re] cox det of: me- fo Wi l5 and decrease the overall density of land. The large lots overzoned could be a differentiated product with higher quality features which could potentially be sold by the developer to customers at a higher price. Ohls, Weisburg and White in an article on "The Effect of Zoning on Land Value" argued that zoning could either raise or lower aggregate land values. On one hand they argue: Fiscal zoning enables suppliers of land to act in discriminating monopolist fashion. They can charge a high price in that submarket where demand is inelastic by using zoning to restrict supply and channel the left—over supply into the market with elastic demand. Furthermore, aggregate land value can be increased even when the two submarkets are interrelated, if demand for the restricted use (in this case apartments) is sufficiently inelastic (1974. p.162). They also argue that zoning could lower land values depending on the relative elasticities of different residential types. This is ‘consistent with rent theory and indicates the importance of demand in determining economic rent. Several empirical studies provide-data and analysis of zoning effects. The Regional Plan Association's (RPA) study of zoning in the metropolitan New York area is reported in Gold and Davidoffs' report for the President's Committee on Urban Housing (1967) and combined with other research in Sagalyn and Sternliebs' Zoning and Housing Josts: The Impact of Land Use Controls on Housing Pribe (1971). [he RPA report indicated the following trends: 1) Of the vacant land -n the region 75.7 percent is zoned and 90.7 percent of that for ‘esidential uses and 99.2 percent of that for single family residential mes hence 8 percent for multiple family housing, 2) "We now see that 90 per: lots 0: zoned : p343) and St same 8 lot si lot si counti correl requi] contn to ca; have ‘ great land Mice 5i? If H Cana tho I, 16 90 percent of the area zoned for single family housing is zoned for lots of one-quarter acre or larger, while tw0—thirds of the area is zoned for half acre or larger lots" (Sagalyn and Sternlieb, 1971, p.343), and 3) there has been a strong trend to up zone or to increase minimum lot size as an area begins to undergo urbanization. Sagalyn and Sternlieb compared the RPA data with a later survey and found the same strong preference for large lot zoning. Sagalyn and Sternlieb provide the basis for isolating minimum lot size as a crucial variable in zoning. Their study found minimum lot size statistically significant in explaining housing costs across counties in New Jersey. In addition, minimum lot size was highly correlated with front footage requirements. So while minimum frontage requirements, population density limits and other specific zoning controls may be important, minimum lot size requirements seem likely to capture the intent of the zoning ordinance. It can, therefore, be hypothesized that low density zoning will have the following relationship to site price and appreciation: The greater the percentage of low density residentially zoned land in the land conversion market, the greater will be the appreciation and site price in that market. Sewer Provision Local government practices with respect to public sanitary sewers can also influence site price, appreciation and land use. There are two key aspects. mad men If 3 0th are. per int dic url sen be de ex bu de —————arr-uwm-~w a—~—\ 1? One is the level of provision. If the supply of sewers is restricted or controlled, the supply of building sites, given require- ments for adequate sewers, will be restricted in that jurisdiction. If policies responsible for controlling sewer supply such as, among others: 1) sewer moratoria, 2) refusal to sewer, or 3) a small land area included in a Service Policy Area are in effect in a large percentage of the land conversion market, developers could be forced into a leapfrogging pattern of land acquisition by moving to juris- dictions which are less restrictive and, perhaps, further from the urban area (Tabors, et.al., 1976). This also increases the price of sewered land, which is in limited supply. Use of septic tanks is becoming more difficult in many areas, primarily because of health department policies (Downing, 1972). The process could be described as follows. If seWers have been extended to large areas of undeveloped land, developers are likely to buy and construct on large tracts where land is cheaper. The resultant development will be a low gross density and probably a low net density. The rate of development and infilling will depend upon general and relative demand. However, if sewer provision is still further increased because of demands on other areas of the metropolitan area and fringe, complete infilling might never occur. The infilling would also relate to other issues including the zoning by a local jurisdiction. If a suburban jurisdiction is settled with a certain more or less homo- geneous group, the zoning may reflect a desire to maintain that homogeneity. This may not be the case where communities on the fringe encourage development. A moderate level of sewer provision might not --.~.__ haw ISS' gre mod low the SE of 18 have the same result. Supply of sewered lots would be somewhat restricted, therefore the price of a sewered lot is likely to be greater. Demand will determine the price and density. If demand is moderate, prices will not be as great, but density is likely to be lower, though not as low as with over-supply of sewers. Finally, when much less sewer system is provided than is demanded, the land which is sewered will be highly priced. If the price is too great, developers may find it reasonable to find cheaper land much further from the urban area where other alternatives such as package sewer plants or septic tanks provide a reasonable financial alternative. The general hypothesis is as follows: The greater the percentage of land in the land conversion market where sewer provision is controlled or restricted, the greater will be the site price and appreciation. However, it is necessary to differentiate between restrictions associated with too much or too little sewer provision. Restricted supply, too little, should increase appreciation while over-supplied would increase price because of the sewer component of development cost, but lower appreciation. Another source of appreciation above agricultural opportunity costs occurs because sewers are provided for less than they cost. A proportion of the value of land is based on the availability of sewers. To the extent that the sewers are limited in supply and made available without or below costs, their value becomes capitalized into the value of the land. For example, sewer service may be provided to new areas at the same price as the central city area even though the cost may be highc sewe owne: of a a re the cont Perc marl POL 8817 con Pro Via wit alt 19 higher. It is the owner of the sewered land who benefits because the sewer services are capitalized into the land values but cost the owner very little. Schmid noted that, "The asset appreciation reflecting the value of amenities provided in limited supply at less than cost appears as a rent from the developers point of view, but is monopoly profit from the point of view of the whole economy, in that it results from a contrived rather than natural restriction of supply." (1968, p.37) Clawson elaborated on the issue as follows: To the extent that the house purchaser evades any of the costs of public services to his property, the raw land price will be higher than if he had to pay them. The house purchaser will have gained little or nothing by evasion of these costs, nor will the builder have gained. Virtually all of the gain from costs evaded by the purchaser will have passed on to the owner of the raw land. To the extent that the house purchaser does not pay all the costs associated with his property, some other taxpayers will have to pay them. (1971, p.162). Therefore, the hypothesis is as follows: The greater the )ercentage of subsidization of sewer services in a land conversion Iarket, the greater will be the site price and appreciation. One caution will be mentioned at this point. LOcal government olicy on septic tanks may affect sewer policies. For example, if eptic tanks are restricted, it may either increase density in the and conversion market or perhaps extend the boundaries of the land onversion market to areas where septic tanks are allowed. Subsidized rovision of sewers might increase density if septic tanks are not a Lable alternative. In addition, the over-building of sewers combined -th the subsidization issue should also relate to septic tank or any .ternative treatment system by making the use of sewers more attractive. l broa CODI sys1 fees P102 the age: use An lit: 801 20 Operationalization of this issue is complex. There are three broad categories of revenues used to finance sewage: Service charges, connection fees, and property value taxes. Service charges are periodical charges to the users of a sewer system, or presumptive evidence of such use. In contrast, connection fees are one time expenditures paid when the user begins service, while property value taxes may be either one time or periodic and vary with the assessed value of the property. Property taxes may be levied against users and non-users. These categories can be further characterized by the type of revenue base used: 1) general revenues, 2) special ad valorem assessments, 3) lot size and frontage assessments, 4) flat rates and modified flat rates charge, 5) user charges based on water use and, 6) user provided facilities. (Tabors, et.al., 1976). ‘ Very limited data is available across metropolitan areas on the use of sewer charges. Some data for 1960 is presented in Table 1. Appendix A contains data for some jurisdictions in metropolitan Detroit. The issues which arise in the pricing of a joint impact good are varied}+ One effect noted in a report entitled Interceptor Sewers and Suburban Sprawl: The Impact of Construction Grants on Residential Land Use was that connection fees forced developers and the local :ommunity to attempt to reach their population projections, "without Joint impact goods are goods which enter tw0 or more persons' .tility irreducibly. The marginal cost of another user is zero over ome range. See Schmid (1979, pgs.70-87). Metrn or U' Bost Chic: Cine Clevl Detr Milw Sour 21 TEES—1 PERCENT ALLOCATION BETWEEN PROPERTY TAX AND SEWER CHARGES FOR SEWAGE UTILITIES. SELECTED METROPOLITAN AREAS OR UTILITY DISTRICTS IN THE UNITED STATES, 1960. Metropolitan Area User Charges Property Tax or Utility District (%) (%) Boston 0 100 Chicago O 100 Cincinnati 100 0 Cleveland 78 22 Detroit 100 0 Los Angeles Sanitation District 0 100 Milwaukee O 100 New York 78 22 Philadelphia 100 0 Pittsburgh 100 0 Portland 57.7 42.3 San Francisco 0 100 East Bay Municipal 60.7 39.3 Utility District Toledo 100 0 Washington, D.C. 100 0 Buffalo 37.5 62.5 Green Bay, Wis. O 100 Madison, Wis. 86 . 14 Rahway Valley, N.J. o 100 St. Louis 91.3 8.7 Source: Downing, 1969, p.145. new COIlI Any sys value i T able to They wi relatic 22 new connections, the bonds the area has sold cannot be paid off." The sponsors of Harkey Creek Interceptor System in Tulsa, Oklahoma financed their system by selling developers debentures which could later be applied only against connection fees, and were not convertible into cash. This places great pressure on developers to sell lots as quickly as possible after the system has been installed. (Council on Environmental Quality, 1974, p. 77). Any system of financing has distributional as well as density or land value implications. The econometric and comparative case study analysis will not be able to test for the complexities of the inter-actions described above. They will, however, attempt to test the linkage and direction of the relationship between land values and sewer policies. Propgrty Taxation There is considerable variation in the extent to which the property tax is used among urban areas (Netzer, in Beaton, ed., 1974, .159). In addition, there are wide disparities within an SMSA on the Level and adequacy of the tax base and the level of property taxation. fins can lead to what Gaffney (1975) referred to as municipal or fiscal Ierchantilism, competition for a better tax base. If Political units .re sufficiently small, the location of particular high property tax .ncome generating business can have a significant effect on municipal 'inances. The competition for high paying properties can have an effect n the aggregate land conversion process regardless of whether the actic works or not. Indeed, James and Windsor (1976) argue that their esearch shows that zoning patterns do not relate to the fiscal ideal hey established for various types of communities. affe rese If t the proI apyn inc1 the be pro hel res ler 8.112 pe: V8. 31] Va bt 23 From the point of view of the landowner, the property tax can affect the price of land through the present value (holding costs) and reservation price and, therefore, the supply of land at any given time. If the reservation price exceeds the market price, the owner can hold the land for further gains, though prospects may be uncertain. High property taxes can make it unprofitable to invest in land to hold for appreciation. "It also should be noted that while a property tax increase can lower prices to lot consumers,it may not necessarily reduce the amount of appreciation above farm value. Since, if the property tax rate increase is general, the price of agricultural land could also be expected to fall." (Schmid, 1968). Therefore, the impact of a property tax increase is to reduce the reservation price of land being held for future gains. Lowering tax rates for agricultural land will result in raising present values and reservation prices for fringe land and could be expected to increase landowner gains. The research by Schwartz and Hansen (1975) on preferential taxation supports the analysis that expectations of gains by landowners are greater than the perceived tax benefits of such a policy. Deferred taxation as well as use value taxation also encourage land withholding. From the point of View of the property tax effect on housing values, it would be expected that high property taxes lower home values and, therefore, site values. The true value of a home includes site alue plus construction costs which is a function of operating costs. erefore, higher property taxes reduce the derived demand for homes ut may decrease the costs of development. of land effect proper or def increa site p values correl educat averaé 0f the indie; subure eXamp: becau: Fame: becau; incom abili Pr0pe Donn Duh. 24 To summarize, high property taxes should increase holding costs of landowners and increase the supply of land for urban uses. The effect on appreciation depends upon the extent to which the high property taxes are applied to agriculture. If there is preferential or deferred taxation for agriculture, while other property taxes increase, then appreciation could become greater. Operationalizing the relationship between the property tax and site price and appreciation is complicated by variations in assessed values and property tax rates across jurisdictions as well as the high correlation between the average per capita property tax and income, education, and public service variables. However, the variance in the average per capita property tax between cities might also be indicative of the ability of landowners to withhold land. A larger variance might indicate less pressure on landowners because of lower tax rates in the suburbs, assuming that the higher rates are in the central cities. Nevertheless, other factors cloud the issue still further. For example, income levels are important in calculating the effect of taxes because of the property tax write-offs on the federal income tax. Farmers often pay a proportionately large share of property taxes because of land and other capital intensive investments. When the income of farmers increase, as it did in 1973, 1974, and 1975, their ability to withhold and even buy land increases. Furthermore, the property tax and, indeed, the entire tax structure are related to the population growth rate and the provision of public services in the area. With this degree of conceptual and operational complexity, the research involving the property tax is exploratory and potential results >erhaps directional rather than definitive. for ' of e tan whic cert that the beca cons sup; the] the use in: tin of; Fri als SPr Sic 25 M This chapter has established the theoretical conceptualization for the relationship between the site price and appreciation measures of economic rent and profit and local government zoning, sewer, and tax policies. It is argued, at the general level, that the degree to which these local government policies restrict the supply of land for certain demanded urban uses will affect the price of the land sold and that part of the appreciation in land value is attributable to monopoly profits. Specifically, if zoning for large minimum lots is extensive in the land conversion market, the price of smaller lots might be greater because of limited supply. On the other hand, large lots might be considered a differentiated product and sell for a higher price than supply indicates because of potentially inelastic demand. Given that there is likely to be variation in the attitudes of jurisdictions in he same land conversion market, the excess demand for a certain land se limited in one jurisdiction will move to another, less restrictive, 'urisdiction which raises the price there. The result of this institu- ional interdependence is to not only affect the speed and direction f residential location and growth, but, perhaps, keep the relative rice differences between jurisdictions essentially unaltered. The egree of variation in policies followed across jurisdictions should lso affect relative prices. If supply restricting policies are wide- Pread, prices should be higher than when such policies are rare. Sewer provision and pricing could be similarly analyzed. Provi— 'On of sewered and zoned land for different uses at supply levels appr sion simi Rest of 5 eff. wit the acr and hr: of rie la in 80 26 appropriate to demand results in moderate prices. Over or under-provi- sion of sewered and zoned land in proportion to demand can result in similar leapfrog land use patterns but different price structures. Restricted sewer supply should lead to higher prices and over-supply of sewers should lead to lower prices. A complete examination of the property tax is beyond the scope of this research. Property taxation in the land conversion process affects both demand for more services and the ability of landowners to withhold land from the market. Property taxes are quite complex theoretically and empirically, however, the property tax variation across a market will be examined. Zoning, sewer provision and pricing, and taxation have significant interactions which complicate the hypothesis further still. The unit of analysis of this research is the land conversion market across cities. The fact that this market will generally consist of many local government jurisdictions adds to both the complexity and richness of the model. Previous research has barely investigated the ‘implications of economic and institutional interdependence within the land conversion market. Theoretical and empirical examination of the interdependencies in the land conversion process and the role of local government policies is the essence of this research. varie over time is1 the ot? We CHAPTER III SURVEY AND CRITIQUE OF EMPIRICAL STUDIES OF INTER~URBAN LAND VALUES Introduction The determination of urban land values has been studied in a variety of forms: 1) the effects of changes on a particular location over time, 2) comparison of different sites within a city at the same time, 3) comparisons of different locations within a city over time, and 4) inter-urban comparisons of aggregate variables. Much of the theory used to analyze intra-urban land values is based on von Thfinen's model of agricultural land rent. The basic idea is that location rent is determined by transportation cost savings and the concentric zone model of urban land use. Moreover: Modern economists have inserted the tools of micro- economic theory into this framework and adapted it to an urban setting. In the modern versions, Thfinen‘s town becomes the Central Business District (CBD) of a city; his crops become such urban uses of land as finance, retailing, housing. The object is still to show how competition determines the price of space which is shown to be a declining function of distance from the center. An optimal pattern of land use is determined that is still a sequence of rings, one to each urban use. (Goldstein and Moses, 1973, p.475). The work of Wingo (1961), Alonzo (1964) and Muth (1971) among others is rooted in the von Thunen model. Their empirical results are eeak. The problem with this approach is the changing nature of cities Lnd assumptions which ignore the complexities of the land market. 27 W 1977) r increas from th values. who, wi diverse and the factors recreat differe have e: Other e Prices r€1atee Muth ( States (1965) State . Vines housir Sagal) Calif ( 28 Wendt and Goldner (1966), Romanos (1976) and Ottensmann (1975, 1977) raise similar critiques of the von Thfinen model based upon the increasing complexity of spatial pulls which have replaced distance from the CBD as the transportation determinant in residential land values. The character of this diversity is discussed by Andrews (1971) who, within the sub—discipline of urban land economics, raises the diverse factors influencing the determinants of residential preferences and the factors developers must consider in location decisions. Those factors include the location of schools, business districts, recreational amenities, etc. The recent studies which have attempted to explain inter-urban differences in land values by the use of multiple regression techniques have explored FHA data on residential site prices across SMSA's.5 Other studies looked at per capita land values or residential site iprices across states.6 Several other complementary studies examined related factors.7 Land value appreciation, one of the approaches 5These include Maisel (1963), Mittlebach and Cottingham (1964), Muth (1971), Witte (1975) and Ottensmann (1977). 6Keiper, et.al. (1961) estimates per capital land values across states based on the land component of taxable real property. Gottlieb (1965) used FHA data on the average residential price of a site by state. 7These include a study by Van Vuuren (1976)On Canadian land values using Spearman rank correlations, a cross-sectional study of housing costs and zoning regulations in New Jersey by Sternlieb and Sagalyn (1973), and a cross-sectional study by Miller (1977) of three Californian cities. used '11 Schmid analysi methode T eXplail sectio: regres: indepel studiee clarif; for ge 0f the on Sta KeiPer used 1: most c Metro} and Ge These Varied SiZe ( 29 used in this research was first developed and empirically tested by Schmid (1968). This chapter will consist of both a survey of the analysis behind each aforementioned approach and the econometric methodologies used. With one exception, previous research efforts have attempted to explain residential site values by SMSA's or states by use of cross- sectional analysis. Witte (1975), however, developed a pooled regression using time series as well as cross-section data. The independent variables used in these studied have varied. These studied will be reviewed then key issues will be summarized to clarify the take off point for the econometric analysis of this study. Dependent Variables Geogggphic Definitions There are three approaches, based upon the source of the data, for geographically defining the land value variable as well as most of the associated independent variables. First is the research based on State variation done by Keiper, et. a1. (1961) and Gottlieb (1965). Keiper et.al.’s data was from the Census of Government. Gottlieb used Federal Housing Administration (FHA) State data. Second, and most common is the Bureau of the Census definition of a Standard Metropolitan Statistical Area (SMSA) used by Maisel (1963), Mittleback d Cottingham (1964), Muth (1971), Witte (1975), and Ottensmann (1977). see articles used the FHA SMSA market price data as the dependent ariable. Observations varied from year to year based upon sample ize criteria in data collection. Ottensmann also used data provided in Sch‘ This (1 metrop homebu SMSA's Census Each e implic estim: capit respe- betwe agric two i found was (i Gottl into While Variz impl use , is 1 that 30 in Schmid (1968) from the National Association of Home Builders (NAHB). This data was an aggregation of local homebuilder associations by metropolitan areas. The exact geographic definitions of the local homebuilder associations are not known but probably correspond most to SMSA's. Schmid's econometric analysis also used the Bureau of the Census definition of urbanized area in conjunction with the NAHB data. Each of these geographic definitions imply different analysis and implications. The State based studies of Keiper et.al. and Gottlieb used an e estimate of the land component of taxable real property on a per capita basis and FHA average price of residential sites in states respectively. Keiper et.al.'s results show a significant relationship between the dependent variable and income, population density and agricultural output variables. Gottlieb was somewhat successful with two income variables and a growth (employment) variable. He, however, found agriculture values insignificant. Keiper et.al.'s research was directed at explaining the geographic distribution of land values. Gottlieb argued that his approach would "yield some valuable insights into functioning of our urban land markets." (1963, p.4) However, while use of State data both for the dependent and independent variables can be useful for analysis of some questions, such as the implications of demographic shifts or State policies affecting land use, the degree of insight into the functioning of urban land markets is limited. Goldstein and Moses in their "Survey of Urban Economics" noted that "Researchers exhibit an understandable tendency to avoid defining the difficulties the land con‘ where land 1; agriculture, those indivi urban uses: institutions There sociologists 0f defining (Kurtz and I SinCIair anc‘ Manderscheie definitions Variation 1; depending 11 market is i. Suburbs in conversion with each a Hesse difficultl. theoretiea] 31 defining the relevant unit of study for their models because of the difficulties of obtaining adequate data." (1973, p.172). Conceptually the land conversion market is that area around a metropolitan area where land is in transition from non-residential use, generally agriculture, to residential or other urban uses. It is defined by those individuals and groups whose function is to convert land to urban uses: developers, landowners, land speculators and public institutions through regulations and policies. There is a wealth of literature, primarily by demographers and sociologists, which deals with the conceptual and operational problems of defining rural, urban, suburban, rural-urban fringe, and sprawl. (Kurtz and Eicher 1958, Gibbs 1961, Fuguitt 1962, Lieberson 1969, Sinclair and Manderscheid 1974, and Macura 1975)- Sinclair and Manderscheid (1974) and Macura (1975) applied different commonly used definitions of rural and urban, respectively, and discovered a large variation in the population which would fall into each category depending upon the definition applied. -Of course the land conversion market is in transition. What is fringe today is most often city or suburbs in the near future. So while the concepts of fringe and land conversion market don't necessarily overlap, the problems associated with each are similar. Research in the area of land conversion has one of the common difficulties in working in the transition area: "Frequently the theoretical and the empirical categories have been at variance since the fanmer t! the latter u: or political The co for this res I distin gories of mix still inters exhibi reside i.e., withor is not p.35). It is diffic ition. The 12 amnwfn essentially and the fun: areas along dez). An some discon (1965) defi Smawlg l) and 3 ) 198$ Categori es Part1 Qular 32 the former tends to focus on general social characteristics whereas the latter usually emphasizes physical, geographical, demographic, or political attributes" (Kurtz and Eicher, 1958, p.32). The conceptual definition of fringe which seems appropriate for this research focuses on land characteristics. Land use in the fringe is of a unique nature which distinguishes the area from all other residence cate— gories. This unique characteristic is the existence of mixed rural and urban land use--much of the area is still in farmland and residence of non farm dwellers are interspersed among the farms. This mixture of land use exhibits no consistent pattern of farm and non farm residences; if a consistent pattern of residences exists, i.e., if there are solid groups of residential homes without interspersion of non-farm dwellings, this area is not considered fringe area. (Kurtz and Eicher, 1958, p.35). It is difficult, however, to find data based on this kind of defin— ition. The land conversion market can take on various forms. It can be a narrow fringe or a broad belt. Also, "long ribbons of what is essentially urban development, both as regards the form of buildings and the functions performed in them, extend far out into the rural areas along the main highways." (Shryock, Siegel, and Associates, 1971, p.162). Another example is marked leapfrogging to the extent that some discontinuities occur in residential patterns. Harvey and Clark (1965) defined three spatial patterns commonly associated with urban SPTan: 1) low density continuous development, 2) ribbon development, and 3) leapfrog development. These can be considered descriptive categories in a static sense. They may all be occuring in any particular metropolitan area but will change over time. The de i.e. populat One could ha at a particu depending uP Resear of defining are both of the metropol potential in all land use development ratio of lan low density of residenti The grid or analysis. F showing ribl with ribbon Characterist While and ribbon c for sites t: develOpment less exPens: inadvertent: development 33 The definition of the unit chosen to express some relationship, i.e. population density, will influence the results of the research. One could have population dispersed throughout the area or concentrated at a particular point in the area and get the same average density depending upon the grid chosen. Research into land use patterns must address empirical problems of defining the density of urban development. Gross and net density are both of interest. Gross density, as used here, is the ratio of the metropolitan area to the total population. This approach has some potential in picking up leapfrog and ribbon development by including all land uses, though the variation in land used for non-residential development will exacerbate problems in analysis. Net density, the ratio of land for residential uses to people, might be able to pick up low density continuous development though here again the distribution of residential mixes from one metropolitan area to another will vary. The grid or grain chosen for analysis has direct impact upon the analysis. For example, one would have to have a grid capable of showing ribbon development in order to examine hypotheses associated with ribbon development. Other grids would be needed for other characteristics of concern. While it can be argued that low density continuous development and ribbon development might explain higher land values, land values for sites transacted over a large grid will, in the case of leapfrog evelopment, include the expensive closein land transactions and the ess eXpensive, more distant transactions. Therefore the data might ‘nadvertently indicate that appreciation is lower with leapfrog evelopment. — One We patterns is area. Nort] land uses, j Both report: lower popul.‘ insight on ‘ presently v. either stud; questions w newness and Previ and positiv density . M land hence tion or sit density. '1 preferences density and services wi taxation. Idle but remain: . . .1: urba- estii and : 131er larg 31+ One way of capturing some of the variation in settlement patterns is to know the gross amount of vacant land in a metropolitan area. Northam (1971) and Niedercorn and Hearle (1968) surveyed the land uses, particularly vacant land, for various American cities. Both reports point to the proportion of vacant land being greater for lower population size cities. While the numbers generated provide insight on the past development patterns, and could indicate if presently vacant land is filled in later, it seems unlikely that either study rigorously defined the geographic area to which the questions were directed. Therefore, significant variations based on newness and size of the cities could exist. Previous research has also found a statistically significant and positive relationship between site price and gross population density. More intensive use could indicate greater competition for land hence greater appreciation. On the other hand, higher apprecia- tion or site price will decrease the quantity demanded and raise density. This suggests a simultaneous relationship. Income and preferences also enter into this interaction. The implications of density and spatial patterns on the issue of the cost of public services will be dealt with later when discussing sewer provision and taxation. Idle or vacant land on the fringe remains difficult to measure but remains a concern in land value analysis as Clawson noted: ...land within the suburban zone not actually used for urban purposes typically is not used at all. Our best estimate is that there is about as much idle land in and around cities as there is land used for urban purposes. In the suburbs the idled land is an even larger proportion. (1971, p.318). If this idle expectations interest. ( percent of c (1975), how actual gains Anoth: concept als< definition < density ind: "If the sub1 and therefo: urban and r1 restrictive and Associa; Standard Me conceptuall; urbanized a: data. for or ment in the lost; e.g. density rat commercial 0n th alternate s the land cc 35 If this idled land is held for speculative purposes, the level of expectations and uncertainty associated with particular markets is of interest. Ottenmann's (1977) model associated expectations with percent of change in population. Schmid (1968) and Hansen and Schwartz (1975), hOWever, indicated the possibility of expectations exceeding actual gains. Another aspect of operationalizing the land conversion market concept also presents a dilemma. On one hand one can use the Census definition of urbanized area. The basic concept is a population density index. However, according to another Census publication, “If the suburbs are viewed as a peripheral part of the physical city, and therefore entirely urban, rather than as a traditional zone between urban and rural territory, then the former (urban fringe), more restrictive definition would be the preferable one." (Shyrock, Siegel, and Associates, 1975, p.130). Other census definitions such as Standard Metropolitan Statistical Areas (SMSA's) are also problematicl conceptually because of their basis on political units. Unfortunately urbanized areas and SMSA's are the basis of most of the available data for cross city comparisons and so serve as the basis of measure- ment in the econometric model. Much relevant area information is lost; e.g. areas showing potential for population growth and increasing density ratios and areas showing marked leapfrogging of residential or commercial development will not be captured by these measures. On the other hand, within the comparative case study research, alternate systems for operationalizing the heterogeneous features of the land conversion market will be examined. Comparison of these — detailed apj data. used 11 validity of method seler critical to hypotheses. and pragmat The g explaining indicate th geographic be using SN. informatior geographic Commison Then studies bei property v: price of r assessed b lls'lng raw The assessment 0f Govern] Value . Ilr PTOPSrty 3 36 detailed approaches when compared to the exigencies of the secondary data used in the econometric model should provide insights on the validity of this and earlier research. It is clear, however, that the method selected for operationalizing the land conversion market is critical to the definition of all variables and formation of all hypotheses. It is necessary, nevertheless, to be somewhat arbitrary and pragmatic in the choices made. The geographic problem with most of the previous efforts at explaining residential site or raw land values is the failure to indicate the problems associated with data using any particular geographic definition. As with these earlier efforts, this study will be using SMSA and urbanized area data but will interpret the resultant information in the context of operational difficulties in the geographic base of the definition. ‘Compapison of Dependent Variables There have been four types of dependent variables used in the studies being reviewed here. They are:' l) the land component of property value, 2) the price paid for raw land by developers, 3) the price of residential sites either received by developers (NAHB) or assessed by the FHA, and 4) land value appreciation which is calculated using raw land price or site price. Table 2 summarizes previous research. The estimate of the land component of taxable real property assessment ratios used by Keiper, et.al.'s (1961) study from the Census of Government for 1957 was highly correlated with total real property value. "The Spearman coefficient of correlation between land and property rankings in 1956 is .96" (p.157). Moreover, the regression / still/lull z m mum . l mm, at. al. (mn‘ m mm III Ill-I :- mm (Mud 0- II all." c( m In! mu: ll until all may [at 1956) -” ___._._————~—.——— you (my) 1 In "ch cl sum sun ll 9M: W em (Im’ IM .01 37 TABLE 2 smwm 0' WIIIQL leLS 0' FILVXOUS IESWCII y/vau 2 131327910!" VARMILLS‘IY VARIABLE TY" AND HATISTIUL SlCMl'lWCI WEI Avumu Not)” VAl‘IAlLl l (5 nine- ln pummel“. sand-rd "for ozhntvl cho-I Fowl-Mon focal fowl-un- Agricultural OIIIII other Den-Hy yopul-uan Chum. Valu- n. u. ll. (1961)1 In rapt:- Papal-um .1... u c In Ind v- luu Person-l DIllltly by Annular-J. nu- (b-ud on an Inca-I by sun ‘7“ .by I." o( c c 311‘. Stu- n...“ out .31 .u ptflptlty for b1:56) .15 01 (“63) 1 Fowl-(ton Chun- h Yahoo a! nu m:- nice of Dun-12y, fowl-um A'tlculllllll a! low din; sun SPSA. (1) 1930—1960 sun Noun. SHSA y. J6 1960 1 95A .29 .36 J0 .11 (1.61) (‘.27) - (L31) (1.98) . m4 Avon;- Inca—- out ugh-s: uni inh- (1964) pct "hold, Pom-um who on SBA Pris. of A. 60 536‘. 2.10 «nun sin .11 .J) .1! (2.51) (1.17) ( .u) 1.1:: (l9::)‘ an. Annual Crunch L- um Supply at 0-5". of of For Cantu mgrkulmnl tun-Junk. aunt. Lee. rover-b1- and; 51:. by Inco- b [aplenty-1.91 60L. «4 by St: . Ann—u 1961 .22 sun. 1961 (inunflt- at “an: .13 (“Inning-(1) on and) .4 (“sullen-t) (ull'nlflcnc) .l (llllr lulu-l) rerun! 0| fowl-lion Ilkh Into-a Above 3 0 000 by Sun. 1961 .‘l' (Ind “ll, significant) 5 in an“ no l-tlon To 1 In: It on." r :n An- L232 t,- (3) y "ran." ropululo- 1n Populate: Fowl-(1°- VIN-"Id u I... 1960 “thank-l nu- wunuu 19 19 win: In And. 1960 , “a" An- Dru-had Ann. 1960 Urban"! Ann SEBA u... Lot "1". .12, n“, 1960 .00]. -.‘93 nu Both .00 J1 1- C“! in. Hull, .91 . 1M "60 .l). .21. SEA 4.029 .30 . nut-u l—uy hul Plrcut Chan" hue-c Inco— ueo (a) pawl-non in Input-non DUI!!! C117 City. 1960 1950-1960 City 1- u“ .013 -.cm 1‘ I}! An- 1950- ,1) _°. . 1,50 Una-Ind n <3]! 51 lore-n: In: ring L: "In". 1960 i. 9 Ml! 2 “m." Ruth (1911)5 Lo: o! the Mm" Mu of Int- «add me: 1- m's. 19“ I" um: (limb umnl W of the MCI he 51"" Foot of humid Slm km! 9M“- ms .a mm um. mm mm ' Inmate: o! laud in 0f “H Price In hum foot of holding“! :lm mm son'- mum Von-M Mtg-AI, um) um km land Mn. 19“ 8mm: Son of the Inter- m Mot-nun fr [Mu mum:- an not I m Milan “m (1913, um um p. 362). . iii‘?..‘.‘.‘.i‘l‘ "’°" i" ‘Sehu's real: “but 3 ch out the 10".." m Him (“75. p. 3:1? “I ‘ “:10 (ms. 9. 351) r t 7 mm. «mum of H 0"Mu- (19m "to!" 2 :onl'd. {x911 5 Nu Flllly Pr I r Con-truc- ' the Aveng- Inco-n Squ-r- too: :19: Cu: at lesl- 1960 o! "la — [man I], Slnl (II .128 [1.1 Slit. .lJ7 1m .76 (1.61) (.52) .Al‘ (5.56) (1975) Mun Inca-c Fowl-lion Pcrzntngc Av::lg| Vllu‘ AVIIIIC 31cc mrlgllc I Log oi mug-g: Den-11y. SBA Dung- 1nd A;r1:1::ur:1 Sln In: (nu: : Prlcn mm: 1970 Fowl-lion -.A9 | It. SFBA . 1960-197° (8-11) -.ll :1 .27 (3.12) DEA (1.11) (1J0) ultlll (L71) .1! Acron- 0-3” I . . 8] Win-plum: qutmtlgl lea-Aug AIIII‘C I410. SBA. Gang. 1: Ag. Loan to 1970 non-uh".- .0) run . pal-clan ( .Jl) uuo (1.30) 1960-1970 -.09 .05 (1.1!) ( .70) Ann.- Tcr— la HIKuIlIY (4)) (1975) Pooled Indian Inca-c Fowl-do- Pncnu‘c Av-ugn Val-u Avon'- D—y 1967 ulna a! at Mrt'I‘Ol Milly. SPSA Chang. 1!: 1 Acre o S“. Stu I1 hog ol the or "IA an. 1970 Fowl-u. Agriculturll -.A9 My 196! hr Squar- PBA .20 1960-1970 sun (20.13) D! I'lldmtlll .15 (3.71) 734 .27 3—1 196’ mu SKSA'c (11.78) .1) (10.9” [969 .7! (Lll) In“, rh- l—lly :1 "(cat Gun ) [mo-c. 1960 Fowl-non 1n Papal-r no: lav land 1960. s SPSA ‘ 196‘ .55 .l] 1.32 1950-1960 .‘5 .1‘ 17.71 ll.ll {he lnfonutlon on other nudit- M (M- tabla :1.- Iron Htuc (1915). on." r such“:- we" no: rev-"ted. t-v-l uu I (1975. 1).;62). (Hob dld not. noon gut-unte- no u Hut: (1975. p. 361) is "Ira-lot! analyses." Soc-c ( All. inbmuon lro. Inn (1975). Tnlo J. var u checked lulu: [In "1.1" ‘H'I "-11“ "port "uh-“rd error: and It. thus no: la ninth-nu. ’4 ul lhc log of All varlnbh- 1- M. rqrultn wlyuu. - (1:75. p. m) I (1915. p. )5?) r! port-d I nri Ivluo 4-“an 1m 9! In uc' I vurllblcn Ipnll’ 1- A! nil-Ann (1977) upon: land-rd Irrati- .. at cm.- «action-1 rqnollzl (or your: 1966-1969. mu A. “pulled hut "lulu (owned. (cc. Sui-1 loom flu gnu-1180:! «nation (or l969 1- prucncd Inn. Hiro- I-u-n that LN “guru 1n punch-n- below KM nun-ion rnulu 1n Kittlob-ch and continue“ uru luv-dare dulnum d 5".“ null- App-Ind 1n Sch-1d (19:0) and menu-Ann (1911). m -l acct-nary to ebul- - unlo- af “gunk-net ol vale-u v-rlaolau lm Ml Hutu-Hon at the rot-:1” Ruth's «duel-nu In "(mt-d «lug umlwllzu no: “much-d “rt-nu." model worked land compone arose from r between stat different 12 The be the price p data may ex: 1964 repres. data are re Resid internal co OPPOI'tnnity EElCh of the can eXplain independent average sit Products (14 Schnj The compute begins Wit} the lot is total is S, of absolut. Percentage 39 model worked better for total property values (per capita) than the land component estimate. To Keiper et.al. the most troublesome issue arose from not only the lack of consistency of assessment practice betWeen states, but also within states, hence, offsetting effects of different land market practices. The best data operationally for site price or appreciation are the price paid for raw land by residential developers. While this data may exist in scattered studies, only the NAHB data for 1960 and 1964 represents significant systematic gathering of such data. This data are reproduced in Schmid (1968) from NAHB sources. Residential site prices as reported by FHA and NAHB has several internal components: 1) development cost of a site, 2) agricultural opportunity cost (e.g., agricultural value), and 3) size of the site. Each of these factors imply different policy questions. They in turn can explain the reason for statistical significance found in such independent variables as the construction cost index (Muth, 1971), average site size (Witte, 1975) and value of agricultural land or products (Keiper, et.al., 1961; Maisel, 1963; Witte, 1975). Schmid‘s land value appreciation is derived from site value. The computation process is as follows: For each city the analysis begins with the price per finished lot. The farm value of the land in the lot is computed and added to the lot improvement costs, and the total is subtracted from the finished lot price, to obtain the amount of absolute appreciation. The appreciation is then expressed as a percentage of the farm value. Ottens notes that: Q L. over i land I smalls land 1 large P394. This argume1 challenge t] The l 1) Income 5 2) Total pc When eithe] 3) Value 0: three out . 1+) Other v, size, Cons Incc has affect studies . but two 8, Populatio] £01m (vit- income . variabl e S 40 Ottensmann, in commenting on Schmid's appreciation variable notes that: Schmid's dependent variable has percent appreciation over farm land values. This is highly correlated with land prices themselves, since farm land prices are much smaller and vary less. However, any error in the farm land price data is magnified by this procedure, producing large variations in the appreciation variable. (1977, 13394)- This argument notes the measurement difficulty but does not directly challenge the underlying theoretical concept. Independent Variables and Regression Results of Previous Research ‘ The following results of previous econometric research stand out: 1) Income and population density were most often found significant. 2) Total population and population growth were often found significant when either income or density were found insignificant or not used. 3) Value of agricultural land or output was found significant in three out of four studies explaining state or SMSA site variation. 4) Other variables found significant in various studies related to site size, construction cost (indices) or price of complements. Income and population seem to have an inter-relationship which has affected which variables have been found significant in these studies. Average income of one sort or another was significant in all but two studies (Maisel, 1963; Schmid, 1968). In both of those cases population change was found significant. Also, in only one case out of four (Witte, 1975) did population density enter the equation with income. In other words, income seems closely associated with size variables, total population or gross population density. As the population t family shouj is complex . purchase la: costs of 001 of course, I the problem: major geogr. urbanized a: the part of What the bu willingness Perce Studies . '1 Value of ag Studies and tance of ag °‘~“ Competir The < of new home ASsuming S< costs, the COHStI'ucti. affect Clem; 1+1 population (size)of the metropolitan area is larger, then income per family should be greater. The relationship between density and income is complex. Higher income is associated with a greater ability to purchase larger lots but the cost of living in dense areas and the costs of congestion are also associated with greater incomes. This, of course, leads back to the problem of geographical definitions and the problems of mixing different characteristics in any of the three major geographic definitions and even within the fringe area of the urbanized area. Income may also indicate a degree of market power on the part of sellers, either direct or through expectations, to charge what the buyer can pay or inversely a measure of the degree of willingness of buyers to pay. Percent change in population is statistically significant in four studies. This also perhaps indicates some role for expectations. Value of agricultural output or land appeared significant in three studies and insignificant in one other. This demonstrates the impor- tance of agricultural land value as an indication of opportunity cost or competing uses of land and, hence, a supply characteristic. The other variables were residential costs, site size and price of new homes. Construction costs could affect both supply and demand. Assuming some relationship between construction costs and development costs, the supply of lots will be affected. On the other hand, Construction costs associated with the price of new homes will also affect demand. Site size was found significant by Witte (1975) and indicates a relationship between per unit prices and size of the lot. The is by Witte: ‘ price per s: a more nearl Witte state! sion analys not standar The i relation be Situation t nature of i unambiguous one in this Only if 001 any direct which can Miller, 19 there has Sliperior ”t Muth Seems 0f ounce]:I implicati< 42 Issues of Functional Form in the Dependent and Independent Variables The issue of functional form was raised initially in a footnote by Witte: "The logarithm rather than the unmanipulated value of the price per square foot was used in order to give the dependent variable a more nearly normal distribution." (1975, p. 357). On the other hand, Witte stated that, "Muth used the log of all variables in his regres- sion analysis. Muth's coefficients are estimated using unmanipulated not standardized variables." (1975, p.362). The issue of functional form is related to the hypothesized relation between a dependent and independent variable. In any given situation the researcher cannot know with complete certainty the nature of the functional relationship. "Ideally, his theory tells him unambiguously which to choose; if he fails to utilize the appropriate one in this situation, his estimates will be biased and/or inefficient. Only if complete searching of the theory does not give the researcher any direction should he proceed to use the following ad hoc procedure, which can never completely substitute for a good theory" (Rao and Miller, 1971, p.105). Certainly in comparing the research to date there has been little theory and no clear evidence that log forms are superior to linear forms of the equation. The practice of Witte and Muth seems to have been to use non-linear functional forms to take care Of concerns about heteroskedasticity without concern for the theoretical implications of these functional forms. The t‘ and Ottensm These two s research in considered, Ottensmann equations v pooled reg: reference 1 be conside: 43 Comparison of Studies by Witte and Ottensmann The two most recently published studies, by Witte (l9?5)(l977) and Ottensmann (1977), can be contrasted to raise several issues. These two studies represent a contrast in several areas of approach to research in this area. These include 1) number of variables considered, 2) functional form, 3) regression techniques tried (e.g. Ottensmann used a recussive model and attempted some simultaneous equations while Witte had a series of cross-sectional models and a pooled regression), 4) theoretical arguments, and 5) results. In reference to Witte's article Ottensmann raised the following issue to be considered here. Other alternative explanations of the level of land values have been provided; however, the derived demand model developed and tested by Witte (1975) is one of the best examples. She has achieved higher coeffi- cients of determination but only at the eXpense of considering a greater number of independent variables. The simple straightforward model tested here, with but three independent variables, must be considered as a valid alternative. (197?, p. 389). Ottensmann’s three independent variable model offers very little new including the theory justifying their use. On the other hand, Witte presents little rationale for the use of variables or for the Lometimes fanciful proxies chosen. While this may have been a unction of publication space it also seems that little attention was aid to the implications of each specification. Regarding variable zlection Witte noted. In many cases, a number of alternative measures of the determinants of residential site prices were found and that measure which gave the best explanation as measured by the adjusted coefficient of determination was the one utilized. (1975, P-356) The r1 development conversion ‘ population variables. application simple mode and on the beyond the Otte: little imp in regress transforma regressior Witte used achieve a The need com a recursi equations number c an The results of both studies, however, are not reassuring for the develOpment of instrumental variables designed to influence the land conversion market. As do earlier efforts they do point to income, population growth, size and population density as significant variables. These alert us to the need to study in more depth the applications of urban shape and structure. We have on one hand a simple model capable of multiple explanations or vague generalities and on the other hand a finely manipulated model with little theory beyond the concept of derived demand. Ottensmann experimented with a log functional form and found little improvement in results. "Different functional forms Were used in regressions for some of the variables. For example, a logarithmic transformation of the population variable was tested in all of the regressions. None of the tests were conclusive." (1975, p.395). Witte Witte used a log form of the dependent variable in an attempt to achieve a more nearly normal distribution. The regression techniques tried by each of these studies also need comment. Ottensmann, while reporting cross-sectional results of a recursive model using OLS also attempted a system of simultaneous equations. He reported: TWO-stage least-squares procedures were used to estimate the parameters, with population, income and the population change variables considered as exogenous. In each case, the parameter values associated with the original three predictors of land prices were hardly changed from those obtained with the recursive model, while the parameter associated with density of develOp- ment was insignificant. (p.395) Witte used standard multiple regression techniques. Given the mber of variables used by each it seems that the techniques used were approp squares met left out of large numbe determined of each apI simultane 01 In a: reviewed t' Oped an es assumption a high ela Particular results we resf for mucl boti thi- men pri Per be of 45 were appropriate to the other model. For example, two—stage least squares methods raise issues of a critical nature when variables are left out of the model. Witte, however, failed to take advantage of the large number of variables used for either a recursive or simultaneously determined model. There are, of course, advantages and disadvantages of each approach but theory clearly indicates that some variables have simultaneously determined characteristics. Elasticity Issues Raised by Witte In another article using the same data base, Witte (1977) reviewed the elasticity assumptions of previous research, then devel- oped an estimate of the price elasticity of demand and tested earlier assumptions of the constant elasticity of demand. The results indicated a high elasticity of residential site price with respect to income, particularly for middle income groups. In addition, the following results were noted. Our findings indicate that the elasticity of residential site price with respect to size is hi for SMSA's with medium-sized lots (approx. % acre much lower, indeed highly negative, for SMSA's with both small and large lots. If, as hypothesized above, this is due to large lot zoning and other local govern- ment policies, it would seem possible to lower site prices by appropriate alteration of these policies. Perhaps the most promising of such alterations would be restrictions on large-lot zoning and encouragement of the redesign of older subdivisions. The relatively low estimate obtained for the price elasticity of demand for residential sites (-.7) provides a potential explanation for the prevalence of land withholding, particularly in rapidly growing areas. Assuming that holding vacant land is profitable, as it normally would be in a rapidly growing area, this inelasticity of demand would encourage land withholding since this would increase the capital gains of the land holder. Such land withholding could be made less $622.1 proff Local by e: rais: P8-‘ The for certai made in Ch however, n action of Pre\ total pop1 etc” to e aPPI‘eciat: arise fro: exPlanato. is arbitr aSpects 0 Character trends of can Qmpy to be tea and the 5 data. H: manize< market r. 46 profitable if land holding costs were increased. Local governments could increase land holding costs by either increasing the tax rate on vacant land or raising the assessment on such land. (Witte, 1977, pg . 408-409) . The conclusions drawn by Witte regarding the elasticity of demand for certain land uses are supportive of the elasticity assumptions made in Chapter II. Zoning and property tax conclusions drawn are, however, not sufficient to handle the actual complexity and inter— action of local government policies in a metropolitan area. Summary Previous research has concentrated on demand variables such as total population, median family income, percent change in population, etc., to explain a range of variables related to site price and/or appreciation. Variation in the unit of analysis and problems which arise from these difficulties were reviewed. In both the selection of explanatory demand variables and units of analysis, the final choice is arbitrary. Because of the complexity of the urban structure many aspects of community characteristics are interdependent with other characteristics. While each listed variable is indicative of different trends of interest, selection of the appropriate group of variables can only be made after an analysis of the specific theoretical model to be tested. Geographical units are less amenable to such decisions and the selection of which unit is used often depends on the available data. However, where possible this research uses the Census urbanized area as most appropriate to capturing the land conversion market related characteristics. Find: to follow. potential government selection develOps t essential] ——————m‘"‘“d"“ ‘ 47 Findings in earlier research function as a base for the chapter to follow. While emphaSis in this research is on development or the potential for development of instrumental variables related to local government policies, earlier models provided much direction in variable selection and empirical problem identification. The next chapter develops the data and measurement for site price and appreciation, essentially updating ideas developed by Schmid (1968). llris variables, initiaquv. (FHA) axni are then r ciation am site Size, The calcul concludes and its 1‘! Ide (resident tion or 1 raw land Others, 1 by this : Site as 0f hOHSe CHAPTER IV SITE PRICE AND APPRECIATION Introduction This chapter discusses and operationalizes the dependent variables, site price and appreciation. Site prices are discussed initially. The issues related to the Federal Housing Administration (FHA) and National Association of Home Builders (NAHB) data sources are then raised. The various components needed to calculate appre- ciation are then sequentially discussed. These components include site size, development costs, and agricultural opportunity costs. The calculation of appreciation is then demonstrated. This chapter concludes with a brief summary of the land conversion process and its relationship to site price and appreciation. Site Prices Introduction Ideally, the price paid for land to_be converted to urban (residential) uses should be used as a base for calculating apprecia- tion or used on its own merits. However, data on the price paid for raw land is nearly nonexistent. Clawson and Stewart (1966), among others, has dealt extensively with the woes of analysis brought about by this scarcity. Market price of a site, or site price, price of a site as a percentage of total house and land value, and average value of house and site, along with value of land and buildings in 48 agricultur With the e Census of Administre As f over 300 j 1948 to l? percentag' it was in 1948. Th in the co increase Priced FE Pri value. 1 ment cost policy 11 time can developm. due to a Opportun If the h and bett Costs ha priVate increase 49 agriculture are reported for the U.S. from 1960 to 1977 in Table 3. With the exception of agricultural land value, which comes from the Census of Agriculture, the data came from the Federal Housing Administration Data on Cities and States. As Table 3 shows, the mean market price of a site increased over 300 percent from 1960 to 1976. Similar increases were noted for 1948 to 1964 by Schmid (1968, p.7). While the site value as a percentage of house value remained near nineteen percent in 1976, as it was in 1964, it still represents a jump from the eleven percent of 1948. The high of twenty one percent in 1972 preceded the rapid rise in the cost of a house, as indicated by the average property value increase of eighty four percent from 1972 to 1976 for moderate to low priced FHA insured homes. Price of a residential site has been used as a proxy for raw land value. Problems arise because other sub—components, such as develop- ment costs, of the site price have different characteristics and policy implications than raw land. Variations across cities and over time can be explained by agricultural values, lot size and quality, development costs or land value appreciation. If the variation is due to agricultural values, the implication is that agricultural opportunity costs are high and more land is withheld from the market. If the higher prices are a result of lot size and quality, then larger and better lots are being demanded and purchased. If development costs have increased, then the source of those increases, whether private or public, should be examined. HOWever, if appreciation has increased, then the sources should be isolated and examined. While Year 1960 61 62 63 65 66 67 68 69 7o 71 72 73 74 75 76 77 Source : TABLE SITE PRICE AND RELATED STATISTICS, 1969—1977 Price of Value of Market Mean Price site as land and Year Price of of house percent of buildings a site and site property in agri- ($per site) ($) value culture ($per acre), 1960 2223 14326 15.5 116 61 2477 14855 16.7 118 62 2725 15460 17.6 124 63 2978 16189 18.4 130 64 3130 16522 18.9 138 65 3416 17289 19.8 146 66 3544 18099 19.6 158 67 3776 19163 19.7 168 68 4161 20116 20.7 179 69 4214 21186 19.9 188 70 4961 23710 20.9 195 71 5066 24373 20.8 203 72 5307 24321 21.0 219 73 5051 25112 20.1 247 74 5372 28488 ' 18.9 310 75 6329 33376 18.5 343 76 6963 35600 19.1 390 77 n.a. n.a. n.a. 456 Source: FHA Data on Cities and States, 1960-1976; U.S. Department of Agriculture, Economic Research Service, Farm Real Estate Historical Series Data: 1850-1970, ERS 520, Washington, D.C. June 1973. these four and intere: Empiricall; the relati Data Sourc The published Standard F; Associatic for 1960, Of these 6 discussed As 1 metropoli' depending samP1e is This sele tudinal s Si“11316 cr 51 these four causes of increased site values are examined, the theory and interest of the research in this thesis is on appreciation. Empirically examining both site price and appreciation will clarify the relationship between the two. Data Sources The Federal Housing Administration Data for Cities and States has published site price data for selected housing areas defined as Standard Metropolitan Statistical Areas since 1948. The National Association of Home Builders has also published data for site prices for 1960, 1964, and 1969 and raw land prices for 1960 and 1964. Each of these data sources, for the following empirical research, will be discussed below. As mentioned, the FHA reported site price data for selected metropolitan areas. These metropolitan areas varied from year to year depending upon sample size. It's likely, therefore, that the FHA sample is biased towards active, growing areas and/or larger areas. This selection process also severely limits the sample size for longi- tudinal studies and is likely to increase the aforementioned bias over simple cross-sectional analysis. Other aspects of the FHA data are succinctly discussed in an appendix supplied by Witte (1975): The ideal dependent variable for the testing of the model developed in this paper would be the average price of single family building lots in well defined housing market areas for sites of standard size, quality and location. The actual dependent variable used in this model is only standardized for size and is for somewhat well defined market areas. It is not the average price for all building sites but rather the price for a 70 to 100 percent sample of new homes insured by the FHA, under 203(b). Sampling errors can be considered minimal due to the large percent of population samples; however, the FHA tends to this rese 11$ng Sch NAHB , T11 homebuilc is proba} Urbanizec membershi non-reSp. NAHB dat: ITurket , than FHA resident Smctmm Outside aVailabl 6‘ “Seful 52 insure middle or moderate income housing and hence their figures for site prices may not be representative of the upper or lOWer portions of the housing market. In addition, the lagged adjustment of FHA mortgage ceilings probably means that FHA site prices understate rates of change of site prices in periods of inflation. The new homes which the FHA insures are primarily located in the newly developed tracts lying in the suburban parts of each SMSA. However, since the land built on in each year and in each SMSA will not have the same locational dispersion, an unknown bias is introduced into inter-temporal or cross-sectional comparisons. In addition, the share of housing loans, insured by the FHA, in each market varies from year to year and from SMSA to SMSA and hence the degree to which average FHA site prices represent the true average site prices for an SMSA in a given year will vary. A major advantage of FHA data is that the value of the site is estimated by trained appraisers in order to make important loan decisions. (Appendix A) As noted in Chapter III, the FHA is the prime source of data in this research area. However, Schmid (1968), and later Ottensmann(l975) using Schmid's monograph, used data from homebuildersurveys of the NAHB. The geographical definition of the survey was based on local homebuilder organizationsbut it could be expected that the NAHB data is probably closer to the SMSA political delineation rather than the urbanized area or land conversion market. Responses from the large membership surveys involve an unknown bias on non-membership and non-response as well as possible regional variation in responses. The NAHB data also picks up a wider spectrum of the housing and land market. Since NAHB site size and value data is consistently larger than FHA data, the NAHB data probably reports more higher valued residential development than it does the lower valued part of the spectrum. NAHB data, as FHA data, will reflect land substantially outside of the land market of interest. Nevertheless, for the available years, the existence of both FHA and NAHB sources provide a useful contrast and complement to each other. Site Size Be: The low ' in Table Year “ 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 Source: Natlom 53 Site Size Residential site size has been decreasing steadily since 1966. The low period was registered during the 1974 recession as demonstrated in Table 4. TABLE 4 AVERAGE SIZE OF LOT, SINGIE FAMILY, FHA 1966-1977, U.S. HIGH AVERAGE STATE AND LOW AVERAGE STATE, 1966-1977 Average Size of Lot High Average Low Average State (sq.ftJ rggr U.S. (sq. feet) spépgp(§gpft.) 1967 9,796 22,526 (Conn.) 6,930 (0a1.) 1968 9,186 22,633 (Vt.) 6,788 (Hawaii) 1969 9,580 17,215 (Miss.) 6,325 (Nev.) 1970 8,611 16,587 (8.0.) 6,336 (0a1.) 1971 8,558 17,090 (S.C.) 5,973 (Cal.) 1972 7.731 14,770 (Ala.) 5,515 (Cal.) 1973 7,502 14,476 (5.0.) 5,676 (Penn.) 1974 7,456 15,239 (5.0.) 4,251 (N.J.) 1975 7.972 15.556 (11.1.) 2.395 (M) 1976 7,983 16,470 (3.0.) 6,508 (0a1.) Source: FHA Data for Cities and States 1967—1976. National Association of Home Builders data in Table 5 demonstrated that at least through 1969, the average size of a finished lot had been increasing. Source : While it housing t unless re Neverthel over time Of the s;‘ 0n 2.6 lots site Siz resPecti PET site B< rmid rz‘ regardec of a $1. increas. the SCO 54 TABLE 5 AVERAGE SQUARE FEET OF FINISHED LOT, NAHB Year 39. Feet 1950 7558 1960 8932 1965 10312 1969 12839 Source: Sumicrast and Frankel (1970, p.58). While it could be expected that the NAHB lots are for more expensive housing trends since 1969 should be expected to parallel FHA trends; unless restrictions, such as large lot zoning, restrict supply. Nevertheless, the data indicate that as the price of a site has risen over time the size has decreased. This does not speak to the quality of the sites and the development costs of a site. Operationally, in the case of the NAHB series, the figure of 2.6 lots per acre develOped by Schmid (1968) is used. For the FHA data, site size by metropolitan area has existed since 1966. These respective figures are used to calculate agricultural opportunity cost per site. Development Costs Both in the literature and in discussions with developers the rapid rise in the costs of improving a lot for urban use has been regarded as an important reason for the increase in the market price of a site. Government regulations are often blamed for much of the increase on these costs. A detailed analysis of this issue is beyond the scope of this research. However, because of the importance of develOpmex of the is: Sag: subdivisi of standa explainin communiti state or associate of Home I Skepticis intra-reé Washingtv However, 011 the 3; Ne 55 development costs to the calculation of land value appreciation, some of the issues will be sketched and some data reviewed. Sagalyn and Sternlieb (1973) concluded that, "The impact of subdivision improvements, given the present uniformity of a high level of standards, does not appear to be statistically significant in explaining selling price variation." (p.56) Their study was for communities in the state of New Jersey. This might indicate strong state or even regional consistency of requirements and costs associated with subdivision development costs. National Association of Home Builder data does indicate strong regional variations. Skepticism about that data arises because of the suspicion that intra-regional variations (say between California, Oregon, and Washington) can be as significant as inter-regional variations. However, neither national nor cross-city data exists which can improve on the available data, primarily from the NAHB. Nevertheless, data available for the city of Detroit allows some perspective on the increase of development costs over time. Table 68 of estimated subdivision development costs demonstrates the rapid rise in development costs from 1963 to 1976 of a total of 8Table 6 summarizes data from Appendix B prepared primarily by Robert H. Carey, President, Thompson-Brown Company, with the exceptions of 1958, 1959 by Ross Campbell and 1975, 1976 by Roy Russell, both associated with Thompson-Brown. While there are some difficulties with the data base because of changes in definitions and a few apparent inaccuracies, this time series seems quite unique. The data were calculated for each year and does not involve retrospective calculations The articles for which the data was prepared appeared in a number of publications including the Michigan Buildor. Furthermore, for the years 1969-1974 the articles written by Mr. Carey included sewer and water use and tap fees for selected municipalities in the Detroit metro- Politan area and are reproduced in Appendix A. 56 I 3 033 :12. £3353: 3.9.50: .Elcu 50.49.3885... .0555... £25 HR. ER. ..n Ar...“ team. 43...: ASE 6R: .nRo— .22: .38 .33 ..38 .Roo .2...“ .33 ...c.. m. in .8 .Rn ...Rw deem .8...“ .83.. .39. . :~.. .83 .Nman .39 .93 .82 .2»... ed Ru d... d? .Rn. .2: .3... an.» .2: .Rs .56 ...ee den 59. .E: .8: as as .n: .e... .32 .68. .sa 8.. ...en .Sm .5... ...D. .eR as .62 en «fin a... .no .e:~ . 2e .3... ...mn .R: .8: .nR .9n .:n .Rn ...3 .e- .8~ n. - 1... .8 . 3.. . Si .Rmn . NSN Aim .28 .RS . R... ...eD .29 .3...— . i: . .8 «6 Ta. .3 .3 .32 .RE .86 ...8 .5. .si .25 .8“ .9R 5% .ch .3“ 6.2 5.2 .~.. .8. .38. .85 6.3 .3... .53 .82 .mn: .nk: .68 .3: .86 .18 4 JAN oNuno .flm.” .mwm. swan... as.5.o as R ..n E S E as $ 8 he 8 we so ...Enufi— ..c .mgvuuph ...Zoufl— ..EEE .n 5:535: :_ 35. he Fran—3 .= .132. 3 215:... $2.12.. . och-En fittiili .n~.: .cD: ..Rn .SN dew .nR 63 .95 no H¢U< a! BB Q.N §=fl>< 5: 2M: .8 835 3: a .8 aim .5.— :2 63732 a: 39 51.415 E86 25:35": 8.9328 snags..."- o Sufi u 5 c... :12.— 283. .1; 15.1.26 2:. ago mas—«m £55: a: :50 ..:0,L:—,.:cv 1.: . :5: 53.48::— _.=.... 5.1:; .— . 7.:— givcfiuiu secs—r: 1:...— smmcn-E: .5: 5.5: leach: .725: Isa—Sm .515... Le... 279 percer percent re and indie: tively in This coul< requireme: role for from the to 1974 s 41 and 3', rose by 2 user Cha: faster a; 0pment c A Gated, h tllP'lcal 17 less Water a in term: and not. While n regiona 57 279 percent. Sewer system and water system costs rose 104 and 98 percent resPectively. This was a slower rise than in other costs. and indicates a drop from 19.7 to 10.6 and 13.3 to 6.9 reSpec- tively in the proportion of those services to total development costs. This could indicate that the source of increased costs via government requirements are in road and drainage. Further evidence of a limited role for costs associated with sewer and water regulation is drawn from the Appendix A of sewer and water charges and fees. From 1969 to 1974 sewer system and water system development costs increased by 41 and 37 percent respectively while sewer charges and water charges rose by 22 and 17 percent respectively. This could indicate that the user charge aspect of government regulations were not rising any faster and even much less than other factors associated with devel- opment costs. A recent report by the General Accounting Office (1978) indi- cated, however, that, "In the 87 communities sampled, we estimated typical savings of about $1,300 a house if communities would allow 17 less expensive requirements for streets, sidewalks, driveways, and water and sewer systems." (p.15) That report details the components in terms of street width, number of inches of concrete deep, etc. and noted the large variation in requirements for its sample size. While no breakdown was available by community, a few hints of regional and intra-regional variations in municipal development, utility and building fees are presented in Table 7. Census Select Northe it Ne North C] S. C South 21>- West Hfhh—a Sour TABLE Z SELECTED MUNICIPAL DEVELOPMENT, UTILITY AND BUILDING FEE RANGES, AND AVERAGE FEES FOR ASSOCIATED SMSA, 1978 Census Region and Range of Development, Selected SMSA Utility and Building Fees for Selected Communities ($) Northeast Philadelphia 307-1495 Nassau/Suffolk 526-2485 North Central Chicago 200-1293 St. Louis 73-1302 Cleveland 192-1144 South Houston 56-1048 Atlanta 293‘9094 Washington, D.C. 1476-3265 West Los Angeles/Long Beach 1003-2274 Seattle Everett 434-1949 ’ Denver Boulder 1402—3172 Average Develop- ment, Utility and Building Fees for Associated SMSA§§Z 1025 973 775 639 543 2398 1418 2275 Source: General Accounting Office (1978, p.27). the i seems inter cos get the loc The de 59 The data on and empirical research into development costs and the impact of government regulation are sparse. The need for research seems great although conclusions seem to be drawn and normative interpretations implicit. Site development costs have been one of the most steadily increasing components of housing costs generally over the past ten years. This has happened because higher, more costly standards have evolved and because costs formerly the reSponsibility of local government and not included in the purchase price of housing have now been shifted to the developer, who passes them on to the housing consumer. Site development may include the costs of grading and clearing; construction of on-site or off-site utilities (water, sewer, gas and electricity); storm water management; dedication of land for on—site community facilities, such as schools and parks; payments in lieu of dedication; and various fees, charges, and other assessments. (HUD, 1978, p.23). Table 8 summarizes the development costs estimates used to calculate appreciation. It provides regional data for development costs per front foot multiplied by the national mean lot frontage to get development costs per lot by region. This regional number is then used to calculate appreciation for metropolitan areas which are located primarily in a state associated with the defined regions. The difference in the national average site size between the NAHB and FHA sites in 1969, the FHA was .72 of NAHB, was used to calculate the development cost for an FHA site. Agricultural Opportunity Costs It is argued in this research that the market value of agricul- tural land for agricultural use represents the opportunity cost of the land. This section will briefly review major factors operating in the rural land market, factors which affect the ability of the landowners, speculators or farmers, to hold land. 60 TABLE 8 DEVELOPMENT COST ESTIMATES BASED ON 1969 NATIONAL ASSOCIATION OF HOME BUILDERS DATA Mean Land National Mean Development . Development Cost Lot Frontage Cost per Lot Region Per Front Foot ) ($) by Regions($) New England 25. 90 2250. Mid Atlantic 36 . 90 3240. South Atlantic 29. 90 2610. East South Central 27. 90 2430. East North Central 37. 90 3330- L West North Central 33. 90 2970- West South Central 24. 90 2160. Mountain 32. 90 2880. Pacific 44. 90 3960. Region : States New England : Maine, Vermont, New Hampshire, Massachusetts, Rhode Island, Connecticut New York, Pennsylvania, New Jersey Maryland, West Virginia, Virginia, D.C., N. Carolina, S. Carolina, Georgia, Florida East South Central : Kentucky, Tennessee, Mississippi, Oklahoma East North Central : Michigan, Wisconsin, Illinois, Indiana, Ohio West North Central - N. Dakota, 8. Dakota, Nebraska, Kansas, Minnesota, Iowa, Missouri Mid Atlantic South Atlantic Mountain : Montana, Idaho, Wyoming, Nevada, Utah, Colorado, Arizona, New Mexico . .. Pacific : Washington, Oregon, California, Hawaii, Alaska Source: Sumicrast and Frankel (1970). Mean land Development Cost per Front foot by region from Table 29, page 150 and mean lot frontage from table 14, page 110. n Iii l. 61 Research which attempts to explain farm land values has found the following six factors statistically significant: 1) amount of land in farms, 2) farm transfers, 3) number of farms, 4) previous year net farm income, 5) rate of return on nonfarm investment and 6) land prices in the previous year (Healy and Short, 1979). Capital gains or appreciation of value and the expectations of capital appre— ciation are an important part of increasing farm real estate values. Certainly inflation and the increased product prices of 1973-1974, followed by increased farm enlargement, contributed to the increasing farm land values (Herr, 1974). That part of the appreciation affected by farm related factors should influence supply of land for urban uses. As farming becomes more rewarding the supply of land for urban uses should shift and the price increase. Schmid pointed to the following factors which affect the ability or desire of landowners to withhold land from the conversion market. "There is no a priori reason that the reservation or asking price set by sellers should not be found to exceed the present value of actual future values, and no reason that this price could not persist over considerable time, even if there is no overt collusion and no comination by a few sellers." (1968, p.39). This reinforces the previous discussion of the role of expectations. Taxation can also affect the investment potential and affect the capitalization of land values. Finally, the cost of capital and degree of uncertainty and risk can affect the supply of land. Van Vuuren (1976) argued that appreciation gains made by farmers in selling their farms for urban uses is not extensive because the 62 the opportunity cost should also include attachment to and knowledge of the land, transactions costs, etc. While many of these factors may be involved, the size of the appreciation seems to large to be explained solely by these factors. Farm value data is published in two places: The Farm Real Estate Market Reports, published by the U.S. Department of Agriculture,and the U.S. Census of Agriculture. The first exists for the U.S. and states and is available on a yearly basis. The second is done every five years. The data are by county and states. While the Census of Agri- culture data is superior and will be used in the empirical work to follow, success has occurred using The Farm Real Estate Market Reports. Again, Schmid noted that the farm land price rose 150 percent from 1946 to 1964. (1968, p.9). From 1964 to 1976, however, the price of farm land increased by over 300 percent, keeping pace with the percentage increase in site price. This data was presented in Table 3. So while the percentage that farm values represented of site value was 2 percent in 1960, it was about 2.2 percent in 1976. Conceptually the calculation of agricultural opportunity cost should entail finding the value of agricultural land without urban pressures. However, land values reported near urban areas have the urban pressure component included. To estimate the opportunity cost, the land use map of National Atlas of the United States (U.S. Geological Survey, 1970) was examined by counties and the counties listed which had a similar land use (i.e., field crops, irrigated farming, crops and grazing) to the agricultural areas around the urban areas of the state. The mean land values of the agricultural 63 counties in a given state was considered the agricultural opportunity cost for cities in that state. A finer, city by city, analysis was not possible without detailing the agricultural land uses around each city; a costly venture. In Michigan, for example, seventeen counties were listed.9 The average land and building value of these counties was $563 while the average for the state was $459. This conceivably indicates that the better agricultural land is around urban areas, which is the case in the lower peninsula of Michigan. The process of selecting the counties was essentially by examination of the land use map.10 This process seemed to expatiate several existent problems with state averages such as the inclusion of much forest or desert land and, in very urban states such as New Jersey, excessive inclusion of urban land. These counties, calculated from the 1969 Census of Agriculture, were also used for 1959, 1964 and 1974 calculations used in the pooled cross-sectional time series analysis.11 See Appendix C. 9The counties were Allegan, Barry, Branch, Calhoun, Cass, Grat- iot, Hillsdale, Ionia, Isabella, Lapeer, Lenavee, Mecosta, Montcalm, St. Joseph, Sanilac, Tuscola, and Van Buren. 10In a few instances prior knowledge by the author in such states as Oregon and Maryland and the knowledge of fellow graduate students supplemented the process. In one case, Colorado, a fellow graduate student described the sale by his father of a farm undergoing urban pressure and the purchase for quite a bit lower price of a similar farm further from the urban area. llBecause of changes in the definition of a farm reported in the regular Census of Agriculture in 1974, the preliminary report data was used for calculating agricultural opportunity cost. The defini- tions of farms were, therefore, consistent for 1959, 1964, 1969, and 1974. 64 Appreciation Site price appreciation is calculated as the percent appreciation over agricultural opportunity costs after development costs and agricul- tural opportunity cost adjusted for site size, are deducted. Thirty- six cities were common to both FHA and NAHB data. For these cities the calculation of appreciation from NAHB data is shown in Table 9 and the calculation of appreciation from FHA data is shown in Table 10. Site price and appreciation from both sources are summarized in Table 11 for thirty-six cities in 1969. Birmingham, Alabama will be used to demonstrate the calculation. The NAHB site price for Birmingham was $5451 in 1969. Development costs are estimated for Alabama as $2430 and the calculated agricul— tural opportunity cost for Alabama was $82, both for 1969. Development and agricultural opportunity costs are added together for a total of $2512 and then subtracted from site price leaving a difference of $2938. Appreciation is then calculated with respect to development costs at 117 percent and agricultural opportunity costs at 3570 percent. Appreciation varies between the FHA and NAHB data not solely because of site price. Agricultural opportunity cost is adjusted by 2.6 lots per acre for NAHB data and appropriate metropolitan area site size for the FHA data. Development cost from the NAHB source was adjusted on a national basis to FHA sites. Therefore, FHA site size information is likely to be more reliable than the constant 2.6 lots per acre assumed in the NAHB calculations for all cities. 0n the other hand, the development cost data from NAHB sources, while still a poor approximation across metropolitan 65 .o 5.35.72 ..— 1on—nsEE:w .Er. >_ ..e >a._z=...ze..._c ._<~=:.._:U~ zc< z..._>c zo-.—.< —Uu~_._._< b.2383: U as..Acc-.:. memo: au>c z:.e<_cm=aa< ezavau. ..— A=»e.z=ezoaao meme; hzezaq.m>m= c mm.e_u x_m->e=_=e area .c=< zb:.<~uux..:< MUS..— ms—Lm a man—SH I .2...»va seesaw 1:: uwZ—u .5»... 3:. .NN— .: .mu .nn .mn .mn .nc- .9: .nc— whmsu >._._ 2355...... ._<==.—.._:U _ =c< = as: <=L MU:E MPH—m .>_ seas—Eu 5 12.1.1...“- 5 nous—Sm :=c: use—.0 ..— 10:22.35 r... anus—=35. mdccc >~_.=S..9_._c _:.:5_:u:a< so... moo—u; mug—m Eocm:..§. 13.957. 355:... .— >..u ox;— ..em 3:522 :tv. .5225: ..Zc: J... a... ...c o___>:m:z m_—_>xc:x 55—5.: .7. 9...... Tee...— 1... 1::— :2.— 5.2.... .3 .3 5.5;...— ..c c: 1...: 255—... 3. Hut... 5.5:: 3.5.. .5. .3 .u E... J. ...— 1.5.5:...2 :Escz 1.5.... a 7......0: e.g.... :5. Sic—u: 3oz n.— _ 23:...— 1:93:51:— cuce. .5 nu::_.< 5.5:... :c..u._._=: .3 .525: :um.u.:.£..._ :5.» Ca-.D:.flh.:wm c. 2:: 5.2.555.— :— .c» .f‘ .x w~c=fi can cm~< .aCNm.. .ox::.. 1:: umuuu.E:m so». .u x.1=c:;< =u sen—hueeam 1:: >~ uwuaw:u =4 tv:—=~axw m: oucum >; =w.:.=o.=c can: mazcu .:u§::.a>¢c . c ...csu.c:;c .=L:..=u .m -;:.- .o.a.w:.u.=a.. 1:: .m:.u.5=m so.. may... by.“ .ccu.:c . m .cch ..e .ocs. .n.o~ ..maw .n . .»I;.«I:,.JI aims :r::.a .53. .i .3: .32 .2... .nw .MMMM his... .5 . .amnu . .Nn .Nac. .e..~ ...c~ .nc .a:~q .. . c. m ...z .m. .ccmm .mc ..va .Nam— .mmm— .~¢ .mncN ..W :w:. ..:m .b“m: .c. .cNQ ..e ..m.. .amo. .a... .c.. ...cm o.._>x:.” .5. ...cw .ce .9... .n=.~ .nmn~ .c. ..Nman :a.=am..__ .WN ...nm .N. ..oo. .NoHN .HMHN ..m .9... =.;..;==__;; ..N .cun .:~ .ewm .NeON .—mw~ ..~. .~..U m=::= .N. .oan. ..w .ecm. ..HNN .un.w .ao .no.. z_.c.ao:=_= .e. .cma. .we ...e. .mm¢~ .maMN .om .co.¢ :1..cz ===.c .m. .wc<~ .Nw .~.~. .am.~ ..m.~ .nm .-.m ..s..cc .._ .am.~ .c.. .NoNN .Nco. .m.m. .mm .qewq u.os_..:= 4. .que .GNN .eamm Hacou .emm— .mm .v¢.m m:=c..o 3L2 .N. (u .nemN .m. .emn. .Now. .9... .nn .om.n o...>m.:=. .. (U .ade .n~ .m~¢ .waew ..omw .aa .N.0N m..ca::n.1:. .o. .cma~ ..o .o.n. ..¢~ ..anu .Q. .Nwom oxau.;u .a .cmm. .mc .a¢n~ .ewm— .ahwd .nw .m—nn c.=c—.< .w ..o.. .Nq ...w ..n:. ...m. .mn ..n.~ «game ; .nmm— .nn .¢m¢~ ..mm— .ONEu .nm ..OHH =c.u=.E..3 .c .e.w. .cc ..~¢. ..acn .n.o~ .w. .m_mm .c>=oc .m .e... .o.. .aa.n .~nm~ .omaN ..a ..nne cum_u=:.. ==m .. .¢m- .nc .m¢w— .DmaN .CmmN .NQ .nQNv .......:=c._..fim .h .~m- .ee .am.. .o~n— .o... .e. .oxmn u._;:: .N ...a ..n .<.e .m.a. .a<.. .e. ..mqm Ea;.:_a..= .. .N. .N. .0. Am. .m. .w. A». co...=.-.u. .mcu ...zapzc..a . .o_=. .<.=h.=u.=c< oc...=.-.=. pwoo quz.c.m>u: =u>c mpmcu =a>o . + mhmcu z:..<.ou...< zc..<.u....< .:-<. amen .s.z=b¢c..c mpmou ...z=.zc..o so... Hzgczu. pzauxm. zc.b<.uw...< .<==h4=u..c< hzu=.:..>.= .y=.=p mca. .:n ez< smoo ...zzhzc..c .<¢=b.=u.zu< u>oe< zo.h<.um¢..< mu... a..m m. m.a«m . 67 TABLE 11 COMPARISON OF LAND VALUE APPRECIATION AND MARKET PRICE OF SITE FOR THIRTY-SIX CITIES COMMON TO BOTH NATIONAL ASSOCIATION OF HOME BUILDERS AND FEDERAL HOUSING ADMINISTRATION DATA 1969 SITE SITE PRICE PRICE APPRECIATION APPRECIATION CITY NAHB ($) FHA ($) NAHB (Z) FHA (Z) 1. Birmingham 5451. 2343. 3570. 879. 2. Mobile 4507. 3586. 2423. 2252. 3. Sacramento 5798. 4783. 906. 2130. 4. San Francisco 10478. 6731. 3468. 4710. 5. Denver 5877. 3518. 8102. 7976. 6. Wilmington 8875. 3391. 3982. 1852. 7. Tampa 4371. 2754. 1339. 1401. 8. Atlanta 7281. 3313. 4681. 1587. 9. Chicago 9847. 3982. 3505. 2050. 10. Indianapolis 5897. 2912. 1564. 419. 11. Louisville 6543. 3159. 4061. 2543. 12. New Orleans 7021. 5144. 3225. 6472. 13. Baltimore 5931. 4244. 1662. 2758. 14. Detroit 8172. 3722. 4110. 2407. 15. Grand Rapids 4500. 4160. 917. 1955. 16. Miami 7240. 4103. 3295. 1889. 17. Kansas City 5101. 3130. 2199. 1568. 18. St. Louis 7027. 3607. 4276. 3003. 19. Omaha 4918. 3148. 1622. 1547. 20. Buffalo 5231. 3445. 1631. 1677. 21. Cincinnati 7781. 4368. 2859. 2493. 22. Dayton 8227. 4479. 3156. 2109. 23. Oklahoma 5307. 2442. 3653. 1909. 24. Tulsa 5275. 2801. ‘ 3615. 2709. 25. Portland 5515. 3547. 895. 524. 26. Philadelphia 7068. 4359. 3429. 3330. 27. Pittsburgh 7441. 3852. 3773. 2071. 28. Knoxville 3117. 3047. 533. 826. 29. Nashville 4384. 3280. 1701. 1521. 30. Dallas 6410. 3182. 5537. 4099. 31. Ft. Worth 3546. 2722. 1738. 2652. 32. Houston 5428. 3403. 4235. 4874. 33. San Antonio 3757. 2678. 2018. 2580. 34. Salt Lake City 7034. 3209. 5675. 2559. 35. Richmond 5663. 3428. 2305. 1535. 36. Seattle 6274. 4215. 1925. 2049. Sources: Sumicrast and Frankel (1970) for NAHB data. FHA, Data for Cities and States, 1969. Also see Table 9 for NAHB data and Table 10 for FHA data. 68 areas of variations, would seem more reliable for the NAHB calculations than for the FHA calculations, given that the FHA data is from a different segment of the market. Ideally, site size, development cost, and agricultural opportunity cost Would exist for each metro- politan area. Lacking that ideal, the best available data for each data set has been used. ngmary of the Land Conversion Process There are a series of steps in the land conversion process which will be summarized from the sections above. It is generally assumed that most land converted to urban residential uses was initially farm land. The active farm value is the initial point of departure, but the opportunity cost of alterna- tives available to the farmer in terms of farms with similar agriculture not under urban pressure represents the base from which the process begins conceptually. In this research an effort is made to estimate the value of land and buildings, though land alone would have been preferred, of farms similar to farms under urban pressure, at least by State. If the farmer sells to a speculator or holds on to the land for Speculative purposes, the role of expectations with respect to the general and specific growth of the metropolitan will come into play. Economic rent arises from this urban population growth. The ability of the farmer or non-farmer speculator to keep land off the market until prices rise depends upon various holding costs such as, among others, the property tax or in some cases sewer charges. 69 The next step is the purchase by the subdivider. After a holding period which can vary substantially from several months to three years, the developer puts in government mandated or consumer demanded improvements for urban uses. (Sumicrast and Frankel, 1970). The developer then sells the land, or perhaps, builds a house on it. The size of the site will often depend upon the price (and vice versa) and development costs will vary according to demands from buyers and local jurisdictions for different improvements. Appreciation represents both economic rent and monopoly profit after agricultural opportunity cost and the normal returns and development costs are subtracted from site price. The chapters to follow empirically attempt to explain the variation across cities and over time for site price and appreciation for both the FHA and NAHB data. While this is an effort to discern supply, demand, and instrumental features, it is unlikely in the near future, given the data situation, that a single model will be successfully identified. CHAPTER V CROSS-SECTIONAL REGRESSION ANALYSIS Introduction This chapter consists of three primary sections. The first section introduces the operational form of the independent variables. The second section estimates a cross-sectional regression model for site price and the third section estimates a cross—sectional regres- sion model for appreciation. A brief analysis of sewer financing precedes the conclusion. There are four objectives of the cross-sectional regression work to follow. First, more recent data will be used to complement earlier results for basically the same models. The estimates of the models with more recent data will be compared to earlier results and changes noted. Second, the analysis will entail use of both site price data and the calculated appreciation estimates. This will allow for continuity with most previous econometric,research. Third, variables associated with the institutional issues of interest in this research will be introduced. This will include property taxation, sewer provision, and sewer financing variables. In addition, some variables thought to have institutional significance in the other econometric studies, reviewed in Chapter III, will be examined. 71 Fourth, and last, the analysis will use both FHA and NAHB data for the dependent variables. This will highlight variations in the data sets which could influence further analysis which uses one or the other source of data. Independent Variables Chapter IV developed the city variation and trends over time for site price and appreciation, the dependent variables in this research. This section will present a detailed description of the independent variables and the hypothesized relationship to the dependent variables, consistent with the theory presented in Chapter II and the literature review presented in Chapter III. This section will be concluded by a formal presentation of the site price and appre- ciation models. Cross-sectional regression analysis will follow. The independent variables can be roughly categorized into three groups. The first group consists of general demand variables such as total population, change in population, and mean family income. These variables appeared in the earlier research reviewed in Chapter III. The second group is composed of characteristics implicit in the land site such as site size, development cost, and agricultural opportunity cost. These variables are used in part to calculate appreciation from site price. The third group consists of the instrumental variables related to sewer provision and tax policies. The variables are instrumental in that the variable reflects policies subject to political decisions. Each of the independent variables in these three groups is discussed below. Each variable is also 72 summarized for both the cross-sectional and pooled regression analysis in Table 12. General demand characteristics are introduced into the model through the use of variables for total population, percentage change in population, mean or median family income and a dummy variable to represent cities in California. Each of these is discussed below. Size of the metropolitan population encompasses several aspects of demand. Larger cities cost more to live in and often provide more amenities so prices in general are higher the greater the total population. Moreover, because of the generally larger area or 1 increased congestion, some amenities such as schools, businesses, parks, cultural activities will be spread further from a given location or site. Lots on the fringe of a larger city will be further away from some of the amenities in the metropolitan area whether the metropolitan area is monocentric, policentric or even banana shaped. For example, a Detroit residential fringe lot is further away from the Renaissance Center in downtown Detroit than a fringe lot in metro- politan Kalamazoo is from the Kalamazoo Convention Center. More undeveloped land is within easy reach of most metropolitan amenities in a smaller city and traffic congestion is often less of a problem. Indeed, there may be fewer amenities in general in the smaller city which, therefore, could mean less demand for close-in fringe land. It should be noted, however, that the shape or structure of the metropolitan area does introduce some ambiquity into this relationship. The dispersal of amenities throughout a large metropolitan area will also disperse demand at the fringe or expand the area considered the fringe. The exact size and implications of this effect on site 73 TABLE 12 DESCRIPTION AND SOURCE OF INDEPENDENT VARIAB¥ES USED IN SITE PRICE AND APPRECIATION MODEIS Total Pepulation (persons), Urbanized Area, 1970, U.S. Census of Population, 1970. _ Change in Popplation (percent), Urbanized Area, 1960-1970, U.S. Census of Population, 1970. Mean Family Income or Median Family Income (dollars), Urbanized Area, 1970, U.S. Census of Population, 1970. Calculated Agpicultural Opportunity Cost per acre (dollars), 1959, 1964, 19 9, 197 , U.S. Census of Agriculture. Calculated for counties which have similar use as urbanized counties but are free of urbanizing value impacts. See Chapter IV for detailed discussion. Development or Improvement Costs per lot (dollars), 1969, National Association of Home Builders (Profile of the Builder and His Industry) 1970. See Chapter IV for detailed discussion and Table 8 for calculation by U.S. region. California Dummy Qualitative dummy variable where value of one if city is in state of California and zero other wise. Percent All (Single Family) Homes With PUblic Sewer, SMSA's and places over 100,000 persons, 1970. U.S. Census of Housing, 1970. Percent New (Single Family) Homes With Public Sewers, SMSA's, 1957—1975, FHA Data for States and Selected Areas 1967-1976. SMSA's reported each year varied because of sample size criterion. Prpperty Tax Propprtion of General Revenue (percent), 1967, Census of Local Government, 19 7. Site Size (square feet), 1967-1976, SMSA's, FHA Data for States and Selected Areas, 1967—1976. SMSA's reported each year varied because of sample size criterion. J————i [A K. L. 74 Property Tax Rapge: Calculated by subtracting low county average per capita property tax in SMSA from high county average per capita property tax in SMSA and dividing by overall SMSA average per capita property tax thus resulting in a relative range. Where one county consists of entire SMSA, the range is zero. Census of Local Government, Financial Characteristics of Selected Metropolitan Areas, 1964-1975. Sewer Range: Calculated the same as the property tax range but uses the per capita sewer capital outlay, high, low, and overall averages. Census of Local Government, Financial Characteristics of Selected Metropolitan Areas, 1964—1975. lVariables used in the pooled cross-sectional time series regressions include total population, percent of new homes sewered, property tax range, sewer range, development costs, site size, and agricultural opportunity costs. When data were not available for a specific year, linear interpolations or extrapolations were made. Selection of the number of cities and years chosen for analysis was made based on minimizing interpolations or extrapolations because of their effect on the degrees of freedom. IlIIIIIIIIIIIIII-------'------aiiiiiiEE=a--I-----------------‘.llq 75 price or appreciation is unclear given interaction with other variables associated with total population such as income. Percent change in population can be viewed as both an indication of recent past increases in demand and as a portent for future growth. The greater the percentage increment in population the greater will be the demand,hence, price for residential lots. The degree of that effect could depend on the housing stock, etc. but this might be mitigated somewhat by the ten year period used in the population change variable. The initial housing stock effect might wash out over the ten year period. Furthermore, landowners at the fringe would naturally look at past growth trends as an indication of future growth and economic rent. As expectation of increasing demand is greater, then some greater proportion of the supply of land is withheld in antici- pation of increased prices, increasing current prices. The best form of this concept would be an indication or projection of future growth. The mean or median family income variable indicates by size the strength of demand or buying power of an area. People with higher incomes are able to pay or bid more for lots with more amenities or locational advantages. It's generally recognized that the mean weights the extreme values more heavily while the median is more stable and generally lower than the mean. The implications of the two income variables will be discussed with the results. Regional variation in population growth, weather, demography, etc. could lead to variations in site price and appreciation either through demand or development costs associated with weather, input costs, etc. Preliminary statistical analysis indicated the 76 possibility of land market conditions in California varying in size if not characteristically from the rest of the country. Therefore, limited examination of regional variations was indicated by the use of a dummy variable for cities in the State of California. The second group of variables concern site characteristics and include development costs, agricultural opportunity cost, and site size. There are two possible interpretations of the relationship betWeen development costs and the dependent variables. If, as is generally assumed in this research, the supply of sites is fixed, then price is demand determined and appreciation is a residual affected by the size of development costs as demonstrated in Figure 2. The supply of sites is unaffected by development costs while appreciation is inversely related to development costs. 0n the other hand, development costs could affect the supply of sites as shown in Figure 3. Development costs in this case are positively related to site price and appreciation. High development costs, in this case, would be associated with a smaller supply and higher prices farsites and” presumably, greater appreciation. This research will use the assumption of a fixed supply where price is demand determined while recognizing that the assumption of supply determined by development costs could also be valid. Agricultural opportunity cost should be positively related to site price because of the higher price needed for urban uses in order to meet the offer price of the landowner to cover opportunity costs. If the variable definition has captured any demand characteristics, those characteristics will add to the positive relationship. rnuflfillmim- I. 77 FIGURE 2. DEVELOPMENT COSTS WITH THE SUPPLY OF SITES FIXED P Appre 'ation Development Cost D, agricul- D, urban tural use sites 0 Q. FIGURE 3. DEVELOPMENT COSTS AS DETERMINING SITE SUPPLY S, high development costs S, low D, urban sites tural use % I 78 Finally, over time, the size of a lot has decreased as the price of the site has increased. Moreover, it can be assumed that across cities that as the price of the site increases the size of the lot will decrease, ceteris paribus. I The instrumental variables used in the cross—sectional analysis include percent all homes with public sewer, percent new homes with public sewer, the property tax proportion of general revenue (percent), the property tax range and the sewer range. Either percent all homes with public sewer or percent new homes with public sewer could indicate the sewer supply policies followed in a metropolitan area. Percent all homes with public sewer could indicate the cumulative influence and historical policies of local jurisdictions while percent new homes with public sewer would indicate recent policy. The greater the percentage of either all or new homes with public sewer, the more likely that sewers are being supplied which require hook ups. Site price and appreciation should be higher if this is the case. On the other hand, liberal provision of sewer services when there is demand could be the policy followed. While the greater supply of public sewers could lead to scattered residential location as well as an under-supply, the raw land price might be less with over—supply as the increased improvement costs show up in the site price. On the other hand, under-supply also raises prices. Therefore, while either of these variables could indicate past sewer supply policies, the variables are somewhat ambiguous on price and appreciation. However, the role of sewer costs in site improvement might override the supply characteristics and indicate that a larger percent of lots with public sewer would lead to a higher —7—f - _._._.__ . 79 site price. It should be noted, though, that if this facet of sewer provision is important, it could lead to smaller appreciation. Care should be taken with these variables. The lack of sewer supply variable data for new sites results in the use of these poor proxies. The property tax variables include the property tax proportion of general revenue (percent) and the relative range of average per capita property tax paid across counties in a metropolitan area. The property tax proportion of general revenue is an indication of the importance of the property tax vis-a-vis other local government financing methods. The property tax is a less direct way of financing new infrastructure than connection fees or service charges so the greater the use of the property tax the greater raw land prices will be. According to Clawson, Fthe extent that the house purchaser evades any of the costs of public services to his property, the raw land price will be higher than if he had to pay them." (1971, p.162). The percentage of sewer financing with user charges is also tested for data available for seventeen cities in 1960. But also note, that high property taxes, per se, lowers home value and, therefore, site value. True value of a home includes site value and construction costs. Therefore higher property tax reduces derived demand for homes but may decrease the cost of development. The property tax range was calculated by subtracting the low county average per capita property tax in the SMSA from the high county average per capita property tax in the SMSA and dividing by the overall SMSA average per capita property tax thus resulting in a relative range. Where one county consists of entire SMSA the range is zero. The sewer 80 range is calculated the same as the property tax range but uses the per capita sewer capital outlay high, low, and overall averages. Since the basic units are the counties in SMSA's, the great variation in county size from one SMSA to another SMSA raises serious issues about the usefulness of the relative range in this context. The range of the average per capita property tax can be indica- tive of two phenomena. First, the range may indicate the diSparity between the central city and the suburbs. If, as seems likely, the central city has the greatest average per capita property taxes paid then the flight to the low tax suburbs could be indicated. This would indicate that the greater the range the greater the demand, hence, price. Also captured in the range is variation across subjurisdictions. It is, however, a crude measure since the range is across counties in an SMSA, where as many jurisdictions are involved. As a proxy for the variance which cannot be compared across different populations because distributions cannot be assumed the same and small number of cases for some cities, the range captures several.effects which make the inter- pretation of the relationships of interest ambiguous. However, as the average per capita property tax is related to income and demands for amenities associated with the property tax, it can also be assumed that the range can also be indicative of income and have a positive relationship to site price and appreciation. The range of the average per capita sewer capital outlay should indicate another supply variable. If the central city can be considered to be completely sewered (not counting replacement), then the greater the range the greater the current outlay on sewers in 81 the fringe. Following the theory of the other sewer supply variables, the greater the supply the greater the site price and appreciation. A final note is needed to point to some of the interactions between the independent variables. For example, there should be a strong correlation between sewer supply variables and the total population of the metropolitan area. As population becomes greater and generally denser, the need for public sewers becomes greater because of the inability of the land to absorb waste with septic tanks and other techniques. Tax variables could be highly correlated with income. As population and income increase, the demands for services also increase and so taxes should also increase. With a greater income range the variation in demand (and tastes) might also be reflected in tax and sewer range variables. This review of the Operationalization of the model and variables can be summarized for both the site price and appreciation models in the following equations: Expected Site Price ‘= a , Relationship Bl Total POpulation (persons) + B2 Change in Population (percent) + B3 Mean or Median Family Income (dollars) + + 34 California Dummy B5 Site Size (square feet) B6 Improvement Costs (dollars per site) - B7 Agricultural Opportunity Cost (dollars) + B8 Property Tax Proportion of General Revenue (percent) + B9 Property Tax Range + Blo Sewer Range + B All Single Family Homes with Public Sewer(%) or ‘2 l . . 1 New Single Family Homes w1th Public Sewer(%) ? 82 where a = Constant to be estimated. B1 to 311:: Coefficients associated with specified variables. The appreciation model below consists of two groups of variables. The demand variables used in the site price model are carried over since demand is a key determinant of rent and profit, the theoretical components of appreciation. In Chapter IV, site size, development cost, and agricultural opportunity cost were used to calculate appreciation from site price. These variables cannot be used to explain appreciation since they were used to calculate appreciation: site size less agricultural opportunity cost (adjusted by site size) and less development cost (adjusted by site size) is equal to land value appreciation. Several variables will be tested only in the site price model. The California Dummy and the Property Tax Proportion of General Revenue (percent) were not considered accurate enough variables to be tested in both models unless they were reasonably strong in the site price model. In this case, as will be demonstrated, neither variable was statistically significant. The other instrumental variables are the property tax range, sewer range and percent all or new homes with public sewers. These variables are intended to explain appreciation via the hypothesized relationships associated with public policies restricting supply and leading to profit. 3:3 83 The model is: Expected Appreciation = a Sign +Bl Total Population (persons) +B2 Change in Population (percent) + ‘tBB Mean or Median Family Income (dollars) + +B4 Property Tax Range + -+B5 Sewer Range + -+B6 All Single Family Homes with Public ? Sewer (percent) —or— New Single Family Homes with Public Sewer ? Where: a = Constant to be Estimated. B1 to 36:: Coefficients Associated with Specified Variables. Site Price Model Introduction Both the site price and appreciation models were estimated by ordinary least squares. Because of missing data for some variables, the entire model is not examined in any one equation. Statistically insignificant variables are dropped so that the equations which follow have fewer independent variables than they would normally have if the entire model were examined in one equation, hence gaining degrees of freedom. This section builds upon precedents discussed in Chapter III. There was, as earlier, no strong a priori suspicion of heterosked- asticity in the cross-sectional analysis. An examination of the residuals did not indicate a problem. However, as with previous research, there is an indication of some multicollinearity. While 84 this raises issues about the interpretation of the variables it was not judged serious. The independent variables thought to be associated with site prices are examined with data from FHA.and NAHB for 1969. The results of the fourteen equations used to estimate the site price model are displayed in Tables 13, 14 and 15. The following analysis consists of two approaches. First, a step by step description of the selection criteria for dropping or adding variables will be made. This will proceed through the model using NAHB data and the variables related to that model. Then the same will be done using FHA data. This will be followed by regressions of the same independent variables and same observations for each data set. Secondly, analysis will be made of the statistical significance and sign of the independent variables as reflected in the fourteen equations. Description of the Method To reiterate, there are two reasons for the sequentially estimated equations to follow: 1) to test relationships with the largest number of cases possible and 2) to examine issues raised in earlier econometric work. This process will now be briefly described. Ottensmann (1977) estimated an equation using 1964 NAHB site data similar to the equation estimated in Table 13, regression one, which is, however, for 1969 NAHB site data. The results are similar to those of Ottensmann with significance levels of 5.0001 for total population, 5.0001 for income and .478 for change in population. Ottensmann's coefficient of determination (R2) was .53 for 1964 compared to .42 for the 1969 data used in this section. The signs were positive. ‘—"'__ ‘ 85 4:0» 0.. 05...: 0.... e afieu0>c 1:0 .39# .suq: ae—uzo ~euuaeu u030m eudaeo H0; 0:u can: use 0u=au xea saunas»: me 05:: 0rd cc.:_:;_:c 9: A: .ouon 0u03 ooauuu surge -a p=u 0:0 000: unencuuqeu no 0ueam 0su :u =0_uqu onus: oasc.ee> as ess. ess. ess. ess. acoo.w. Nae. sac. A.aco—v a~cnm.s_—v Amomc.qcnv Acmee.euv Aomec.dv Aocm~.v chco.v an am. No. .ewcm: mmao.m_n . qcec.s——N eosn.—e ~a~o.a once. macs. awn. cam. sea. sac. acccaw ¢_n. was. A.eesao Anse.-_o Ae~_~.ae~c .Nsee.n—. A-ea._c Amae~.~_o A~=s=.. on um. am. .ccn—x comm.~- anew.ooo~ eccn.ec ooa~.m chc.¢: macs. noseuw nee. see. acec.w ~_c. .ccc.w A.asmc Anenn.memv Aceo~.o Aomns.o Andam.ec A_ooc.o Ned cc. Ce. .mnmN ¢¢n¢.ODI ach. ONm—.m N:N9.D— OCDC. use. .oocaw see. .cao.w 292.3 83:; 8.3:: Cece; Nee es. ~e. ee.~m- seen when n ess: Ilatilzwmww acau Auc0uu0mv uamiaa Anus—dovv Amueducvv Anus—«ovv Aucoummuw {tnNNmmmHmmv a0w=nx ~0w=m¢ nah chad s030w coon uuecu aeo— acou Geo. oeoocu o~o«;cc:_ cha— u—ucaa :uqz e.guouuqso u=0anc~0>0c saucsuuoeao a—_Eeu cc—ac_:;oa :c..:—:acm 1-:Igtu039m mau050ua a0Ec= -< nosey—scuuw< :u—toz 2. 0w:s:0 —eucs more: x x assumecc ~o>0n oozsuduucw—m .m—e0;u:0uea =~ gonna peep:mum .ncofio—uuocu ecdezouuox Ho u w: N mo~n=—us> u:01:0a0p:. 'lnlllli.‘.- ’cll‘tl. it‘ll: lllll'lnl‘t'u‘l‘ l Ilrlll I "IIII mm:sas~ue—0u e s“ manu~=w0u a::u so» susoaCaa eu_acu so: oueuo>e c m=~tu>qe e:s e ausacu :u—s acne c au:=cu zoq w:_uu:u.;:a a: to.s_:t.:h N .5... 0>. :3 . .55 - .- . ...5... ::_mecuw0= 90.. .Z. :c_c30n§0= 33h ==._;%V.ucz 0:: :c.mc.....uc~_ Amcc._ctv ace. mh07-_:c 6E3: be comet—:cuc< —==c_ucz 9.; 5 etc. ..r.._:....g :: _ autumn: \0_;n_u:> €83.93: 86 This equation provides the initial reference point for the other equations to follow. It also indicates problems mentioned in the review of other research on the correlation between some of the independent variables. Specifically the simple correlation of income with total population was .40 and .21 for percent change in population. Table 13, regression two, was estimated, therefore, without income but with agricultural opportunity cost, development cost, and a California dummy added for the 162 cases. The results showed percent change in population now statistically significant at the .017 level and agricultural opportunity cost and total population statistically significant at the 5.0001 level each. All of the signs were positive. Development cost and the California dummy were insignificant, hence, dropped. Both of these variables could be considered as proxies for regional variations across the U.S. Their lack of statistical significance might indicate the need for an even finer grain in regional analysis. A conclusion supporting the importance of regional variation came from comments by developers interested in national variations in development costs (Chapter VIII). Regression three of Table 13 included only fifty—six cities with the addition of percent all with public sewer, property tax range and sewer range as well as total population, percent change in population and agricultural opportunity cost. As expected, total population and agricultural opportunity cost were significant (.008 and.g.OOOl, respectively) as well as percent all homes with public sewer (.007) and the property tax range (.007). All of the significant variables had positive coefficients. Sewer range was insignificant as well as change in population. Regression four,Table l3,replicated regression 87 three save for the replacement of change in population by median family income. This was done because of the possibility of income factors entering with the new variables. The addition of income not only increased the R2 from .57 to .62 but income was statistically significant at the .012 level and had a positive coefficient. Table 14 traces four equations estimated for 1969 FHA site price data. Regression one of Table 14 begins the FHA site price analysis for 104 cities. The independent variables were total population, change in population, mean family income, agricultural opportunity cost, site size and the property tax proportion of general revenue. Total population, change in population and agricultural Opportunity costs were Significant at the 2.0001, .001, g .0001 levels reSpectively. Mean family income and percent new homes with public sewer were significant at the .165 and .013 levels and carried over to the next equation. The signs of these five variables were, as hypothesized, positive. Site size and the property tax proportion of general revenue were significant at the .401 and .456 levels and not carried further. Regression two in the FHA site price analysis added the prOperty tax range and sewer range to the five variables carried over. There were fifty cities in regressions two, three, and four. The R2 dropped from regression one (.55) to regression two (.51). Total population and agricultural opportunity cost were significant at .019 and 550001 levels respectively. Change in population declined to a significance level of .125. Percent new homes with public sewer was significant at the .318 level with a positive coefficient. The income, property tax range and sewer range variables were insignificant. .s0mmu0>s ~a~u0>c ucs red .;Mu; aways: uu—qu can 0:» m0m= u=a 0w==u xzu sameness 0;“ me 05mm o:a =0L:_:;.:; .cHeu nu 0m:wn 0;a u==ou 0:o 0n0:3 .0u:ua 0>~uu~0n u :. w=¢u—=m0u e::d xna sunoncu; :g_::c no; ouuu0>= c 0;u x; u=_=.>d1 ccm uu0gcu; ea_azv .0; oweu0>n au:=cu :wu: ecu . e ss=:eu zed u=_aeete;=e as eereiwimiw. w ass. ”as. can. ”AN. aeoo.w see. ems. e: A.meeo geese.noao A~ne~.e<~o Aeqoe.~.v Anaem.~o Aieas.__o Amcccc.o . he‘s as es. an. .see enem.e_- mesa.n- seee.n. aena.h sace.ea ..ceo. =c_eeer e: one. ens. ome. sn~. deco.” moo. ace. , Q0 A.cshc Assem.me_o Aneea._e~o Anaee.no Asc~m.ao Aimes._lo Amcoss.o .eewee 90 en es. in. .eena ~m~e.s- hasm.ea econ.e s~me.e caee.e_ e_cco. =e_h... as use. see. ecu. was. noso.w ,nce. , m~_. sec. A.=e_eoa-a~.e=_o AeNse.ee~o seems.nv AeNee.aC A~sn_.. Aeeoo.wao Amcooo.v see an as. an. .haoi ceme.h_- Noen.na sane.“ aaee.e News. NINs.sa s_ccs. =e_eee.aea age. n.s. one. .ce. soco.w me_. _oc. .ecc.w . . A.aecv Righa.nv Aa~ma.~o Acans.o Aoae~.av AeNee.v A_ese.sc Aesecc.v . ize so. am mm. .eam- osso.ol ess~.~y seas. enss.e ae~.. meee.a_ Niece. =e_ee..eee i|wmww «Illwmmfl Nummwmwuw Nmruunud% IIINwmmmmwmw IIAmMMH Amumfifiovv Amuea~0uv Au:oouu;v, Am::muo;v Aztc_.:=v eco— Neseez lumeee see. sees cues u==e>e= oue=ehc esea taco cae~ seea-oee_ oae_ =c_heree_ eczuw xnh uuzam nozom uc~0:0c we area >u«::aue;;o msooc— =c_u:_:;ca :c_o:—:sca ::_E=< n:_u::: zunoacua u.—a:; u—ua=m =ciuuoaoua 0n—m men=u—:v«uw< a—_Ecm cu ew:c:u .euch _:te=oa .cu_LL :a—Z sud: xeh suuwccum 0n_m :co: 0u_m nexus: namsv mesa: -< ¢0Eo= :02 ac ~ x z u:cum=ou n0;E:z :ounmoumwz N1 N \0—;=_uc> u:c=:c;n: acan .<=m .mzc—mwumUm: mums; Sham «q mgnu d—cuu>c so: .Jo~ ”coon a. once» 0;u uuc~0u u :— acqufisauu mega xcu auuoaoha ea_a=c C0; ouctc>c : >5 aucacu :wu: loin e auczcu 3c— m:_uo=ta;:c >5 ton=_:c_:: d weN. . . -aq::xcse---...--.-. - A.canaV me.a my ANA” ~oc. man. one. emu. .lne.- . e and any -s~.~ .Ne.=~v Aeema.o A..oee.. ..m ea .mfl ma.~a same.e em.e_- eeee. n_eee.- ee.eee.e.= <=t ace. . A.aeo~o so». Nae. Nae. s-. was. .eN. .Aeee- .aeeo Ase.~a. geese.~o Aen.e_o Aae-.. A~e_eee.o e._e sewn aa.@¢ encn.m ec.cml «Nam. QH~OOD. ::_rctuuo= =:0; 0uzeu—u—suuw .eumosuc0ua; cu manna euoeceum .u:0_u_uuocu :c—ceouwox 503.....2 :: _ Eric at: e0—seuue> u=0ecoa0ccd \ccazcm c.e: Iota. . N0—;=.»c> usctzoaca aca~ ash <=u cz< n:huu=k zOL szHmmamumx mU—KL whum 'llun‘ll'l'lll nu udnn -uuu>o ace .3o~ .:m.: zm—uao use—mac nwiem :u—auo he; may mum: a: a ux==n xsu synago5: u:a we came 9:4 cans—:c._ N . .oums a. 62::n uu>= e ess >4 u:_=_>_c gnu nu: em: = . o;u uuadon u :« wcuudzmuu m=;a x5. manage»; mu. m. uosbuaac a: =Qa=_:L_su E? SE 5.5 5 53 3.2.3; 525 3.. 3.52:. 35.8 :3: so: fizm 5 x5 35%... 31.5 .2. $526 3.58 :3 fil...lmll-l¥t«q$¥t : . _ coN. mNm. “No. aces.“ e.<. eachmmuc atr;e A-.a~v Aso_s.ac Am-~.n. Amsag.v Ammsascny Aaswmsosecc. =c_¢=e_u;= <=a on “N. as.s~- cano.~- o«~s.a owes. l~s_oo - as . see. one. ans. __n. c~s. H~s. . cab Asses... Aeneas.av Anmng.nv Ansso.v Assn. V Ahmmcccoo=.v =c.n=¢_uo= <=l on sN. NM. g~s~.~| na_s~.- sass.n ns_~. NaNs. Nassceccs can. one. —c¢c.w =CCH 892v 232:5; GEE; £2382. v : autumn s: .5. R. Z. 3... $325. :2. 83:95:. 2 t. , nae. ego. Nae. Noe. sac. sac” em.;s Asm.=~v Ana.-~. Ans.sshv Aa~.mac games. Ac~_~_=cc.v =. muu.a:a == ===; >uu0=CHL mesa: nfi< ow::;o «rack acmsu ac>ua wo:su«u_=waw .m—musucwusa =— ucuumxeu==:nuw .uswdu_uhucu scummwumwz nu;E:z :c_.r;_moz we mo—au—ne> usovcoawfi:~ \muu::m fizz: ‘ NM N: u=nun=cu , \m—a=_u:> u:c=:oae: mead .<=m .==_s asst s; 1t»s-=o.£: .cum=.u>u fi~uuv>c e=a .ch .:aq: Acuuso fisuua: a owcuu>s .uuqou u =« w=uuqsnmu ¢::a nu» xenoncbm uu_auu aw acne cuqnwu awn awmuv> u:u u==ou eyes: u:=oo :u—g Esau uuw e=s .uo=c.= w-suu no; vmuuw>s a lilllllb|t .MN. nag. . as“. use. Nam. can” _::a A.e-~v Ame.eseuc Aumnc.-. .Nsno.-v Arse~.v Aads~eoc.v =:_mm¢.uo= <=a mm. .nnnn: eu.s_cu ~n~s.mn emm~.na uned.| agnocoo u-. n~e. nos. sac. as“. ~ann our;r A.sn-v A.oao~v Aa~.u~v As_.a~v Anasa.v Asnsaoco.v =c.=:u.mtz == ==es manages; =ouuuq=zaa u-n=; :udz adusah saw: :_ uw=m:u muse: ~—< Luss=z ::_mmou:uz a a. do>u; wusmuuu_:wqm .c—mo:u=wuma =— mean: vhn$=mam .uamuoduhuou :c—mm0umw= \webacm can: u Pump-59 DUHdehB> HGO—uEDLU—ua; \G—_~=—.:..> “cw—y’ZHLL: cco— .<=h =z< ==hzu=H com mzcummwzcmz zc-v<—umzam< Nu Nq= . 333 .333 1 .sam .Hmm .mss .mme .mmn .aem .Hsm .aem .amm .ome Asa ones son oneness W .snows. .aammm .aerme .ememoa .asese .amomw .nasmn .Hmans .mmwme .moooaa Aav sane sea omssoes .samma .monoma .ammes .eiaama .msssa .nsamsH .mmaom .msnama .mmama .smmmna Aaoooav : nmeaeaasm s puma mo osHs> .aem .mmm .msm .wss .mnw .33: .sam .sse .ase .Hom sosssz : comm soeo . _ modem spa: ments or .dwa .owfi .mma .me .mna .Hma .wna .sna .mma .NNH momo< I seem my soNHm omnHo>< .emommm .mmeomm .namnwa .namasa .mnaewm .mnnaem .mmmmmm .Hmamnm .anmomm .smmmmm noses I msssa as snag .sama .Hama .nmaa .NNHH .mea .omsa .mwsa .mmwa .mmma .Nmma sopssz : mshma HH< seam emmm mama rams moms same 6063 news (lates mmoa sensor MMdSoU finance >pssoo >psdoo sources counsnanx sopswao sopem sanmsH 16863 835.3 62323 isms .meaa memes zsonon .zomeess mas oomszsqsa .qumzsq are zH mmHezsoo moa_ nmmaqdadm ma<8m 92¢ ZOHBUHQmHMDH ho am»? am mmxoq me aflzmczoe mama .a>oq xme mafia mmoa “monscm .363.66¢.3 66.:n .665.666.3 6n.6m .6-.66N 6N.63 .333.6n3 63.n .san.m3m.nm ss36 scoezouaa .6aw.ma6.am 6a.66 .ma~.mcm.a3 a6.m: .sas.~a6.a 65.63 .swm.6m3.m 63.n .6n6.n6a.63: ss36 ooamse3a: .mam.ans.3 mm.m: .666.3m~.3 ma.am .mn6.nn 63.3 .6sm.m63 63.n .636.:m6.mm maxme gangs; 3.? sores; seam and? 3.3 .8263 63A sedans 2.22.: {p.863 3.? .mikmm 3.66 Amman 66.3 656.633 636 .6N6.mmn.mm 66333>£ .mma.s3s.m 36.6: .nmm.n3m.w 3n.3s .N6:.6:3 66.N .3nm.mmm 63.n .66:.m:m.6a oaasnmo .amm.aaa.6 m3.6m .ms3.a63.m a:.~a .363.6N6.3 6m.6 .6~N.m36 63.n .Nm6.6am.6w3 oosasa3ax 63.3.6.3 and: 63:59.3 mien $9.66 66.3 666.38 63% .66m.wn:.am H6366 .666.336.m 66.6: .6:6.3a6.: 6~.6: .666.36m 6m.m .3n6.w6n 63.n .666.66¢.m33 goesmsoo .3266: 63.6: .865? 66.6: .3m6.63 66.3 633% 636 $848.63 x2336 Swazi: so“ an .m> as. an as. an an. an .n. ma. Teena... 33m H.309 Hoonom 5.50 no Awnmczom. #3500 ANOH . QHNHiDGm whim Q5. ZOHBOHGmHE—fi mo Ea. wm mafia. Qz< mug—km v28 Emaomm mama. 176 .hpsmdoue mo .amon .nmwfisoflz mo oamnm .soammdssoo xme madam .z>oq xme macmszoa meH .s>oq xa? >a3o 5593 _oohscm iiiii .mao.m66.m3 36.36 .666.m3n.a 3N.n: .mn6.6~6.~ :6.a .emn.~6s.3 n6.6 .66N.n3s.63m seasons 58.3636 9.3m .mwadmwfi No.9 366.33 8.3 20.63.56 3.6 $368.33 fissm 636.5% .3 who: Amman; 5.6: .38.3m 66.3 .Smdmm 3.6 $3.86. 3m no.3: 33am .ma6.6~a.3 an.aa .6n6.6::.3 36.66 .aom.6n 66.3 .Nm6.NmN n6.6 .N:6.~mm.6m 66w3oaaz .mma.336.n 66.6w . 336.3638 3.3: .68 .83 on. N .666.Nmn 36 .6N366N5. 3:63 stigma and: 63.6%.: No.3: 68.663 66. 3 566.63 3.6 69.68663 62x38 26666: as. an as an an. an an. an s Haves Hoozom huno no mazmczos handoo R63 is: 253865.? 53:3: .2623. as moa a3; 633.263 Egan Qz< ZOHFUHanman ho Hub. Hm 952m. 92 mum—5.3m 55H. arm—momm H493. 177 assessed value or average per capita would seem more appropriate. Regardless of the potential for measuring holding costs, it is more likely that variation reflects preferences for different levels of public services. Indeed, the second major point made by developers was that taxes, per se, did not seem to be important in residential decisions. Other neighborhood, amenity or school characteristics were more important. If public services were great in these areas, the property tax would be higher and holding costs also greater. But demand would be greater too. As in the case with sewer provision, no clear conclusions can be drawn because of the necessity to look at issues such as the tax base, assessed value, per capita taxes paid, and tax rate. A complete picture might not be feasible given these data needs but there is little I evidence to suggest that the property tax range or property tax proportion of general revenue reveal sufficient information on the effect of the property tax. Undeveloped land is often underassessed and programs designed to relieve agriculture from property taxes encourage land holdings in some cases, for urban sale. This is also encouraged by the significant appreciation in land values between agricultural and urban uses. Conclusion The econometric chapters presented a model explaining site price and appreciation with demand, site characteristic, and instrumental variables. Though not statistically significant in the research, site price was interpreted as a proxy for zoning by Witte (1975). Percent all and new homes with public sewer were generally quite robust while 178 the property tax range held promise. This chapter has compared factors related to the dependent variables such as site size, development costs, and agricultural opportunity costs for three Specific locations in Michigan. Site price and appreciation were compared across the three metropolitan areas in addition to an examination of zoning, sewer, and tax issues. Closer scrutiny of national land value data via interviews with planners, developers, and assessors indicated that the results of the national model should be analyzed guardedly. Total population, percent population change, and income indicated that Lansing and Kalamazoo should have similar appreciation levels while Jackson should be lower. However, appreciation in Lansing and Jackson were similar and higher than appre— ciation in Kalamazoo. The reliability of the Jackson data is dubious but developer comments supported the difference between Lansing and Kalamazoo. Agricultural opportunity cost, site size, and development costs do not appear to be significantly different betweeen the three areas. There was some variation in the structure of the home building industry. Jackson had large developers from Detroit because the local financing and home building market was not large enough to support larger developers. In examining the instrumental variables, it is concluded that the degree of zoning restrictions did vary between the three metro- politan areas. In Lansing two of the growing townships either limited the location of development or zoned primarily for large minimum lot sizes. Kalamazoo, on the other hand, appeared to have a sufficient amount of land available for a variety of uses. Zoning around Jackson 179 was amenable to just about any kind of growth. These comparisons account for both the ease of zoning changes as Well as static comparisons of nominal zoning. Sewer provision and property taxation were peripherally examined. Taxes were not considered to be important locational factors by developers, though further reflection and examination might have indi- cated more of a role with respect to the ability to withhold land from the market in anticipation of future gains. Sewer provision, and to a lesser extent sewer financing, were recognized as important market delimiting factors. For example, where sewer provision was limited and septic tank policies were strict, leapfrogging was occurring. It is interesting to note that neither SMSA or urbanized definitions would have picked up this particular leapfrog pattern. Furthermore, the same metro- politan area, as defined by the census Urbanized Area, included an incorporated jurisdiction which was still growing. The combination of zoning, sewer, and tax policies and character— istics indicated that there were more problems in getting land for anything other than large minimum lot residences in Lansing than in either Kalamazoo or Jackson. It can tentatively be concluded that these policies account for some of the variation in appreciation between Lansing and Kalamazoo. With respect to the econometric results, Kalamazoo had a relatively low percentage of homes sewered but was actively providing those services in areas where demand existed. Therefore, to a certain extent, the percent all homes with public sewer did reflect the expected supply situation in Kalamazoo. On the other hand, sewer 180 provision was not that difficult in the Lansing area, though controlled more than in Kalamazoo. The tax rates were lower in Kalamazoo and Jackson than in Lansing. This could also be another growth related factor which helps to Shape the urban environment. Descriptive data at least indicate quite a bit of variation within urban areas. Though the variation is generally from central city to the fringe, there is still significant variation around the fringe. Site size as a proxy for zoning can be brought into question by noting that where zoning or other supply restrictions do not exist. Jackson, for example, consumers can presumably buy bigger lots or, at least, fulfill preferences for bigger lots. Large site sizes could then be the case for both restrictive jurisdictions or areas or unrestrictive areas. The comparative case study supports the general model of land conversion while raising questions about the interpretation and policy implications of the econometric results. The comparative case study also is suggestive of ways to operationalize the instrumental variables or at least directions future research could take. For example, the extent of unused and planned sewer capacity would be better indicators of provision policies than either past or present homes sewered. Finally, developer behavioral comments supported an important theoretical concept develOped in this research. That concept stated that supply restrictions arising from local government policies will increase appreciation in that jurisdiction, while the excess demand will shift to other jurisdictions, raising appreciation there too. The degree of this effect will depend upon the variation in policies 181 among jurisdictions. Nearly all developers who were interviewed indicated that they were not building in specific jurisdictions because of the attitude of that jurisdiction towards different kinds of development. This process could increase aggregate land value appre- ciation but could keep the relative position of jurisdictions the same. This increased appreciation creates further pressures on the land use plans of all jurisdictions involved. Costly court suits to get zoning charges and this appreciation are common. It was interesting to note that the concept of a land market was easily understood by developers, perhaps because of the aforementioned process, while planners had difficulty with the concept and seemed inured to thinking in terms of political jurisdictions with little recognition of the interaction of one jurisdiction's policies with other jurisdictions. ___—‘-~ CHAPTER IX CONCLUSION Introduction This chapter has three objectives. The first objective is to briefly summarize the contents and results of this research. Many specific interpretations of both econometric and comparative case study results have already been stated. This summarization will note the major themes and results. The second objective is to determine the policy implications of these results and answer the question of what do individuals or groups support in terms of government policies if they want to keep prices and appreciation down. The final objective is to suggest areas where further research should be directed. Summggy .Chapter II noted that there continues to be a large and growing residual land value which contributes to high residential site prices which is not explained completely by agricultural opportunity costs, site size, improvement costs, or general price inflation. This residual is defined as land value appreciation and is demonstrated in Tables 9, IO, and ll. The underlying economic concept in this research is the distinction between economic rent and profit. Appreciation consists of both. Rent arises from natural limitations in supply which cannot be competed away. Profit, on the other hand, arises from factors which limit supply due to market structure or institutional restrictions. It is these institutions, exemplified by zoning, sewer 182 183 provision and pricing, and property taxation, which are analyzed in this research in terms of their supply restricting effects. Specifically, hypotheses were developed for zoning, sewer provi- sion and pricing, and property taxation with respect to appreciation. Zoning policies which restrict the availability of land for certain types of residential development, such as high density, single family housing, raise prices and appreciation. The degree to which this effect occurs depends upon the aggregate and jurisdictional variation of zoning policies in any given land conversion market. If a large share of the market has such zoning restrictions, the greater the appreciation. Much the same process could also occur with sewer provision and pricing. The hypothesis is that sewer provision policies which restrict supply relative to demand increases appreciation. Pricing policies such as high user charges can affect the location and movement of demand in a land conversion market. It was also noted that sewer over-supply could lead to a diSpersed land use pattern with less expensive Sites at the end of the sewer lines being developed first. Finally, the property tax was discussed with a major focus on it's effect on holding costs of land. If taxes Were low for undevel- oped land then it was easier for the landowner to keep land off the market in anticipation of higher appreciation. While zoning, sewer, and property tax policies are not the sole shapers of the urban land market, they present a range and sample of the local government institutions most likely to affect the market. In general, the model presented tries to advance understanding of the land conversion market and explain variations in appreciation and site prices across metro- politan areas. 184 Chapter III presented previous research which has concentrated on either intra-urban land values or the demand characteristics involved with inter-urban land value comparisons. The economic theory of rent argues that rent is demand determined. The results of previous research on demand variables were used as a basis for specifying those variables in the model. Because of the use and availability of site prices across metropolitan areas, this variable was also analyzed. This was done to examine the stability of some of the demand character- istics and, more importantly, as a contrast with the measure of appre- ciation developed by Schmid (1968) and applied in this research. Site prices available from both the National Association of Home Builders and the Federal Housing Administration, which capture, in part, different sectors of the land market, were used as the basis for calculating appreciation. This process examined the component data and variables needed to calculate appreciation; site size, development costs, and agricultural opportunity costs. Site price and appreciation were the dependent variables. Appreciation was designed to measure both economic rent and profit. Because of the data difficulties with the calculation of appreciation, site prices might be considered a crude proxy for rent and profit. Unfortunately, components of site prices, such as development costs, raise other policy issues and compound difficulties in interpretation. Nevertheless, site prices help to maintain a check and comparison with appreciation. Chapter IV developed these dependent variables. Chapter V operationalized the independent variables. The inde- pendent variables used to explain site prices consisted of demand, variables, site characteristic variables and instrumental policy 185 variables. Since the site characteristic variables were used to calculate appreciation, only demand and instrumental variables were used in the appreciation model. The demand variables included total population, percent change in population, mean family income, and a regional dummy. The site characteristic variables were development costs, site size, and agricultural opportunity costs. The instrumental variables were the property tax proportion of general revenue, property tax range, sewer range, and percent all or new homes sewered. The property tax range and sewer range were proxy calculations of the variation in policies which used the high, low, and average per capita property tax paid or sewer capital outlay. The method used to test the theoretical model consisted of a series of cross-sectional regression equations, pooled cross-sectional and time series equations, and a comparative case study of Lansing, Kalamazoo, and Jackson, Michigan. The regression analysis explored demand variables and the hypothesized relationship between sewer provision, pricing, and property taxation and the dependent variables. The comparative case study examined the geographical basis and depen- dent variable definitions of the regression work with a finer grain of analysis. Zoning was explored in some detail while sewer and tax policies received cursory treatment. Chapters V, VI, VII, and VIII presented these empirical results and analyses. The results of both the econometric analysis and the comparative case study are now considered with respect to the zoning, sewer provi— sion and pricing, and property tax hypotheses. 186 The zoning hypothesis stated that the greater the percentage of low density residentially zoned land in the land conversion market, the greater will be the land value appreciation and site prices in that market. Zoning was examined in detail in Chapter VIII, the comparative case study. The results of examining nominal zoning as well as comments from developers and planners suggested that other factors needed consideration in addition to nominal zoning. The effect of zoning on actual opportunities to build did affect the location decisions of developers. Developers aISO noted the pressure on land use regulations. This supported the concept of the competition, through the political process, for economic rent and profit. Most important, however, was that differences in land values and appreciation between the three cities could be related to a holistic, qualitative, and quantitative measure of zoning variations. If it were possible to aggregate various characteristics of zoning such as nominal zoning, nominal zoning changes, time and red tape involved in zoning changes, and the uncertainty of the process, then this holistic variable might prove supportive of the hypothesis. But its interpretation Would be difficult. Based on interviews with developers and planners substantiating behavior consistent with perceptions and reactions to restrictive zoning, as well as examina- tion of nominal zoning, the weight of the evidence, as presented in the comparative case study (Chapter VIII), suggests that zoning restrictions for certain uses, in the aggregate, do affect developer decisions, inter alia, and land value appreciation. Especially significant were developer and planner descriptions of jurisdictional attitudes on zoning and responses by developers in the entire land 187 conversion market. Demand shifts as a result of developer decisions from one jurisdiction to another can lead to higher prices throughout the market. If there are product differentiating effects with respect to large lots, the price increases could be even larger. Similar behavior patterns can be seen with sewer provision and pricing. It was hypothesized with sewer provision that the greater the percentage of land in the land conversion market where sewer provision is controlled or restricted, the greater will be site prices and apprecia- tion. There were several caveats on restrictions associated with too much or too little sewer provision. Restricted supply of sewers should increase appreciation while over-supply would increase price, because of the sewer component of develOpment costs, but lower appre- ciation. The sewer pricing hypothesis stated that the greater the percentage of subsidization of sewer services in a land conversion market, the greater will be site price and appreciation. The weight of the research evidence suggests that local govern- ment policies regarding both sewer provision and pricing affect the supply of land for certain uses in the land conversion market. In the econometric analysis the public sewer supply hypothesis was operationalized by two variables; percent of all homes with public sewer and and percent of new homes with public sewers. Percent of all homes with public sewer could indicate, in the aggregate, past sewer provision policies while percent of new homes with public sewer could indicate present policies. These interpretations have to be guarded because of the importance of other policies such as septic tank use. Aggregate under-supply can diSperse development if septic tanks are 188 allowed. Appreciation is, therefore, low. If septic tank policy is tight, appreciation will be high. If sewers are over-supplied, then development could also be dispersed. While the econometric results were mixed, both the percent of all and new homes with public sewer were frequently statistically significant, at greater than the .01 level, at explaining site prices or appreciation, especially with the NAHB data. Table 13, for NAHB Site price data, presents two equations with percent all homes with public sewer statistically significant with a positive sign at the .007 and .006 level, respectively. Table 15 also demonstrates positive results for the NAHB site price regressions while FHA site price regressions were weak, as demonstrated in Table 14 and Table 15. The statistical significance level for all homes with public sewer in the cross- sectional appreciation regressions ranged from a low of .020 to a high of .001 across eight equations presented in Table 16 and Table 17. In the pooled cross-sectional time series regressions the percent new homes sewered was statistically Significant in the Test of the I Covariance Model, Table 19, and the Appreciation Covariance Model, Table 25. However, results in the site price pooled regressions in Table 20, were statistically insignificant. The results cnfpooled regressions were encouraging and seem to indicate stability of the sewer variable over time. In addition, comments made by pdanners and developers in the comparative case study did support the behavioral implications, in terms of developer location decisions, of sewer supply policies. Nevertheless, the difficulty in separating supply policies from sewer use policies makes interpretation of support for the hypothesis cautious. 189 Sewer pricing policies were tested by the sewer range in both the cross-sectional analysis and the pooled regression analysis. With the exception of Table 16, in which the sewer range was significant at the .08 and .07 level in explaining appreciation (NAHB), the sewer range was consistently insignificant. Furthermore, Table 18, which tested the sewer charge effect on raw land prices, also yielded poor results. On the other hand,the comparative case study indicated a growing awareness of sewer user charges vis-a-vis location decisions by developers and planners. Combined with sewer provision issues, sewer financing can limit residential growth in some areas and encourage residential growth elsewhere. Property taxation was tested in the econometric analysis most successfully by the property tax range. This was especially the case with NAHB data as was demonstrated in Tables 13 and 15 (site price) and Table 16 and Table I7 (appreciation). Since the pooled regressions used FHA data, generally insignificant in the FHA cross-sectional results, it is not surprising that the property tax range continued to be insignificant. DevelOper comments in the comparative case study did not indicate much importance for property tax variations. On the other hand, agricultural opportunity costs in part represent prOperty tax effects on landowner ability to hold land. If the range is large, agricultural land might be taxed closer to use rather than urban market value, thus increasing the agricultural opportunity cost and the ability to hold land. Agricultural opportunity cost was, with the exception of Table 21, consistently statistically significant and positive. A comment by one developer substantiated concern over agricultural land taxation and its effects on land availability. Nevertheless, as ........... 190 mentioned in Chapter V, this use value taxation, either de jure or de facto, is likely to delay and change the results of land conversion and raise prices and appreciation rather than achieve agricultural land preservation in the long run. When the results of zoning, sewer provision and pricing, and property taxation are aggregated, the general model of the land conversion process presented in Chapter II seems generally supported. While the interaction between these and other instrumental variables is complex, the conclusion is that they separately and jointly affect the supply of land and can be a source of appreciation and economic profit associated with land values. Policy and Research Implications What do individuals or groups support in terms of government policies if they want to keep prices and appreciation down? Efforts to lower appreciation will require focusing on all of the governmental jurisdictions in the land conversion market. Figure l, The Conceptual Model. isolated two issues. This research examined the instrumental policies of local governments and their impact on site prices and appreciation. The structure of local government and its impact on those policies and, hence, site prices and appreciation can be discussed only when the role of the instrumental variables is clear. Tentative policy implications of zoning, sewer provision, and taxation are presented here, however, with some reference to government structure issues in order to present a range of institutional alterna« tives and to avoid leaping to conclusions often found in this research area 0 m __> .,___ . 191 As has been noted, zoning policies can increase appreciation which in turn can create pressure on land use policies such as zoning. Variations in the degree to which supply restricting zoning policies exist will affect both developer and consumer location decisions and, perhaps, Speculative activity. The greater the aggregate restrictions relative to demand, the greater the appreciation. For those who would like to lower land values created by this interdependent zoning process, the increased values are a negative, pecuniary externality. The commonly used Tiebout (1956) model assumes no intercommunity spillovers, since people can move to communities offering characteristics they want, without consequences for others. Zoning is one tool for this product differentiation. If it can be assumed that information about the appreciation effects of restrictive zoning will not change behavior, given the interests vested in the current structure and competition for economic rent and profit, then some change in governmental structure is suggested. Jurisdiction by jurisdiction changes to make zoning less restrictive over the entire land conversion market are unlikely to occur because of the benefits gained by those jurisdictions not changing. Various other land use policy options, whether superceding or coinciding with zoning, must account for jurisdictional inter- dependence. Chinitz and Cowing (1977) have analyzed the argument that metropolitan government be created to internalize externalities and recognized the geographical difficulties and value conflicts inherent in such policy prescriptions. Institutional changes designed to lower land values and appreciation will affect other preferences; small, homogeneous, or high income suburbs, for example. 192 Other institutional arrangements exist. For example, transferable development rights, which are designed, in general, to eliminate competition and change the distribution of economic rent and profit by having the winners in the land conversion process compensate the losers by a bargained transaction rather than administrated (zoning) decision, could be exchanged by local government jurisdictions or individuals acroSs the land conversion market. Clawson (1960) suggested several large suburban development districts in a metropolitan area which could have broad planning and infrastructure powers. In California, a similar concept, spheres of influence, is used as a basis for a planning device and organization (Local Agency Formation Commissions). This structure has problems with interacting with existant governmental jurisdictions which limit their effect (Eells, 1977). The merits of any institutional change should be evaluated on many criteria, includ- ing demand articulation, production economies, as well as prices. Many of the same issues arise with sewer provision. TaborS. et al, (1976) argued that, "The stronger and more centralized the control of the institutions responsible for sewerage planning, the more effective the overall policy is likely to be." (p.172). Sewer supply relative to demand and timing also need to be considered within the context of the land conversion market but where the authority is placed and who gets included in the decision process will determine the ultimate impact of any change. The impact of a decision by one juris- diction to limit sewer supply has, perhaps, clearer implications than zoning on other, nearby jurisdictions. With added concern about the appreciation and land value effects, land conversion market 193 jurisdictions could bargain within the context of metropolitan planning agencies which are becoming more involved in public service supply issues as federal funding increases. Agricultural taxation issues might also vary within a land conversion market. The empirical results of this research for the property tax range and agricultural Opportunity cost support conclu- sions by Schmid (1968), Schwartz and Hansen (1973) and McMillan (1973) that lower property taxes lead to capitalization of the lower taxes into the value of the farm. Policies which, therefore, tax agricul- tural land at use rather than market value are likely to lower uncer- tainty and allow for an increased short run ability to With hold land from the market and hence, raise appreciation. Before research into alternative forms of governing different aspects of the land conversion process begins, further research into the instrumental variables analyzed in this research, as well as other policy variables, Should occur. Other authors contribute different perspectives to the analysis of policy tools (Greene, Neenan, and Scott, 1974; Mills and Oates, 1975; Portney, 1976; and Downing, 1977). But the complexity of the land conversion process perforce argues for a complex model. It is clear that the aggregate effects of local govern- ment policies on land conversion are little understood and have been subject to little empirical investigation. The research which has occurred, including the results reported here, have pushed the analysis of generally available data to its limits. Better measures of the relationships of interest, obtained through primary data, Should replace proxies which do not measure the 191+ relationship of interest and which keep researchers from having further confidence in their results. The contrasts raised between the econometric results and the comparative case study should aid in designing research and survey instruments. Further analysis and research into land value appreciation and the impact of local government policies, given the increasing size of appreciation, seems advised. APENDICES APPENDIX A SEWER AND WATER TAP FEES FOR SELECTED JURISDICTIONS IN METROPOLITAN DETROIT, MICHIGAN, 1969-1974 APPENDIX A SEWER AND WATER TAP FEES FOR SELECTED JURISDICTIONS IN METROPOLITAN DETROIT, MICHIGAN, 1969-1974 Robert H. Carey, President of Thompson-Brown Company of Detroit, Michigan reported the data presented in Tables Arl - Ap4 in a series of articles which discussed development costs and local government fees in jurisdictions where Thompson-Brown had developments. Other data on development costs are presented in Appendix B. Mr. Carey noted that "another cost item that should be considered attributable to land, is that of 'tap' fees, 'use' charges, 'capital improvement' charges, or whatever they are called. The variance from one community to another, or even from one area to another in the same community, is unbelieveable." Mr. Carey presented the tables reproduced here without further comment. 1970 and 1971 reports duplicated the data presented in 1969 and are, therefore, not reproduced here. It is possible that the fees did not change for those years, but further analysis is speculative. The data do provide information supporting significant variation in sewer and water financing policies. 195 t‘fl 196 .6em3sc3z .h_cchc6 Sana-.00 sagas—83:63? 3:03.695 £0.50 .= .7336: 3 coup—ohm «is «.3150 .629: 6m 35 .6626: 666 can con ac<§>< :3.6m6 on: cam 66m 33.6nn 6:62 :3.663 6mm ee< can; cause: cum; on: CNN 8m 80 Gem own 0mm .52 of: nzzncz al.6236e 6rm3azecem .3 6n.n63.3 6n.mn: 6m.~n~ 66m 3w6 lezcexccua 6:. 66 _>6z man men mew 663 66: .-czccxmcue oz- .__=6239e ace3>=ezsz was mmm non Ohm ow: o9” oom 1.333: .695 mmm. mmm mm: on.” com chouuxamufi ozu mud—ratio re_6 asa3>=r=6= 666 6mm :ew nu own on 6n3 6mm su< desserts 233.3 6mm sew mm was on man 6mm eu< c3n: n3 . . a_=mzzoe zcsuznz=< 33... E .955 E allie 2:3... E .936 2.3 one... 36 sic A653 :3 H338 :ouuooamsm acme—36mg .8302an has he ewes: no .axe au< no game: A3 .52. 66:36:66 =se< eu< 9.5 63.3.36: 5.2 9.6.. 35:36: L 25233.; gar—gsm Fax MESS; ~>Cz ._ 2.ng "ma—:::...zCz gnu—hamwa .009m 6866::6 5.30 MAJ :::—Oz EL< «mi—mug. 3F3< 0:: AH ._ ___mzli. zO.—.Uz 3:: <6. 5:53.33... 23.—.230 L Emzscp. zC>< .z<.r:=:n= ..F—OFEQ z<.—.~..9_CF—.mz zn $203._.”::mdd:_. QELEAWm sch .NNOH .Emm .22. ::< mm: £5.53 nz< zmzam N16 3365. 6.33.33.36.33. 353633: 336.336.. :3 .363. :::: .o... .36.. a» .\I . 63.36.65... 3.63:; 3.6.3 66.. m .32. 3 3:: .6... .36.. n. as: 3. 00.5 3530.63.56 0.3 .3603 3.6.3.32 6.6.: .6... .393 w. . 3 3 3 6 .\| 5.666 3.66.. 8.6mm 6.3.5.2.... . , . u < 9.5 3.3.66: 666 656. 6 63 666 66: 666 63 .66: 666 a. 6 . 63 666 66. 636 63 66: 666 eu< 6:6: .6636: .66 3 6:6 6 . 63.6236... 6.536335... 6%.... 3663......3 623:6... «3.3,: . . 6 63: 6.66 6 6...... 666 \M 663 6 \3. 663 ..6 .636 6.2.. 66.2.... 66 .666 6:66 2.... ..:\6 66......6 6. 66.6 666 6:6 66.6 6 2.5. 66.6 63:623.... .3366256 666;...3 66.66: 6:62 66.6 666 666 666 6 66:2 N66 6.5. 6.x: 66336: 66.6:6.3 66.66: 6:6: 66.6 666 666 666 6 6:6: 666 6.: 66:6: 666:: 66 66.6 .32.: 666 .6: 6 6:62 636 ...3 66: 6 6:6: 3x6. 6.=6z;66 6663>=y=6z 666 666 czcz 63 663 663 666 6:62 6:6: 666 .6. 6636 .66:.>:6=6z 66.- . 6.3:: .66... ....\m :66 6:6: 666: cam 6:6: 6:6 6:62 6:62 :66 52.63.. 666 6.. 6 6:6: :66 66 366 3: 666 666 6.: Lo>36.666.6 6% 636.3 6...”. 6 6.5.. :66 6... 366.3 3.. 666 666 E: 63:. 63 l 666 6:6 6 9.6.. :66 66 3:6 3.. 663 666 E: .626}... 333.3 6:6 6 6:6: :66 66 666 3: 666 666 6.< 63.: 63 6 _. :6... 26.3 3.666.. 3.... <6 5.6.6 663... 9.3.. ..:\6 5.636 6 :6.6 666 663 666 66. 66.6 6:. 2.36.. 66....23... 666.3 666 63 63 666 666 666 63 6:62 666 :.=m==6+ 26>: 6:3 :3 Afiwuwmu...5 Amaommfia gammg. am 0:34.. New on; Haywomumsu Nguuuflfi Ngmmfimfia avomumgo :::... .392: :3 356... .6663?!— 63 om... 3.6.3.60 omuczo 6M3. on: 33.... o... 663666.663 6:6 6:: Amwufiba ILH 9.236266 3.66.5. =3. :66 236.63.. 2.365633 3.3369556: 63 6.8.6.63663663. 66.366366 :6... .6663 .666... ..E. 3...... mm: 66.5... 62.. 66.666 6% m. 362,—. II]!flHHWWWJW|||||||||||||||||||||||||||||||||||||-- -11IIil1 A 666.3 6336 6333.3 :0... 6.3.3.3 3 199 .AhHHoauasc 6633333 663336u 63 3966 333:: 23:62 .6; me .\~ 66.666.3 666.~ 6:6 ON:.H 666 66.6m6 66.666.3 66.66~.3 666 666 ~66 666 636.3 666 333.3 3.6.636 663.3 3262.8... 66:36:66 CtNfi 033.1— ..:\m .3366 2.3.3 ..:\6 6N3» 2.3.. ..:\6 36:6: 6336 33 WSW Q 6.36 w 666 m om.me 03 66.666 ocoz on.mcm 6662 3mm 0 won 6:6: Ohm 6:6: 6:6 6:6 w ham w wzfl w :66 36 6 666 63 6:33 :3 66:6 66:36 63 mwvmwmndzv fig 350.3. :6 3.3 6033 m3; 63.-6 6N 66m 66: 63:.3 63 use: 66:.3 66. 6mm 66.: 666 63 2.2. 666 6m 666 66: 66 6 6 3 66: 66 6 6:02 con \fl 06% 03 com Own on . 6 666 666: 66w 6 6662 66.6 66 . 6 66m 6:6 636 n 2.2. 666 66.6 666 66~ 666 6 ago: 366 63 66: 663 66: 6 mac: 66: 63 663 663 666 ocoz was: 666 2.2. 66m 2.2. 66.6 mm 2.2. 666 2.2. :66 6 6 366 3: 666 66.6 2.2. :66 6 6 366 . 3 3: 666 666 2.2. :66 6 6 3:6 3: 66 3 66 6 2.2. :66 6 6 666 3.. 6 66 66 6 :6.6 66m 663 666 66 6:02 666 63 666 66: 666 63 ozoz 666 m 3663,3635 Ago—33.3 :3 N w wound—3U N m “womumso N 3063.36.30 A w 363.6339 A a 3663.395 on: . .3 6:60 6936.330 333.33. cm: 336.3. :6 3.366.363; on: 6633 :66<= 66366 66<66>< EL< 6:633:6u663z3—s EL< oxc3 c3huoz 56< 63<3 33:36: 6:52;... 6.67.33.66.36 6.66... 66 63:..63633::3<3 .. 2.6.6.66 5.323.... 5.2 3.6.. 3533...: EL< cwzcz :66:: ...6 .236 .3336... 6 ._...._.:.... 6.... 326.66.. ...6 3.3.36 6.3.3.255: $26.3... Eu< Lc>3= c266: .._.3< o—w: CH s3< 36:2:Ln6 23< 6.3: 63 6.... E 66...“... :62... .. : 63.36... :::. ...6 .3 333m: 3.5.3. zc>< .2635... €366,363. 33.33.339.656: .33 9.63663663663. 66666366 :66 .:6o3 .6666 6<6 62< 66: :66<3 62¢ 66:66 .T.< 666<6 APPENDIX B ESTIMATED SUBDIVISION DEVELOPMENT COS'IS FOR SELECTED SITE SIZES, DETROIT, MICHIGAN, 1963 AND 1965-1976 AITENDIX B ESTIMATED SUBDIVISION DEVELOPMENT COSTS FOR SELECTED SITE SIZES, DETROIT, MICHIGAN, 1963 AND 1965-1976 Robert H. Carey, President, Thompson-Brown Company was the primary individual responsible for collecting the data presented in Tables B-l through B-lé. While there are some difficulties with the data base because of changes in definitions and a few apparent inaccuracies, this time series is a quite unique data source. The data were calculated for each year and does not involve retrospective calculations. The articles for which the data was prepared appeared in a number of publications including the Michigan Buildor. The text which accompanied the table for each year had extensive comments on the source and kinds of changes in each category cost. There was, however, some duplication from year to year. The signif- icance of this data is in an aggregate form tracking the changes in costs faced by a single develOpment firm over time. An effort was made to keep the categories of cost consistent from year to year. 200 201 .:amagoa= .anousca .acmnzoo ::oualcomalosfi .uaoudmoum .houmu .= anonoz ha uauaaoun 6:6 cofiameoo .coumzumw was: .eouzcm mm.mo Hog: ::.mm coo: 95.nw mash w~.mw mace m~.me mafia no.c ca: .m.o am: uh.o mun ~c.w New mm.o ome 5:.n mom mm.m nzw m~.n mam om.m omn ~:.n mo: mm.o wfi: .m.e Hm: an.o mum mo.c New mm.o 3mm c:.= com .b.: mom mn.n mmn nw.n mom mn.m mm: =o.~ can oo.~ mow on.~ mam ~:.~ msm mm.~ mom mu.m nwm ne.~ mam mm.~ cow nu.~ mud n~.H onH nn.m mmm .A.m mmw on.w nan ow.s own mm.m mooH as.“ men oo.w on: H~.w own :H.e :Hw 5H.m Nsn oo.HH owe an.m ans oo.m mam on.m mum no.5 mam m:.m~ 0mm H:.mH onHH No.5“ aged mm.aH mama Hm.H~ Hmnm ..an Emu gamma Em“ 25mm may. an?“ 3......“ ...Eam 3mm 0:.m m:.~ no.m mm.~ oo~.m ooo.a oom.- ooo.n~ ooo.o~ .oNH x .oo .o- x .nm .onu x .om .om~ x .ooH .mwfi x .cmH \ Mu_o amou mcawm hoccz ac amoc a:ow:aacoc a wsoccdmyocmmz cemuomaacu a wzaccaua .qoca Ezwhemcnwcm oozmnmm tad. owes—aha a made: soumhm Hoes: scamam hazcm TEC— ouo< so; méc; Aaoou .rwv wou< do; AscocV mama so; .=ma zohmfisHomsm ame<=aemm Him mam<9 \ AA! ZZCMZ .=awqgoa= .saouscc .hzmaaoo ::oumlcomasozb .acovumouh .houmu .= cacao: an woudnoga was coyaasoc .uouo;gmw can: .ochzcm no.55 mmw: 05.5 Nw: nw.n Haw oN..N. me: o:.: .NWN 30..“ 0: Sum mmm and awn N...“ 9% oo.H~ owe HN.~N nnmfi ~w.»oou Advaoa acouh upu he; .ONA x .ow 2.2. 8% mm . N. 9% mo.n saw mm . N. 9% :o. : mom me.m mom no.~ mam $5 30 co.o on: :N.. 0 RN. m~.- . coma mwvsoou Aware; acouu mom ppm o:.m ooo.o .ONH x .nm 3.3. 33 no.2. nomN. N.~.N. :3 RN. ans mo.m nmm oe.n can N.~. N. :3 an. N. onN. mm.m mwn me.n mom mn.~ mam ~:.~ New mm. N 08 us. A nNH on.o nan oo.n own :6 own 5.0 :3 oo.o mam mm.m one :w. :m 2mm NM . ow thm :me 3mm 23% am.“ 3. N Nb. N oom.- ooo.m~ .omH x .oa .oma x .ooa ~o.mm mesa om. N. 93 mm. n is so. N. oi Nb. n mm: cm.m new n~.~ cum on.c noofi mm.o ~35 no.5 mno NM .3 ER 123 Q34 acouh you won nn.~ ooo.om .50H x .ONH m_S:Pr adaosg uncuc cud cacsuc>o aces mayam hone: ho arcs acomzaaccc d wsoocsHHecmaz :ofioooamcg a wad::Mr— .nocu mangoccdmzm mocuamm QSQ_ o»d:~dun a mate: Emum»m scum: smamam Moron ecu. ouo< no; mac: Aswou .cuv moL< ac; Asooav exam so; .zHQ onmH>Hnm=m an9<=~emfi Nun H4nma chmH>qc¢=m QHEc 0N: :coo woman one ooze: a: one: n9. scam—.3 CCU é mzco:a~q¢onmz can :cmaocgmzn a wcacccTN .Hocd wcapoocmxzm omH mocdumm 1:m. @ONH omacfiwu: a arcs: Nam anzzm Head: woo scanam chcm mem 5:G_ 2&3 H9: ¢Lc< so; are; Aaoca .cov ch< :c; 38: 93.... ..c: .zmo onmasHooom omema chmH>~Qm=m ombo camonc:_ ::ccsoa wu_<;»:r adgcp; mmCL: c:m tco:p¢>: one: copcm nczcz ac :ccv :::mzdgcoc a msoo:mn~ccan: :omaooLCCN £ mzfi:::fi_ .noza m:mpom::u:m Gaza—d: :::..— cadzmcsc é sci:— sv::nm peas: Eogcmm nozcm :::— ouc< kc; clc; Auccu .rcv ocu< ac: Ascoov calm as: _._. ..4.- 7 _ . .:omazcaz .aaca:¢: .ncoasoo ::ounucomasOSB .acooawoum .nouoo .= guano: on cvuoaoua eza eofimosoo .coso;amm «do: .ocozcm 206 on.o n5.oo mono 5o.o Nnm n:.: 50N 5o.o Nnn 50.0 00: mo.: n:N :o.: 0oN do.nH ~o5 5o.5 mN: on.nH nNo n:.nfi 0No 4g 233 sooua um: um; Nn.: ooN.5 .oNH x .o0 uN.® omdonocd omauo>< on.o No.:o o0n0 o:.o 5n0 :N.: oNn o:.o 5n0 oN.0 N5: ~o.n ooN o5.n noN on.n~ o—oH RN. :3 n:.n~ 5oo~ ~:.m~ 0mg: now soon noooo: acoum Hos no; 0:.n ooo.o .oNH x .n5 mSH.w n@.m RN.© $0.5 00o: uo>o omooucz— ::coaca oo. No o0:5 No. no Nm05 n... :o m::o o5. no once: 55:... oN.o 0:5 on.o n05 .::.o ::o 5n.o moo: roach; moon: 1:: cccsuo>c :H.: n5n 0N.: non NN.: NN: oH.: non sooo mo_cm oN.o 0:5 on.o w05 ::.o loo. 5n.o moo” None: oo ocoo No.0 man 00.“ mmm an.“ . hum 0:.n m:0 acoucaazco é m:co=w_~¢ocm: 00.n own ~0.m 0mm ¢:.n c:m 5N.n Non :oaacoamcH a mcazzcdg .Hocm mcauoczamza oo.n CNN mm.N :om 0m.~ 0mm :0.H on mfiw mocmgcc 52¢— om.n~ RNA N0.MH “NNH MHAHH MHMH NN..N~ cmmfi fi: mmwz..d._c “e. EEC: o:.5 n50 No.5 o50 5n.5 5n5 N0.5 ::o ooN scream polo: 5H.NH noofi 5~.NH wood :5.- :5HH.. o5.o~ noN: moo season rmzoo oo.n~ oN:H No.5H nfioa Nn.oH Nnoa on.HN :onN o:¢_ 3:8... 3:3 3:8... 3:3 3:8,,— GFS 238... are: acouk you hon acouh Hob mom acoum My; Mom ::ouh Hon Hog om.~ 0:.N 5D.N Wm.H ooc< no; one; ooo.o~ oom.NH ooo.n~ ooo.oN Nova. .owV oop< so: .oNH x .oo .onn x .oo .onH x .oo: .50H x .oNH Nomoov coon so: .zoo zooma>oooom omac ccoooczq ::ocno; 0d: afiucu; m_c:c% maca: czd tzozgo>c owes acacm no:cz uo :mcc acowcaacoo N czcoca—noomaz comaowaccn i wzaczcna ooN .Huca fizfioeocdwzm mm mozwmdn c:fi_ ommcfidua 0 meco: ficfi Emnmzm seed: RC. 5.21;...“ .535 .=.:w_ ono< uQ. 50c; A0005 .rCV KCL< as; Aswan: csoo .oo .zoo ono~>Hoo=m oaeo a ¢moo mmamn .aocos so acov azomcfia:co w mzomcw—deoun: codaooaczm d mc—ccm—m .accm manhoozmwzm mocc~ma 6:Q_ owmzwmpc e mtno: Ec»mmm um;d: seamam ucxom t:Q_ ozo< Le; 0504 Aéwoa .ccv acu< a0; Aaccuv exam ac; .z< u§.~ mm.~ 5~.H~H .mmom ao.-~ .a5moH n:.~: .mmdn o5.~: .naon o5.5 .mmn n5.5 .moo .m.n .cfi: :m.n .nm: wo.m .5om Hm.n .nam 5.5. Q . :8. 3. .8 . 5cm: nw.5 .:5n co.m .om5 Sufi .86 2.3 .S: ~92 .92: .6.2 .82 2:8,. 3:3 238.. 333 acoum awn hum acouu Hon yon m:.m om.~ ooo.m oom.oa .owa x .n5 .omH x .oo mwum. no.:m~ .meq- H:.n: .momn ow.5 .:wc m:.n .mw: m:.n .nan on.om .man oa.c .om5 wo.n~ .n5H~ o:.- .wHoN 3.8.. 231.. enema awn ups m:.~ oom.- .mmH x .oo uw.~ o~.:- .odamg n:.m: .nan: 5n.5 .5m5 5~.n .5Nn ow.~ .owm mm.oa .mooH nn.m .nmm oo.nH .oonfi n~.:~ .m~:~ 43.3% acouu mom up» 5o.~ coo.nH .omn x .ooa m:.~ oo.:- .cmosfi n:.n: .Hamm Ho.5 .~:e Ho.n .Hce mo.~ .NNN mm.m~ .w~n~ 5n.m .mooH HN.NH .nmzfl mc.o~ .wmmn and? eschh Hma you nn.H ooo.o~ .50. x .oNH owmfi uw>o cmaouccw ;:ccu:& m. 2.... z. anuouh mmcuu d tro;um>o a amoo eofixm .zccos no aaoo azmwsaazoo a msomzmnqoomflz codaofiamcn a acacqua .~0:w wzunoo:dw:e coca—am 62K. amazemgc a 01cc: Emauam Leda: seaczm Foam: tzm_ PIE :9— maca— Aaocu .vwv d¢u< go; Aaccuv ¢N.m do; .zfla zOnmm>Ham=m nfihc; c_=:: am::_ .cmmH:o«z .aaouama .hcmasoo csoumTcocaeosa .acowamopa .houmv .= guano: an wouaaoua cca cmfiaaaco .cmhc;:mm :::: .ocuson 3.32 .535 @562 65:5 2.52 5.}: 8.52 ...uefi N12; .822 862 .592 ":55... 8:3 .mém 2...: .355 29:: 68: -.n: .80: 2.? 6R: no.9. .23 2.3.5 38: e _._..c;..8>o a ocoo mondm .aczcs mo econ 26 .oo: 2.: ...G 2.: .35 no... .25 8.5 .85 2.5 68 ..cowfi...._co e ..zSfiZSB: 8% .onm 36 .2: 3A .53 :56 .En 5n.m dun onfi .08 830822 a. 5.5.5:: .Hozm wcauoc:m&:3 Nn.: .53 cm... 2mm 2.: SR w5.n 6R Sé .Sm 8.5 .c:~ 3:55 .55 8.5.5 652 8&5 .532 9.3 .22 8.8 852 3.8 .38 :92 8:3 amaze... a 2:5: No.5 .8: 23 53 :55 .35 $5 .8w 2.5 :3 2.5 .5: 51m .52.... c3: .55 no.2 :82 9...: .82 :22 .82 5N.i .55.: o...2 .533 2355 Sim om . 2 . 5m: R . 2 . 9:: :c. 2 . 852 3. mm . 38 m H. :N . n Sm 8. 8 . wmmm E E 5.8.. 3:... 228.. 3:3 3:... 3.4. I312: 3:3 42.8.. 3.1.: Aqua.“ dam: acoum Ham- ugh acouk no: 9mg uncut you umh acoph up“ up: 9:0ua kph you acouu non no; Nn.: mi oo.~ §.~ 50.“... mm; PS< .8; 5c: . 8N5 38.5 o8.2 822 80.2 8o.o~ 2...; .5; 8.2 ..q. .oS x .oo .92 x .5 .82 x .oo .52 x .oo 62 x .02 .53 x .8.” 28.: can" ..3 H.z< x... N. 2n.mn2 .2mnm2 ww.w: .wow: n2.o .n2o mm.m ..Nmm ~2.n .N2m no.2m .no2w o2.o .w2m on.:2 .om:2 m2.:~ .n2:~ lama 222.42 ::ouh Hum Hon nb.~ ooo.m2 .on2 x .oo2 25.5 mm.~n2 .cmmn2 0:.w: .numn nu.» .ono nm.n .:ew -.~ .wom mm.2m .ommm NN.¢ .wo22 ~:.n2 .2202 on.ow .emmn 43222.2 $2222 2:92.. 20.2 .29. mn.2 ooo.om .mo2 x .om2 cmom uo>o ocdcpcc2 acoogog w 2.: :. afihoua mmcuc cam twospc>c cc: amoc mo2dm .sc:o: a: sec: anomauacoo Q msooccH20002= :oauowamzw x mzmccm—Q $22532o:2 .mcfluoccam:m mozamo: c:Q2 owwcawu: i mccc: soacam uc:d: sagwmm umzom cza— OLC< has mac; Ages. .cuv dwk< ac; 2.0m2v os2u .c: £82.62: 2205...: .52 ..E 2.8....» B .282. 82.. 2.2.22.3: E2: man—NH»... Ehm Rumba—mm mom .HBOH .880 “tar—5:369 zOHwH>Homam 93¢:ng nun H.292. 22112 .:ww«:od= .auouawa .acdaloo ::ounncomasofir .ncovdmwym .a¢umo .= auoaoz an woumnoun can cv_masoo .ccuwsamw mgr: .couzcm mm.m mm.n 3A9 .25 mmfina £22 mn.m: .amcn n:.n: .mmmm mo.w .nnm no.w .nmw on.o .ncn H:.o .Hc: :~.n .mom mo.n ..cm m~.mm .nHmH ~m.:~ .owwH :H.w .oo: mu.m .mwc nu.:~ .nmm 30.:H .macfi m~.mH .nman ~n.m~ .mazd ‘lawuwoou Navaoa Haggoou Ngammm. acoum up» ppm u=oum use hog Nn.: n:.n oom.n ooc.o .omfi x .ow .oma x .nn um.n -.mn~ .ooNNH mo.m: .Hom: cm.c .oco c:.o .onn. co... . :3 :~.:~ .NQHN mH.a .:~c no.2 . $3 :93 .82 dim a? econ: Hon kph om.~ occ.o~ .omq x .oo .aamco Humfl Iona ma.n omduuocu OMmuo>< an.n No.5na .HH:N~ n~.ma .:=n: nn.w .mwm $6 :3 w:.: .Ho: *m;nm .oaam nfiw :Am 3. i .39 o:.m~ .waom 2mg 2% m:.~ oom.- .mnH a .oo uc.n AQRH $2.2 mod: .83 on.o .o:o Kc.w .Nbo .mwm. .zmm mm.- .nmmm nn.m .nno Nn.:a .no:H n~.:~ .n~:~ Inflammmmm Nuqlmmm acouh ugh hem ooo.nH .onH x .ooH «n.0nH .ommoH mm.5: .nMNh no.m .owm wh.n .dmc mm.~ .mmm o~.- .mncm on.o .onH~ wa.nd .nmcfl mo.w~ .wwmm flav»00h haw go; acouh ppm no: ooo.o~ .nen x .ONH —nmn po>c emdogo:_ azooucm nu_<;;yh. aduouh mmogv t fuezuo>c a amcv mdem .hozoz ac occo occwcmacoc Q m:ceca-ccmfiz :o*»chm:n a mcmczmfia mcnczficcq .wzfiummzwwcs mozmfimm 5::— ommzwmpc é acre: seamzm Lead: eeérzm chom e=¢_ oho< uca mac; Aaccu .cCV ccg< ac; A;omuv cxwm s3; .zmc zOHmH>Ham=m CWP<2fismm 221:3 ~u®.o ~»5.H~ cc.onH .nooo nn.~n .NmHn on.o .mmm no.n .um: om.m .omn Snow . 82 no.m .ch mm.m~ .msa m~.n~ .ammH .udaqmmmm .aaqmmm acouh up; ups Nn.: oo~.n .oNH x .ow wa.o Hum.o~ om.an .mNNHH Elfin . 2% Nb.& .owm om.n .non an.n .n.: ~:.m~ .Hmma n~.o .moo mo.mm .nn~_ o~.m~ .onma .Iamqmmmm Adaflmm acouh up& up. n:.m ooc.a .ONH x .mn .=5w«:04= .adouaoa .hcaasov :zoumtcomnlosr .asovdmOhh .houdo .= amaze: a; doucaoua 5:0 cmuanaoo .cwuozama agwn «Cami ERN. d No.0 Ho ad m~.~n~ .n~on~ mm.Nn .mom: ~m.a .mmm om.m .wflm :m . n . S: on.n~ .manm on.o .mmm ofi.wq .o::~ 4®.nN .nsmw flwvaoou Nuqmmm acoua uma ugh om.~ ooe.o~ .ONH x .oo mun.o Hun Hfi Nn.nnH .nHmmH mm.nn .encz an.o .onm mn.m .Hom 8.... .R: 20.:N .W:NN Sé .55 om.m~ .mn:H 8.8 .Sfi “avaoou Ammuoa acouh hon nah Q:.~ oom.~H .RH x .C& N56 ~mm.- -.nn~ .NmnnH no.nm .nomn d~.a .Hmo .m;m . .amm so.m ..aa o~.:~ .o~:~ oH.oH .oHoH Hm.nq .Honfl om.m~ .mowm goon .dmmmmm acouk uma awn no.~ ooo.m~ .on~ x .oofi av omaouozm ommuo>< .mamoo c::a oubm\ccc.me c: cccz=m .omoo 6:64 ca mmawucca “UN aces—oz? msa ~um.- Nn.~nfi .wman mi? 51% id .32 o~.n .omm mm.~ .Aam m:.n~ .Hch cH.o~ .onma mo.:~ .momH w~.~a .Hmmn acoum Hum non mm.H ooo.om .mwfi x .o- n .oou:om Nnan po>c mcdw;c:_ ;:cous* C_fin zOHmH>~amzm nfl9<=HHmw 214 .cdmgzcaz .auouao: .hcaasoo ::ouancomnuoza .a:oudmoum .hoadc .= shone: an couanoua $26 comamsco .couo:aau sad: .nwu.t__..vmw mw.~2 ww.~2 mc.~2 R.a2 .22 $.02 $.22 2.02 .292 a~.mm .hmnn on.on .zm:: Ne.¢n .eomm 2.: .22 2.: .wa no.2 .82 .342 .20 3.2 ”Sn 2.2 63 2.6 .8: $6 .3: 2:6 .53 No.2m .ooofi ~:.2n .nmnm cm.on .Nnum 2...» .an 3.2 .mwm 3.2 62. 8.2 .22 2.2 .22 2.2 .32 2. mm . 82 2. 2 . 22 3.9 .22 a: A! an. 2.: m: 8.~ 82m 95$ 25.2 .o2 x .oo .02 x 22. .o2 x .oo ufiu.- enamuocu owduo>< 2.2 2.2.2 .252 2.8 . 2% 8. 2 .30 2.2 .2o 26 ...mn 3.2 .22 2. 2 .25 2. 2 .32 8.2 .22 .uflmqmmmu _flaqmmm acouh mom Huh 9: 8n .2 32 x .oo we. 2 8.22 A22 8. on .82 2.2 .22 oc.a .emo R.: .R: 8.2 .82 3. 2 . S: No. 2 . 22 8.2.. .2va ...awqmmmm ..«uqmmm acouh ugh Han Sé 822 .o2 x .82 $22 34.2 .938 :23 . 8:. 2.2 .22 RS .2: 22.n .ann 8.2 .22 S. 2 .22 wa.m2 .52m2 2.2 . zen acouh Mom has mm.2 828 .2 x .o2 MAOH um>o ow:cuo:~ acwcuma n. _ <._.o._. saucy; mmcpu w emc:uc2. a anoc mmaam .5020: Lo amoc acowcaacoo d o:oc=c-wom.: someooamca d m:«::d4L wcao=H0:H .mzauoo=auzm ac:cmdm =:¢_ cwmcaagc é cede: Ec2mam peaxz amgmhm gcxcm 2::w_ cpc< L92 mac; A;oom .rmv ach< as; Aaccuv exam ‘0; .z~¢zfi nmown .uOhm>h=m a Macadam .uoocdwca m:«»a:m:oo ..mwm .Hfimmmzz .0 ho: an caudaoua 0:0 cofiaasoo .couosadu «00: .02.:0m $.m Ria 39a RN30 Rim thaw uo>o cadmiocn acoptwg _ $.03 .13: 30.03 .903 0.0.03 .033. 00.03 .803 303 .033 0.063 .8000 00:: 320. 005.0: 02222.00 .m 00.00 .0000 8:3 .000: 00.3.. .003 3.3 .008 00.00 .0000 3.00 .30... 303.. 0080 a 0.3.230 . a amoo momma .hwzaz no amcu Sufi .80 {.22 .30 00.3 .00: 00.3 .3: 00.: .00: 00.: .009 05003000 a. 252.0223. 3.. 2 .80 R. S .08 00.: .0n3 3.2. .33 «0.3 .83 00.3 .35. $0... :03 0&5: fl soa>o= amaaon::z 0:0 wzaumezamcm .wcamo>hsm .mcd::d.a 2.0 .00.. :03. .03 00.... .30 2.0 .03 8... .03 0.0.0 .000 28.80 0308.0 :00 0 pHJ wocmamm acq— .m:ficmuc .mcmudm_c l 2 00.00 .32 00.00 .300 00.00 .38 00.00 .mmmn 3.00 .0000 «ER .90.. 003.20 0w 3:35 0.35m .303: 7.3». 9.3 .000 00.3 .30 on.: .33 :5: .33 00.: .8: 00.3 .003 03...: 080: 00.2 .93 3.3 $03 2.2 .003 No.2 .83 9.: .95 933 .{3 E230 0.5200 3. mm . 003 3. mm . 023 8.9. Aim 8.0m . 3:0. 00.0w . 000m 8&0 . Ram 03... “3080 3:3 3:8... 2:3 3.... 8... 0303 3:8... 3:3 “308.. .303 “308.. 3:3 azouh Hon no; u=0Hm hog H03 acouh uma you ozouh kph Hon acouh Hmm Hmm acouh awn men Nn.: Ban om.~ c:.~ mod an; Eo< 00.. 30.. 000.0 000.0 000.3 000.3 000.3 000.00 39.: .50 .82 03 .03 x .00 .03 x .0... .03 x .00 .03 x .00 .03 x .03 R3 0. .03 :85 0.2.0 05 .zmn ZCHmHZA—mam aver—Hr! main mAma¢;& nwcmn .uoao>u=m and Ho::ddm .uooaawcn msaaaamnoo ..n.m .Haommzz .w to: an coummoum 6:3 coaagsoo .cwmozgww dgxa .9950»,H “n4. 2.82 63: ow.ow .HQH: £32 .8... nm.m2 .mmu 8;. .2... mné: .nofi $.02 .mne 3.2 .3: 3.9 .82 $508 3:3 »:ouh Hem hon Nn.: 8~.~. .ONH K .00 .mamoo anon nonu um.w undemoca omuuo>< ES a: “86 $6 3.2: .39: 3&2 632 9.58 .282 352 A32 inmw . awn Sé. .88 2.2. :38 3.8 .88 Emma .R: _ anfifl .32 no.3 .92 $2.: .5: an .2 .93 3.2 $22 2.2 .32 R.: 4.22 {2. .8... a...“ .Qe on.o .23 85 .oon 8. E coon em. on .8? on. an . 23m aw. R . SR mi: .Rm 8.2 .33 2.2 .32 9x2 .92 2.2 .32 No.2 .22 8.2 . $2 3.2 . 32 2.9 .82 8.8 _ .33 8.8 .Efi 8.3 45% 3:8... $33 238.. 2:3. 3:8: 2:2 2:8... 3:3 uzouh huh Hum ocouh you Hon a:bhh hon Rpm 9:0Hh no; Hpm min oc.m £3 8.~ 95$ 08.2 8n.2 8a.? .o2 x .nm .02 x .oo .22 x .oo .03 x .82 $3 9..an . 3:? 3.3 .88 mm.m2 .anH 2.: 4:2 36 gm: ~:.on .3? an .2 .83 u~.52 .mmom 8.2 .19 23%.... 3:3 28.. you a: an; 95.8 .50H x .ONH mmmq go>o mmdc:cc~ accouo; ”BE.— nh5o a once magma .aoccz no amoc acowcaacoc £ m:oo=mHHocma= moan zeaaoQ_m=H a zed>cz Hammoficzz tzd wcauwwcamca .uca»o>u:m .wcmczxdg mfiouucoc :OdmOHQE Haom d 8.2% 23 £532". £3235 .oumzwwhd a m:amua 590g“ .mqso: coas— mcamz awed: muozcm humaficam €33 cuc< ac; was: Aamoa .ccv aeu< ac: Aa0¢uv exam go; $20555. 3:05.39 .fl=0< can 806» .8 “.80 92¢.— nfixswa EH3 maN—d Bum nag—Jam :95 .050.” .589 hszSm—Ru: chmH>nGn=m GEEHPE calm m4m<5 APIENDIXC CALCULATED AGRICULTURAL OPPORTUNITY COSTS APPENDIX C CALCULATED AGRICUIJUEAL OPPORTUNITY COSTS Table C-l contrasts the calculated agricultural Opportunity costs and USDA data on the value of agricultural land and buildings. The calculation of agricultural Opportunity costs involved an examination of the National Atlas of the United States (U.S. Geological Survey, 1970) on a county basis within states to find counties having similar agricultural land uses to those counties undergoing urban pressures were listed and averaged by states. In a few instances prior knowledge by the author and fellow graduate students supplemented the process. These counties served as the base for the calculated agri- cultural opportunity costs for 1959, 1964, and 1974. 217 218 TABLE C—l CALCULATED AGRICULTURAL OPPORTUNITY COSTS AND VALUE OF AGRICULTURAL LAND AND BUILDINGS, 1969. Calculated Value of Agricul- State Agricultural tural Land and Opportunity Costs Buildings ($per acre) §$ per acre) Alabama 214. 187. Alaska --_ ___ Arizona 157. 67. Arkansas 316. 246. California 524. 487. Colorado 95. 92, Connecticut 859. 863. Delaware 399. 457. Florida 318. 319. Georgia 254. 214. Hawaii --- -_- Idaho 232. 168. Illinois 470. 493. Indiana 401. 417. Iowa 430. 382. Kansas 182. 162. Kentucky 257. 238. Louisiana 380. 302. Maine 222. 150. Maryland 490. 578. Massachusetts 469. 514. Michigan 299. 316. Minnesota 327. 216. Mississippi 255. 221. Missouri 241. 217. Montana 103. 56. Nebraska 294. 150. Nevada 197. 50. New Hampshire 270. 218. New Jersey 737. 968. New Mexico 81. 41. New York 299. 250. North Carolina 328. 337. North Dakota 130. 91- Ohio 391 . 378. Oklahoma 218. 162. Oregon 406. 143. Pennsylvania 282. 326. Rhode Island 549 . 684 . 219 TABLE C-l (continued) South Carolina 260. 259. South Dakota 154. 83. Tennessee 282. 252. Texas 196. 142. Utah 187. . 88. Vermont 258. 198. Virginia 330. 258. Washington 253. 215. West Virginia 139. 127. Wisconsin 327. 213. Wyoming 73. 38. Source: Calculated Agricultural Opportunity Costs from 1969 Census of - Agriculture. value of Agricultural Land and Buildings from Farm Real Estate Historical Series. APPENDIX D COMPARATIVE CASE STUDY INTERVIEW QUESTIONS APPENDIX D COMPARATIVE CASE STUDY INTERVIEW QUESTIONS Twenty developers, planners, and sundry other local government officials were interviewed in the Lansing, Kalamazoo, and Jackson, Michigan areas. The developer interviews were confidential and loosely structured around the questions listed below. These interviews followed examination of zoning maps and ordinances as well as other planning documents and census data. The process was iterative, involving telephone contacts with additional questions and issues raised by other interviews. General: 1. How long have you been a developer here? 2. What is the typical number of lots you've developed per year for the last five years? 3. What jurisdictions? 4. What kind of housing? Land: 5. What is the average price per acre of raw land you've bought over the last five years? average high low 6. What would you expect to pay for land this year? 7. How important is the price of raw land to your location decision? Explain. 8. How large are the lots you develop? average large small 9. 0n the average what are your development costs per lot? 220 221 10. What do finished lots go for? 11. What do you think is the percent attributable to land? Has it risen? 12. What are the size (variation) of tracks of raw land you have purchased for residential development? 13. Are there land availability problems? Zoning: 14. Does zoning affect where you build? How? 15. How does zoning affect your cost of doing business? Vary over jurisdictions? 16. How easy (by jurisdiction) is it to get zoning changes for residential and multiple family? 17. What do you think of the consistency of zoning decisions in jurisdictions you are familiar with? 18. Are there jurisdictions where building anything other than low density residential is difficult? Which? 19. How does zoning lot size (etc.) requirements affect the size of the lot you build upon? Do you always build on larger lots anyway? What is the most important zoning constraint? Sewer and Taxes: 20. Does Sewer availability affect your location decision? How? 21. Are there sewer supply problems? 22. Do sewer tap in fees and charges influence where you build? 23. Do jurisdictions (which) follow different policies on where you can (or how you can) hook on to sewers near other developments the end of the line? 222 24. Are there problems in getting sewer permits? Septic tanks? 25. How do you.think the size and variation of the property tax payments in various jurisdictions affect where homes are built? 26. What is the price difference between sewered and unsewered lots? Concluding General Questions: 27. How have development costs changed? 28. What do you see as the trend in land prices, past and future? 29. What are the main local government policies-in which you see variation which affect the price of land? APPENDIXE EXAMPLE OF DEVELOPMENT COSTS IN LANSING, MICHIGAN APPENDIX E EXAMPLE OF DEVELOPMENT COSTS IN LANSING, MICHIGAN Table E-l presents data gathered by Walter E. Neller, a develOper in the Lansing, Michigan area. This data was supplied to the Urban Land Institute to serve as an example of the impact of development costs upon the price of a housing site. 223 224 IEBLE E-l CATEGORIZATION OF TYPICAL DEVELOPMENT COST BREAKDOWN FOR.SELECTED SUBDIVISIONS IN LANSING, MICHIGAN $ Per 1 $ Per 2 Salable Ft. Salable Ft. Engineering and Planning 3.78 2.50 Water A. Distribution System (Including Taps) 9.02 7.75 B. Area Assessments Charges l.ll --- C. Off Site Expense --- .25 Sanitary A. Distribution System (Including Taps) 8.97 6.25 B. Area Assessment Charges 3.21 3.25 C. Off Site Expense -__ -_- Storm A. Distribution System 4.36 4.50 B. Area Assessment Charges 4.61 1.50 C. Off Site Expense -- 1.50 Streets A. Interior System 17.20 15.00 B. D and A Lanes and Passing Lanes 1.57 1.00 Grading 4.40 2.50 Sidewalks A. On Site 2.74 3.50 Underground Electrical --- 1.00 Entrances .32 .30 Lighting 1°09 1.75 Open Space A. Land 6.46 6.50 B. Improvements —-- --_ Misc. Expense Items 1.01 1.00 Land (Direct Only) 20.00 20.00 Administrative & Promotion 6.00 6.00 225 TQBLE E-l (continued) Deeds, Abstracts and Etc. 1.25 1.25 Carrying Costs 7.50 7.50 TOTAL DEVELOPMENT COST 104.60 95.45 Profit 19.40 19.55 TOTAL SALE PRICE 124.00 11 .OO Salable Subdivision NO. Of Lots Footage Acres "A" Sec. I 122 9,738 39.6 "B" Sec. I & II 222 17,729 67.94 Source: Walter Neller, Developer, Lansing, Michigan. 1Prepared for Urban Land Institute Exc. Comm. 10/14/77. 2 Prepared for Urban Land Institute Exc. Comm. 5/7/76. Alonso, William. (1964). Location and Land Use. Cambridge: Harvard University Press. Andrews, Richard B. (1972). Urban Land Use Policy: The Central City. New York: The Free Press. Babcock, Richard F. (1966). The Zoning Game. Madison: The University Of Wisconsin Press. Bish, RObert L. (1971). The Public Economy of Metropolitan Areas. Chicago: Markham Publishing Company. Bonnen, James T. (1975). "Improving Information on Agriculture and Rural Life." American Journal of Agricultural Economics. Vol. 57. Number 5. _(DecembeI)- Breimyer, Harold F. (1978). "Agriculture's Three Economies in a Changing Resource Environment." American Journal Of Agricul- tural Economics. Vol. 60. NO. 1. (February)? Bronsstein, Daniel A. and Erickson, Donald E. (1973). "Zoning Amend- ments in Michigan." Journal Of Urban Law. (May). Chinitz, Benjamin, and Cowing, Thomas G. (1974). "The Impact of Local Public Service Consolidation on Urban Spatial Structure." in Local Service PricingpPOlicies and Their Effect on Urban Spatial Structure. Downing, Paul B. ed. Vancouver: University Of British Columbia Press. Clawson, Marion, Burnell, Held R., and Stoddard, Charles H. (1960). Land For The Future. Baltimore: The Johns Hopkins Press for Resources for the future, Inc. Clawson Marion. (1960). "Suburban DevelOpment Districts." Jgurnal of the American Institute Of Planners. Vol. XXVI, Number 2. (May). Clawson, Marion, and Stewart, Charles L. (1965). Land Use Information. Baltimore: The Johns Hopkins Press. Clawson, Marion. (1971). Suburban Land Conversion in the United States: An Economic and Governmental Process. Baltimore: The Johns Hopkins Press for Resources for the Future, Inc. Commons, John R. (1950). The Economics of Collective Actign. Madison: The University of Wisconsin Press. 226 227 Council on Environmental Quality. (1974). Interceptor Sewers and Suburban Sprawl: The Impact of Construction Grants on Residen:_ ,tial Land Use. Vol. I and Vol. II. Washington, D.C.: C.E.Q. Council on Environmental Quality. (1976). The Growth Shapers: The Land Use Impacts Of Infrastructure Investments. Washington, D.C.: Superintendent Of Documents. (May). Crecine, John P., Davis, Otto A., and Jackson, John E. (1967). "Urban Property Markets: Some Empirical Results and Their Implications for Municipal Zoning." Journal of Law and Economigs. Currie, John M., Murphy, John, and Schmitz, Andrew. (1971). "The Concept of Economic Surplus and Its Use in Economic Analyses." The Economic Journal. Vol. LXXXI. (December). Delaphons, John. (1969). Land Use Controls in the United States, 2nd ed. Cambridge, Mass.: The M.I.T. Press. Downing, Donald A. (1972). “The Role of Water and Sewer Extension Financing in Guiding Urban Residential Growth. Master's Thesis: University of Tennessee. Downing, Paul B. (1969). The Economics Of Urban Sewage Disposal. New York: Praeger Publishers. Downing, Paul B., ed. (1974). Local Service Pricing Policies and Their Effect on Urban Spatial Structure. Vancouver, B.C.: Universityw of British Columbia Press. Eells, John Martin. (1977). ,LAFCO Spheres Of Influence: Effectivg Planning_for the Urban Fringe? Working Paper 77-3. Institute of Governmental Studies: University of California, Berkeley. Edel, Matthew. (1978). "Rent Theory and Working Class Strategy: Marx, George and the Urban Crisis." The Review of Radical Political Economics. Vol. 9. No. 4. (Winter). Ervin, David E., etal. (1977). Land Use Control: Evaluatinngconomic and Political Effects. Cambridge, Mass.: Ballinger Publishing Company. Fuguitt, Glenn U. (1962). "The Rural-Urban Fringe." Proceedings Of the Forty First Conference of the American County_Life Associa- tion, Inc. Washington, D.C. (July). Gaffney, Mason. (1961) "Land Rent in Welfare Economics." Land Economics Reagarch. Joseph Ackerman, et.al. eds. Baltimore: The Johns Hopkins Press. Gaffney, Mason. (1975). "The Many Faces Of Site-Value Taxation." Paper delivered at the TWenty-Seventh Annual Conference Of the Canadian Tax Foundation. (November). 228 General Accounting Office, Report to the Congress. (1978). Why are New House Prices SO Hi h How are The Influenced b Government Regulations, and Can Prices be Reduced? CED 78-101. (May 11). Gibbs, Jack P. (1961). Urban Research Methods. New York: D. Van Nostrand Company, Inc. Gold, Niel N. and Davidoff, Paul. (1967). "Supply and Availability of Land for Housing." The Rgport Of the President's Committee on Urban Housing. Washington, D.C.: U.S. Government Printing Office. Goldberg, Michael A. (1974). "Residential DevelOper Behavior: Some Empirica1.Findings." Land Economiqg. Vol. I. Number 1. (February). p. 85-89. Goldstein, Gerald S. and Moses, Leon N. (1973). "A Survey of Urban Economics." Journal of Economic Literature. Vol. XI. Number 2. (June). Gottlieb, Manuel. (1965). "Influences on Value in Urban Land Markets, U.S.A., 1956-1961." Journal of Regional Science. 6. Greene, Kenneth V., Neeman, William B., and Scott, Claudia D. (1974). Fiscal Interactions in a Metrgpolitan Area. Lexington, Mass.: Lexington Books. Hansen, David E. and Schwartz, S. I. (1975). "Land Owner Behavior at the Rural-Urban Fringe in Response to Preferential Property Taxation." Land Economics. Vol. LI. NO. 4. (November). Harvey, Robert O. and Clark, W. A. V. (1965). "The Nature and Economics Of Urban Sprawl." Land Economics. Vol. XLI. Number 1. (February). Herr, William MOD. (1977). "Importance of Various Factors in the Land Market." in Farm Real Estate. North-Central Regional Extension Publication NO. 51. —(May): James, Franklin J. Jr. with Windsor, Oliver Duane. (1976). "Fiscal Zoning, Fiscal Reform, and Exclusionary Land Use Controls." Journal Of the American Institute of Planners. (April). Johnson, K. H. and Lyon, H. L. (1973). "Experimental Evidence on Combining Cross-Section and Time Series Information." The Review Of Economics and Statistics. Vol. LV. Keiper, Joseph S., Kurnow, Ernest, Clark, Clifford, and Segal, Harvey H. (1961). The Theory and Measurement 9: Rent. Philadelphia: Chilton Company. 229 Kuh, Edwin. (1959). "The Validity of Cross-Sectionally Estimated Behavior Equations in Time Series Applications." ,Eeonometrica. Vol. 27. Kurtz, Richard A. and Eicker, Joanne B. (1958). "Fringe and Suburb: A Confusion Of Concepts." Social Forces. 37(1). (October). Lackman, Conway L. (1976). "The Classical Base of Modern Dent Theory." American Journal of Economics and Sociology. Vol. 35. NO. 3. (July)- Lackman, Conway L. (1977). "The Modern Development Of Classical Rent Theoryu" American Journal 9f Economics and SOOlology. VOl. 36, No. 1. (January). Lieberson, Stanley. (1969). "Measuring Population Diversity." American Sociological Reviefl. 34. (December). Macura, Milos. (1961). in Urban Research Methods. Gibbs, Jack P., ed. New York: D. Van Nostrand Company. Maisel, S. (1963). Land Costs for Single Family Housipg, Berkeley: California Housing Studies. Center for Planning and Development Research. University Of California. Markusen, J. R. and Scheffman, D. T. (1977). Speculation and Monopoly in Urban Development: Analytical Foundations With Evidence for Toronto. Buffalo: University of Toronto Press. Maser, Steven M., Riker, William H., and Rosett, Richard N. (1977). "The Effects of Zoning and Externalities on the Price of Land: An Empirical Analysis Of Monroe County New York." Journal of Law and Economice. Vol. XX(1). (April). McCahill, Ed. (1973). "Stealing: A primer on zoning corruption." Plannigg. Vol. 39, #11. (December). McMahan, John. (1976). Property_Develgpment: Effective Decision Making in Uncertain Times. New York: McGraw-Hill Book Company. 432 pgs. McMillan, Melville. (1973). Land-Use Control for the Preservation of Open Space on the Rural-Urban Fripge: An Alternative Policy. Working Paper NO. 7. Center for Resource Policy Studies, University Of Wisconsin. (December). Miller, Kenneth. (1977). An Intertemporal Cross-City_Comparison of Single-Family Housing Prices. Research Reports on Public Policy, Number 12. Urban Economics Program, University Of California, Santa Barbara. (April). 230 Mills, Edwin S. and Oates, Wallace E. (1975). Fiscal Zoning and Land Use Controls. Lexington, Mass.: Lexington Books, D.C. Heath and Company. Mittlebach, Frank G. and Cottingham, Phoebe. (1964). Some Elements in Interregional Differences in Urban Land Values. Los Angeles: Real Estate Research Program, UCLA. (June). Muth, R. F. (1971). "The Derived Demand for Urban Residential Land." Urban Studies, 8. Netzer, Dick. (1974). "Impact of the Property Tax: Effect on Housing, Urban Land Use, Local Government Finance." in Beaton, W. Patrick, ed. Municipal Needs, Services and Financing. New Brunswick, New Jersey: Center for Urban Policy Research, Rutgers University. Niedercorn, John H. and Hearlean, Edward F. R. (1964). "Recent Land- Use Trends in Forty-eight Large American Cities." Land Economics. Vol. XL. NO. 1. (February). Northam, Ray M. (1971). "Vacant Urban Land in the American City." Land Economics. Vol. XLVIII. NO. 4. (November). Ohls, James C., Weisburg, Richard Chadbourn, and White, Michelle. (1974). "The Effect of Zoning on Land Value." Journal Of Urban Economics. Vol. 1. Ottensmann, John R. (1975). The Changing Spatial Structure of American Cities. Lexington, Mass.: Lexington Books, D.C. Heath and Company. Ottensmann, John R. (1977). "Urban Sprawl, Land Values and the Density Of Development." Land Economics. Vol. 53. NO. 5. (November). Pindyck, Robert S. and Rubinfeld, Daniel. (1976). Econometric Models and Economic Forecasts. New York: McGraWeHill. Portney, Paul R., ed. (1976). Economic Issues in Metropolitan Growth. Baltimore: The Johns Hopkins University Press. Rao, Potluri, and Miller, Roger. (1971). Applied Econometrics. Belmont, California: Wadsworth Publishing Company. Romanos, Michael C. (1976). Residential Spatial Structure. Lexington, Mass.: Lexington Books, D.C. Heath and Company. Sagalyn, Lynn B. and Sternlieb, George. (1973). Zoning_and Housing Costs. Rutgers University: Center for Urban Policy Research. 231 Salter, Leonard A., Jr. (1942). "Cross-Sectional and Case-Grouping Procedures in Research Analysis." Journal Of Farm Economics. Vol. XXIV. NO. 3. (August). Schmid, A. Allan. (1968). Converting_Land from Rural to Urban Uses. Washington, D.C.: Resources for the Future, Inc. Schmid, A. Allan. (1970). "Suburban Land Appreciation and Public Policy." Journal of the American Institute of Planners. Vol. XXXVI. NO. 1. (January); Schmid, A. Allan. (1979). Power, Propeppy, and Public Choice. New York: Praeger Publishers. Schwartz, S. I. and Hansen, D. E. (1975). "Two Methods for Preserving Agricultural Land at the Urban Fringe: Use-Value Assessment and Transferable Development Rights." Agriculture and Epyironment, 2. Shryock, Henry S., Siegel, Jacob S., and Associates. (1975). The Methods and Materials of Demogpaphy. Washington, D.C.: U.S. Government Printing Office. Sinclair, Bill and Manderscheid, Lester V. (1974). A Comparative Evaluation Of Indexes Of-Rurality - Their Policy Implication and Distributional Impacts. Michigan State University. Department of Agricultural Economics. CRMPA Special Report 22. (August). Sumichrast, Michael and Frankel, Sara A. (1970). Profile Of the Builder and His Industry. Washington, D.C.: National Association of Home Builders. Tabors, Richard D., et. a1. (1976). Land Use and the Pipe. Lexington, Mass.: Iexington Books, D.C. Heath and Company. ' Tiebout, Charles M. (1956). "A Pure Theory of Political EXpenditures." Journal of Political Economy. Vol. 64. U.S. Bureau Of the Census. (1959, 1964, 1969, 1974)° Census of Agpiculture. Washington, D.C.: Superintendent of Documents. U.S. Bureau Of the Census. (1970). Census of Housing. Washington, D.C.: Superintendent of Documents. U.S. Bureau of the Census. (1964-1975). Census of Local Government: Financial Characteristics Of Selected Metropolitan Areas. Washington, D.C.: Superintendent of Documents. U.S. Bureau Of the CensUs. (1960, 1970). Census of Population. Washington, D.C.: Superintendent Of Documents. U.S. Bureau Of the Census. (1962, 1967, 1972). U.S. Government Printing Office, Washington, D.C. 232 U.S. Deparment Of Agriculture, Economic Research Service. (1973). Farm Real Estate Historical Series DataL5185O-197O. ERS 520. Washington, D.C. (June). U.S. Department of Commerce. (1975). 1975pBusiness Statistics Washington, D.C. (May). U.S. Department of Housingand Urban Development. (1978). Final Report Of the Task_Eorce on Housing Costs. Washington, D.C. (May). U.S. Federal Housing Administration. (1948-1976). FHA Homes: Data for States and Selected Areas. Washington, D.C. U.S. Geological Survey. (1970). The National Atlas Of the United States. Washington, D.C. VanVuuren, W. (1976). Distribution of Gains and Losses Resulting From Planning Legislation: The Compensation-Betterment Problem. Univ. Of Guelph Publication AE/7678 CRD. NO. 76. (August). Wendt, Paul F. and Goldner, W. Uiam. (1976). "Land Values and the Dynamics Of Residential Location" in Essays on Urban Land Economics. Los Angeles: Real Estate Research Program, U.C.L.A. Wingo, Lowden. (1961). Transportation and Urban Land. Washington: Resources for the Future. ' Williams, Norman Jr., ed. (1978). American Land Planning Law: Cases and Materials. New Brunswick, N.J.: Center for Urban Policy Research, Rutgers University. Witte, Ann Drysen. (1975). "The Determination Of Interurban Residential Site Price Differences: A Derived Demand Model With Empirical Testing." Journal of Regional Science. Vol. 15. NO. 3. Witte, Ann Drysen. (1977). "An Examination of Various Elasticities for Residential Sites." Land Economics. Vol. 53. NO. 4. (November). i I llHIWIIHHIIWIWNIUHIUWMI‘ll\IWHIHHHWHI