On the origins of land use regulations: Theory and evidence from US metro areas

https://doi.org/10.1016/j.jue.2012.10.002Get rights and content

Abstract

We model residential land use constraints as the outcome of a political economy game between owners of developed and owners of undeveloped land. Land use constraints benefit the former group via increasing property prices but hurt the latter via increasing development costs. In this setting, more desirable locations are more developed and, as a consequence of political economy forces, more regulated. These predictions are consistent with the patterns we uncover at the US metropolitan area level.

Introduction

Land use regulations vary tremendously in shape and scope across space and have become more widespread and stringent over time. They can, in principle, raise welfare by correcting market failures. Recent evidence, however, casts doubt on this proposition: land use regulations impose – via increasing housing costs – an enormous gross cost on households that is unlikely to be matched by welfare gains arising from correcting market failures. Understanding the causes of these regulations is thus of primary economic policy importance. Yet, perhaps because a large part of the regulatory costs are indirect and the underlying political-economy processes such as lobbying are difficult to observe empirically, this area of research remains relatively underexplored.

In this paper we propose a political economy model of landowner influence in which land use regulations are the outcome of redistribution motives only. In the model, owners of developed residential land favor additional land use constraints as this raises the price of their land; owners of undeveloped land oppose such tightening because it increases the cost of development. Mobile households evaluate heterogeneous local amenities and housing costs and pick locations accordingly. The model leads to two key equilibrium relationships: first, places with desirable amenities are more populated and their land is more developed than that of less desirable places; second, places that are more developed adopt tighter land use regulations. We find that both theoretical predictions are consistent with patterns we uncover in a cross-section of US Metropolitan Statistical Areas (‘MSAs’ or ‘cities’ henceforth). We also quantify these effects and find them to be economically meaningful.

The spreading adoption of land use regulations is a phenomenon that seems to accompany the rise of urbanization. In the early 20th century, when only about a quarter of the world population lived in urbanized areas, virtually no city had any zoning laws. San Francisco in 1880 and New York City in 1916 were early exceptions. Now that over half of the world population lives in cities, land use regulations are ubiquitous in all developed and most developing countries. Our model and our data are cross-sectional, but the logic of the model suggests that land use regulations are a by-product of urban development.

Formally, we design a theory that complements extant political economy models of land use regulation by showing how development leads to regulation. Specifically, we construct a discrete choice model in which a given population of mobile households has heterogeneous preferences over a set of cities. Some cities are endowed with more desirable amenities than others and such cities have a higher equilibrium population and are more developed. Each city comprises several local jurisdictions, each of which has a local planning board. Each planning board chooses land use regulations that are subsumed in a ‘regulatory tax’ (Glaeser et al., 2005a, Glaeser et al., 2005b) in a non-cooperative fashion as in Brueckner, 1995, Helsley and Strange, 1995, Schone et al., forthcoming.

The working of the model rests on three main features. First, a regulatory tax raises the price of developed land and increases the conversion cost of undeveloped land, putting the owners of developed land (i.e., homeowners and landlords who rent out their properties) against the owners of undeveloped land. Landowners lobby the planning board and the equilibrium degree of regulation reflects these ‘land based interests’ (Molotch, 1976). The marginal (net) value of land use regulations increases with the size of the regulatory tax base of a jurisdiction at the political economy equilibrium, that is, the marginal value of a regulatory tax is increasing in the share of developed residential land (SDL henceforth) of the jurisdiction. This relationship is upward sloping in the SDL-regulatory tax space of a jurisdiction and we refer to it as the ‘political economy response curve’ in Fig. 1. Second, a regulatory tax also increases the cost of living. This effect decreases the equilibrium population and the SDL of heavily regulated jurisdictions relative to the laisser-faire. This relationship is downward sloping in the SDL-regulatory tax space of a jurisdiction and we refer to it as the ‘market response curve’ in Fig. 1. Third, more desirable cities attract more households, all else equal; greater desirability shifts the market response curve to the right, as illustrated by the arrow in Fig. 1. Putting all three features together, desirable jurisdictions are more developed and they charge a higher regulatory tax at equilibrium than less desirable jurisdictions.

The main theoretical contributions of our paper are to (i) develop a formal model of landowner influences that predicts a link between urban development and regulatory stringency and (ii) propose a combination of a discrete choice model for across-city (macro) location decisions and of a standard monocentric city model for within-city (micro) location decisions. This combination provides a useful generalization of the currently available extreme versions of the monocentric city model, whereby each city is either fully isolated or fully open and small (Brueckner, 1987). In our model, both the population size and average utility levels vary across cities and are determined endogenously.

The main empirical contribution of our paper is to establish the robustness of the positive correlation between SDL in 1992 against the Wharton Residential Land Use Regulation Index (WRLURI) in 2005 for our reference sample of the 93 largest US MSAs. Fig. 2 (panel a) plots this relationship; the unconditional correlation between these variables, ρ = 0.31, is statistically larger than zero. Fig. 2 (panel b) suggests that this pattern was already visible in earlier data from the late 1970s, with ρ = 0.34. We show that the positive relationship between SDL and regulatory stringency is robust to the inclusion of a battery of control variables, region fixed effects, and some instrumental variation of the SDL.

In what follows, Section 2 reviews related work. Section 3 presents the model. Section 4 describes the data, provides baseline results, and discusses a number of robustness checks. Section 5 concludes.

Section snippets

Related literature

Evenson and Wheaton, 2003, Glaeser and Ward, 2009 regress measures of various types of land use regulations on historical and other characteristics of Massachusetts towns. Whereas Evenson and Wheaton (2003, p. 223) conclude that zoning seems to follow the current market, Glaeser and Ward (2009, p. 266) find that ‘the bulk of these rules seem moderately random and unrelated to the most obvious explanatory variables’. Our analysis shows that looking at aggregated measures of regulation across the

The model

The set of players and the timing of the game are as follows. In stage 1, the planning boards of a set of local jurisdictions (which differ in exogenously given characteristics) simultaneously choose a zoning policy, taking the other planning boards’ choices as given. Each jurisdiction belongs to exactly one MSA and the set of MSAs is a partition of the set of jurisdictions. In stage 2, households make location decisions of two kinds. They first choose a jurisdiction where to live; a bidding

Empirical analysis

The main purpose of our empirical analysis is to explore whether the unconditional positive correlation between physical residential development and regulatory restrictiveness at the MSA-level – illustrated in Fig. 2 – is robust to adding other potential explanatory variables, estimating alternative specifications, using variant proxy measures, and applying some instrumental variation of the share developed land measure in an attempt to account for the endogeneity of residential development to

Concluding remarks

This study contributes to the understanding of political economics considerations that shape land use restrictions. Consistent with our model of landowner influence, our empirical analysis unveils a strong positive correlation between the degree of physical development of a metro area and the regulatory restrictiveness of its residential land use. Our analysis focuses on residential land use by the nature of the regulatory data available. In practice, zoning also separates incompatible land

Acknowledgments

We are grateful to the Editor, William Strange, to Gilles Duranton and Henry Overman for thorough discussions, detailed feedback and encouragement, and to two anonymous referees for helpful comments and suggestions. We also thank Alex Anas, John Antonakis, Jan Brueckner, Marius Brülhart, Jean Cavailhès, Paul Cheshire, Steve Gibbons, Yoshitsugu Kanemoto, Gerrit Knaap, Rafael Lalive, Tim Leunig, Jim Markusen, Florian Mayneris, Muriel Meunier, Mark Partridge, Esteban Rossi-Hansberg, Michael

References (42)

  • S.P. Anderson et al.

    Discrete Choice Theory of Product Differentiation

    (1992)
  • M. Arzaghi et al.

    Networking off Madison Avenue

    Review of Economic Studies

    (2008)
  • D. Austen-Smith

    Campaign contributions and access

    American Political Science Review

    (1995)
  • R.E. Baldwin et al.

    Entry and asymmetric lobbying: why governments pick losers

    Journal of the European Economic Association

    (2007)
  • Douglas.B. Bernheim et al.

    Menu auctions, resource allocation, and economic influence

    Quarterly Journal of Economics

    (1986)
  • J.K. Brueckner

    The structure of urban equilibria: a unified treatment of the Muth-Mills model

  • J.K. Brueckner

    Strategic control of growth in a system of cities

    Journal of Public Economics

    (1995)
  • M. Burchfield et al.

    Causes of sprawl: a portrait from space

    Quarterly Journal of Economics

    (2006)
  • P.C. Cheshire et al.

    Office space supply restrictions in Britain: the political economy of market revenge

    Economic Journal

    (2008)
  • A. Dixit et al.

    Common agency and coordination: general theory and application to government policy making

    Journal of Political Economy

    (1997)
  • B. Evenson et al.

    Local variations in land use regulations

    Brookings Wharton Papers on Urban Affairs

    (2003)
  • Cited by (129)

    • Homeowner politics and housing supply

      2023, Journal of Urban Economics
    • The geography of mortgage interest deductions

      2023, Journal of Urban Economics
    View all citing articles on Scopus
    View full text