Planners and Economists Debate Land Market Policy

Paul Cheshire, Rosalind Greenstein, and Stephen C. Sheppard, Enero 1, 2003

The land market allocates land and access to urban amenities, and it does so with impressive efficiency. Yet, economists and planners continue to debate the extent to which the market fails to achieve broader social goals, how far regulation can offset for that failure, and even whether regulation results in land market outcomes being even farther from the socially desired outcome than would be the case without any regulation. To examine this debate and the underlying issues, more than 30 economists and planners met at the Lincoln Institute in July 2002 to encourage new policy-relevant analysis on land markets and their regulation, and to foster more fruitful communication between the disciplines.

At the center of the substantive debate was the basic question of regulation within a market economy and the unintended consequences that can result. The discussions touched upon many themes including gentrification, the use of public resources for private consumption, distributional issues, urban form and its regulation. If perspectives regarding market regulation differed between the two disciplines, so too did views regarding the strengths and limitations of the analytic tools that academics from different disciplines bring to such thorny problems. Among the challenges are the basic questions of how to define the problem, how to measure the current conditions in light of limited data, and how to interpret findings. Throughout the conference, the differences in the perspectives, assumptions, tools and references between planners and economists were ever present, in particular with regard to the role of politics in planning and policy making.

Unintended Consequences of Land Market Regulations

Despite their differences, concern for land markets and their centrality to social, political and economic life was the common focus of both economists and planners at the conference. They agreed that land markets are about far more than land. These markets have an important role in delivering life experiences and conditioning the welfare of the majority of people in developed and developing countries alike who live and work in cities. In addition, their regulation has both direct and indirect economic effects that extend into many areas of economic life and public policy. For example, the urban poor are likely to have worse schools and to experience higher levels of neighborhood crime because land markets capitalize the values of neighborhood amenities, such as better school quality and lower crime, thereby pricing poorer households into less desirable neighborhoods.

This power of land markets to reflect and capitalize factors that affect a household’s welfare was revealed in a study of impact fees levied on new development in Florida. Ihlanfeldt and Shaughnessy found that impact fees appear to be fully capitalized into house prices for owners of new and existing houses by redistributing the costs of new infrastructure provision from existing taxpayers to a reduced value of development land. In fast-growing Miami the cost of impact fees was borne by developers, yet offset by the increases they received in higher prices for new housing, “while buyers of new homes are compensated for a higher price by the property tax savings they experience. In contrast to the neutral effects that fees have on developers, landowners, and purchasers of new housing, impact fees provide existing homeowners a capital gain” (Ihlanfeldt and Shaughnessy, 26).

One complement to their story of Florida’s impact fees was illustrated in several other papers concerned with the unintended outcomes of regulation. British participants reported that Britain’s containment policy has generated higher densities within urbanized areas, but cities leapfrog out across their Greenbelts (or growth boundaries) to smaller satellite settlements; the consequence is that development becomes less contiguous and travel times increase. Villages become high-density suburbs surrounded by a sea of wheat: London in functional terms extends to cover most of southeastern England.

In a U.S. example based on an econometric simulation, Elena Irwin and Nancy Bockstael found that a clustering policy intended to preserve open space could instead backfire. Using Maryland data, they simulated the effects of a policy that was intended to preserve rural open space and found that it would instead accelerate development if “small to moderate amounts of open space are required to be preserved (specifically, 20 acres or less) and would slow the timing of development if larger amounts of open space are required to be preserved” (Irwin and Bockstael, 26). Their simulation results yield an interpretation that is highly nuanced and requires careful thought. That is, under certain conditions the cluster policy can backfire, while under other specific conditions the policy can yield an intended policy outcome.

These hypothetical clusters in Maryland may be echos of a real situation that Jean Cavailhès and his colleagues observed in the French countryside, where some urban dwellers moved to farm regions to create a mixed-use area that is neither entirely urban nor entirely rural. These former urbanites appear to value their proximity to a functioning rural landscape in exchange for longer commutes and (surprisingly) smaller residential lots. The authors hypothesize that these peri-urban dwellers benefit in different ways from living among the farmers.

In another example of the unintended consequences of regulations, Donald Shoup analyzed curbside parking. Many U.S. municipalities require developers to provide minimal parking per square foot of new commercial or, in some communities, residential space. The requirement for off-street parking, coupled with a systematic underpricing of curbside parking, has a double impact, according to Shoup. It imposes a substantial tax on affected developments (equivalent to up to 88 percent of construction costs), increases land taking, and means that public revenues annually lost an amount equal to the median property tax.

In these cases of unintended consequences of policy or regulatory interventions in the market, the authors argued for more careful design of both policies and regulations so state and local governments could reasonably achieve their policy goals. Despite the fact that the conference debate tended to pit regulation against the market, there was probably a tendency—if not full-fledged consensus—to favor market incentives and disincentives to achieve policy goals, rather than to rely strictly, or even largely, on regulation. Roger Bolton’s comments on Shoup’s paper cogently reflected this viewpoint. He said that Shoup’s work was valuable because it urges us to pay attention to a whole package of “important and related phenomena: inefficient pricing of an important good, curb parking; inefficient regulation of another good, privately owned off-street parking; and missed opportunities for local government revenue.”

Data and Measurement Challenges

Growth management and urban form were referenced extensively throughout the conference. The paper presented by Henry Overman, and written with three colleagues (Burchfield et al.) provided useful grounding to that conversation. They attempted to measure the extent of sprawl for the entire continental U.S. Using remote sensing data they calculated and mapped urban development and the change in urban land cover between 1976 and 1992. They defined sprawl as either the extension of the urban area, or leapfrog development, or lower-density development beyond the urban fringe. They concluded that only 1.9 percent of the continental U.S. was in urban use and only 0.58 percent had been taken for urban development in the 16-year period covered by the study. Furthermore, during this period, urban densities were mostly on the increase.

This study found development to be a feature of the “nearby urban landscape,” whether that was defined as close to existing development, or near highways or the coasts, and thus was perceived as encroaching on where people lived or traveled. The authors use this last observation to reconcile the apparent contradiction between their finding that less than 2 percent of the continental U.S. has been developed and the fact that containing and managing sprawl is at the center of policy agendas in many states and regions across the U.S. While relatively little land might have been consumed by new development in aggregate during the study period, many people see and experience this development on a daily basis and perceive it to represent significant change, often the kind of change they do not like.

The conference discussion touched upon some of the data questions raised by this work. The paper’s discussant, John Landis, noted some challenges he has faced in working with these and similar data to measure growth patterns in California. The estimates by Burchfield et al. are extremely low, possibly for technical reasons, according to Landis. Among the reasons is the difficulty in interpreting satellite images and the different outcomes that can occur when different thresholds are used for counting density, for example. That is, an area can be classified as more or less dense depending on what threshold the analysts establishes. “Ground-truthing” is required to remove some of the arbitrariness from the analysis, but this is an enormously costly undertaking.

Policy analysts are always faced with data limitations. Sometimes the problem is missing data, while other times it is data with questionable reliability. Yet, all too often researchers spend very little time paying attention to how serious that deficiency is for the policy problem at hand. When the available data is a very long time series with frequent intervals that relies on a well-structured and well-understood data collection method, and where few transformations occur between data collection and data use, most researchers and policy analysts would feel extremely comfortable interpolating one or two or even a handful of missing data points. Econometricians relying on data collected at regular intervals from government surveys frequently face this situation and are quite adept at filling in such “holes in the data.” In the world of limited data, that might be considered the best-case scenario.

At the other extreme we might have data that are collected using relatively new methods and that require significant transformation between collection and use. Data reliability likely decreases under these circumstances. Given the imperfect world in which we live, the answer is probably not to insist on using only the “best data.” However, researchers and policy analysts do have the obligation to use care in interpreting results based on weak data and to convey that weakness to their audience.

Another side of the limited data problem is the translation from concept to measure, and it explains why the conference participants spent so much time discussing “What is sprawl?” For researchers this question becomes “How does one define sprawl in such a way that one can measure it?” Burchfield et al. define sprawl as leapfrog or discontiguous urban development. Landis argues for “a more multi-faceted definition of sprawl, one that also incorporates issues of density, land use mix, and built-form homogeneity.”

Definitions are not trivial in policy analysis. If we cannot define the problem or the outcome, and we cannot measure it, how can we know if it is getting better or worse, and if our policies are having an impact? On the other hand, a very precise definition of a different but perhaps related concept may lead to unnecessary intervention. The new policy may improve the score on the measure but have little or no effect on the problem. For a variety of reasons (perhaps in part the customs and cultures within different disciplines) the economists at the conference tended to favor concepts that are simple and for which the data exist. On the other hand, the planners tended to favor concepts that are messy. In the end, one is left with weaknesses on both sides. The uni-dimensional definition, and therefore the uni-dimensional measure, may provide many of the desirable properties that allow statistical analyses. Multi-dimensional concepts are difficult to translate into measures. Which is better for policy making?

The Political Nature of Land Policy

Planning as a political activity was emphasized by several authors, notably Chris Riley (discussant of papers by Edwin Mills and Alan Evans), to emphasize the importance for economists to recognize this role and the constraints it imposes on significant change (particularly given the capacity of land markets to capitalize into asset values the amenities generated by planning policies themselves). Richard Feiock added there was also evidence that the forms of planning policies that communities selected (both the severity of such policies and the degree to which they relied on regulation in contrast to market instruments) could be largely accounted for by the political structure and socioeconomic and ethnic composition of those communities.

Participants reacted differently to the political nature of land policy and planning. For some this was problematic: it meant that the market was not being allowed to work. For others, it meant that the political process in a democracy was being allowed to work: the people had spoken and the policy reflected the expressed will of the body politic.

Reflections on Debate

The differences between economists and planners will continue, and differences among practitioners in different countries and even different parts of the same country (notably the large United States) can either stimulate or thwart future debates over the study of land market policies and implementation. Perhaps, though, the word debate itself thwarts our efforts. In debates, the debaters rarely change their minds. They enter the debate with their point of view firmly fixed and do not get “points” for admitting that their debating opponent taught them something or that they have consequently changed their own mind. However, one purpose of a professional conference is, indeed, for thoughtful people to consider their own assumptions and to be informed and changed by the points of view of others. In the future, perhaps debates will be supplanted with reflective conversation.

Paul Cheshire is professor of economic geography at the London School of Economics, England; Rosalind Greenstein is senior fellow and cochair of the Department of Planning and Development at the Lincoln Institute; and Stephen C. Sheppard is professor in the Department of Economics at Williams College, Massachusetts. They jointly organized the Lincoln Institute conference, “Analysis of Urban Land Markets and the Impact of Land Market Regulation,” on which this article is based.

Conference Papers

The conference participants whose papers are cited in this article are noted below. All conference papers and discussants’ comments are posted on the Lincoln Institute website ( where they can be downloaded for free

Burchfield, Marcy, Henry Overman, Diego Puga and Matthew A. Turner. “Sprawl?”

Cavailhès, Jean, Dominique Peeters, Evangelos Sékeris, and Jacques-François Thisse. “The Periurban City.”

Feiock, Richard E. and Antonio Taveras. “County Government Institutions and Local Land Use Regulation.”

Ihlanfeldt, Keith R. and Timothy Shaughnessy. “An Empirical Investigation of the Effects of Impact Fees on Housing and Land Markets.”

Irwin, Elena G. and Bockstael, Nancy E. “Urban Sprawl as a Spatial Economic Process.”

Shoup, Donald. “Curb Parking: The Ideal Source of Public Revenue.”