Treating open space as an urban amenity

https://doi.org/10.1016/S0928-7655(01)00055-0Get rights and content

Abstract

In “the welfare economics of city bigness”, George Tolley asserts that the virtual price of amenities can be used to judge the efficiency of a urban spatial land use patterns. Expanding this test to open space amenities is not straightforward because those amenities are especially difficult to characterize. Bockstael and Irwin [Economics and the land use—environment link. In: Tietenberg, T., Folmer, H. (Eds.), International Yearbook of Environmental and Resource Economics, 2000/2001. Edward Edgar, Cheltenhan, UK, 2000] suggest that open space amenities and their virtual prices depend on whether surrounding land uses are fixed or adjustable. This paper estimates hedonic price functions over nearly 30 years to evaluate, whether the distinctions between fixed and adjustable land uses help in measuring the value of open space amenities.

Introduction

Twenty-five years ago concern about the spatial distribution of growth was framed in terms of the centralization of economic activities, and its implications for externalities and congestion. In “the welfare economics of city bigness”, George Tolley presents a model of how commodity and factor markets interact in generating the inefficient resource allocation predicted by Pigou when there are externalities. Indeed, Tolley’s paper anticipated the current literature on urban sprawl. In the decade and a half following the publication of the Tolley paper, much of the literature on the spatial distribution of growth was not concerned with sprawl, but whether the spatial dimensions of cities resulted from an orderly market process. This literature tended to argue that markets served to balance competing claims for land.1 The basic logic of Tolley’s framework argued otherwise—it found that markets did not always “get it right”.

Today, the arguments are in accord with Tolley’s model, though the problem is framed in terms of the decentralization of economic activities. Brueckner (2000) describes the market failures contributing to the excessive spatial growth of cities and notes they include the following:

  • a failure of local governments to account for the social value of open space when land is converted to urban use;

  • a failure of commuters to recognize the social costs of congestion created by their use of a pre-existing road system;

  • a failure of developers to consider all the public infrastructure costs generated by real estate development projects.

The phrases “increased urban sprawl”, “loss of open space”, and “loss of rural amenities” are used to characterize the spatial consequences of economic growth that takes place without signaling these external consequences.

Tolley’s model indicates that the virtual price of amenities can be used to judge the efficiency of a spatial land use patterns, as well as land use policies. Expanding this conclusion to include open space amenities is not straightforward because open space amenities are especially difficult to characterize. Parks and greenways, undeveloped private land, golf courses, and even some developed land uses can contribute to a landscape that is perceived to be protecting open space amenities. It seems that not only is the amount of open space important, but its location relative to other land uses also matters. Measures of the ways spatial growth affects the externalities associated with sprawl and, with them, estimates of benefits to people of avoiding these effects are needed to design the required policy responses.

One might argue we could develop measures using advances in geographic information system (GIS) techniques to characterize the land uses around residential properties, and then to estimate how these patterns affect residential sales prices. Unfortunately, these price models will not necessarily indicate the incremental value that the homeowners associate with open space amenities.

As Bockstael and Irwin (2000) point out, the success of property value methods in isolating preferences for land uses depends, in part, on whether the uses are perceived as fixed. At any particular time, in addition to the characteristics specific to a house and its associated land parcel, the other factors affecting a residential home’s price will include the nearby, fixed land uses, as well as the market expectations about uses for the undeveloped land close to that location. The effect of fixed land uses on a nearby site is expected to be consistent across individuals and time, while the effect of adjustable land uses is expected to vary across individuals and time. The appreciated value of nearby residential properties (due to land that is currently undeveloped and contributes to open space) increases the prospect for that undeveloped land to be converted to additional residential sites, thereby diminishing a part of the reason for the value of existing residential sites—the open space amenities provided by the undeveloped site. Private households and developers recognize these processes are at work.

Thus, the extent to which property values reveal the value of open space amenities depends in part on the relative contributions of fixed and adjustable land uses to the total open space amenities. We have undertaken two types of analyses to sort out these diverse influences on the residential market. First, we evaluate hedonic price functions for recent sales in an area selected because a new interstate “loop” highway has created expectations for changes in access throughout Northern Wake County in North Carolina. An increased pace of land conversions has accompanied the construction and opening of parts of this new road.

Second, we analyze a time profile of sales in this area over nearly 30 years in which higher density land uses associated with the “loop roadway” become more prevalent. Hedonic models estimate how proximity to fixed versus adjustable land uses influences residential property values over time. By comparing the changes in estimates of the effects of proximity to different types of land uses on property values over time, we evaluate whether the distinctions between fixed and adjustable land uses help in measuring the incremental value of open space amenities.

Our analysis focuses on the Research Triangle area of North Carolina. The state is ranked fifth among states in the number of acres of land developed between 1992 and 1997 (Center on Urban and Metropolitan Policy, 2000). Due to rapid population and metropolitan growth, 40% of farms in the Raleigh area (the location close to the interstate “loop”) were developed between 1969 and 1997 (Center on Urban and Metropolitan Policy, 2000).

Section 2 summarizes the problems in using hedonic price functions to measure the incremental value of open space amenities. We describe how the interactions between the various land uses can influence an individual’s locational choices and what we can infer about preferences from a hedonic model. The section outlines our strategy for evaluating the importance of fixed versus adjustable land uses as open space measures in hedonic price functions. Section 3 describes our study area and the data used in this analysis. The estimated hedonic property value models are presented in two parts. Section 4 discusses the estimates for the most recent time period (1995–1998) and Section 5 considers how they have evolved over the period 1980–1998. Section 6 summarizes our overall findings and then returns to Tolley’s contribution on the modeling of city size and discusses the information we would need to consider how open space would affect his analysis of locational equilibria.

Section snippets

Background

The basic logic used to explain how public goods (or externalities) affect the conditions for an efficient resource allocation generally relies on simple, polar cases. In public goods examples, the commodity is usually defined to have the non-rival and non-exclusive attributes that distinguish a public from a private good. Open space amenities are more complex because each agent’s private land use decisions can have consequences for others. That is, at any instant the overall open space amenity

Background

The study area in Northern Wake County, North Carolina lies in a suburban area north of the state capital. As the area around the Research Triangle area has grown, this region has also experienced rapid population growth. It offers a location convenient to downtown Raleigh, as well as to the nearby Research Triangle Park. The decision to construct this outer loop, as well as its location in relationship to other land uses can be considered exogenous to the time period spanned by our data on

Results for the most current period 1995–1998

The first step in our analysis is to consider the estimates of the hedonic price functions for the most recent sales available. These estimates are given in Table 3. Both semi-log and linear Box–Cox models are reported with the adjusted prices. Two specifications are reported for each model. One includes a distance measure for agricultural and forested lands (both the minimum distance and the dummy variable to identify adjacent properties) based on 1999 and 2000 land use classifications. The

Comparative results: 1980–1998

Table 5 provides estimates for our primary specification for the hedonic model (i.e. the semi-log form) for each of five sub-periods spanning 1980–1998. The dependent variable is the natural log of price adjusted for price appreciation using the fixed effect regression explained in Section 3. The specification is limited to structural characteristics and measures of spatial land uses that we can be assured remain in either the fixed or adjustable categories for each sub-period. That is, the

Summary and Implications

Our application has explored the prospects for using the land conversions that accompany new highway infrastructure to study whether fixed and adjustable land uses contribute differently to the open space amenities sought in residential home sites. By considering home sales in one area over time, it is possible to observe changes in the mix of land uses around homes. An advantage of limiting the analysis to one study area, and using several different short time periods, is that the general

Acknowledgements

V.K. Smith’s research was partially supported by the North Carolina Agricultural Research Service Project #NC06572 and the US Environmental Protection Agency. Thanks are due to two anonymous referees for very helpful comments on an earlier draft of this paper and especially to Glenn Blomquist for his careful review and editorial suggestions. We are also grateful to Michelle Holbrook and Charles Fulcher in assembling some of the data used in this analysis, as well as to Ray Palmquist and Randy

References (16)

  • A. Chandra et al.

    Does public infrastructure affect economic activity? evidence from the rural interstate highway system

    Regional Science and Urban Economics

    (2000)
  • G.S. Tolley

    The welfare economics of city bigness

    Journal of Urban Economics

    (1974)
  • Amemiya, T., 1985. Advanced Econometrics. Harvard University Press, Cambridge,...
  • Bockstael, N.E., Irwin, E.G., 2000. Economics and the land use—environment link. In: Tietenberg, T., Folmer, H. (Eds.),...
  • J.K. Brueckner et al.

    The economics of urban sprawl: theory and evidence on the spatial size of cities

    Review of Economics and Statistics

    (1983)
  • J.K. Brueckner

    Urban sprawl: diagnosis and remedies

    International Regional Science Review

    (2000)
  • Center on Urban and Metropolitan Policy, 2000. Adding It Up: Growth Trends and Policies in North Carolina. Z. Smith...
  • M.L. Cropper et al.

    On the choice of functional form for hedonic price functions

    Review of Economics and Statistics

    (1988)
There are more references available in the full text version of this article.

Cited by (123)

  • The value of public urban green spaces: Measuring the effects of proximity to and size of urban green spaces on housing market values in San José, Costa Rica

    2021, Land Use Policy
    Citation Excerpt :

    Evidence is non-conclusive regarding the effects of non-developed green areas in housing market transactions. Some studies have shown that housing prices decreases when closer to non-developed natural areas, urban and peri-urban forests, or brownfields (Tyrväinen, 1997; Smith et al., 2002; Panduro and Veie, 2013; Krekel et al., 2016). However, other studies show that housing prices increase when closer to wetlands (Mahan et al., 2000; Tapsuwan et al., 2009), greenways (Cho et al., 2006), and a main city river (Tajima, 2003).

View all citing articles on Scopus
View full text