|
POSITION STATEMENT
There are few fields in economics where a recognition of the spatial dimensions of the analysis is more important than in environmental economics. As a result, there are a number of areas in environmental economics where more explicit integration of spatial analysis seems likely to lead to new insights. I selected four themes. Two of these relate to areas of established and continuing research and two are relatively new. All could benefit from the insights of those familiar with formal methods of spatial analysis.
I. Transport Modeling and Externalities
A major pre-occupation of environmental economists interested in externalities has been in specifying exactly how the activities of one (or more) agent(s) have negative (or positive) effects on others. For the case of air and water pollution from point sources, these inter-connections are often described with what are labeled transport models or, in simple cases, coefficients linking sources of the externalities and receptors of their effects. These models are an integral part of the definitions for the policies required to promote an efficient allocation of resources in the presence of externalities. Early examples of their role in: defining effluent charges can be found in Kneese and Bower [1979]; designing marketable permits systems in Tietenberg [1985]; and comparing the net benefits of command and control versus incentive based policies in Oates, Portney, and McGartland [1989]. With mobile and non-point sources of externalities, the spatial distribution of activities over an area influences the character of externalities. At the scale most commonly used to define environmental policy, time and spatially time delineated integrals of activity are often required to define the externalities that arise from mobile and non point sources. The economic models familiar to me are ill-equipped to deal with these issues in a way that has any hope of informing policy choices.
While most environmental economists recognize the importance of the technical relationships for point source questions, there are only a few examples of situations where the ways we treat them empirically have been explicitly evaluated. For example, in measuring the cost savings of an incentive based approach to controlling point and non-point sources how much does spatial delineation of differences in soil characteristics that affect surface and sub-surface loss of fertilizer and pesticides influence the conclusions? How does the modeling of the transport processes of streams and a river within a watershed influence the design of incentive based responses to TMDL (total maximum daily load) policies?
My own work and that of a former student, Kurt Schwabe (currently on the faculty of the University of California, Riverside) have attempted to address some of these questions for a watershed in North Carolina (see Smith et al. [1999], Schwabe and Smith [1999], and Schwabe [1996]).
Future research questions that are likely to be especially important in this area involve: (a) the use of statistical (as opposed to hydrological) models to describe how emissions from one location are linked to ambient concentrations in other locations and to judgments about the implementation plans for TMDL in watersheds; (b) are simplified rules defining the spatial rates of exchange in emission credits (or marketable pollution permits), such as what is used for RECLAIM, "good enough" to realize the efficiency gains from these policies? and (c) what is the relevant spatial unit of analysis for environmental externalities? Available political boundaries rarely overlap with the units describing these environmental interactions.
II. Spatial Delineation and Non-Market Valuation
As in the case of externalities, recognition of opportunities to use spatial differences in environmental conditions have been important to the measurement of people's values for changes in environmental resources. The two most common revealed preference approaches that rely on spatial differences in access to resources or in the available amenities are: the travel cost recreation demand and hedonic property value models.
In the first the spatial distribution of consumers around recreation sites defines implicit prices for using these areas for recreation. For hedonic models the spatial locations of homes "delivers" environmental amenities and disamenities. The process of defining what these might be in empirical studies has often been arbitrary and, as a result, rarely subjected to spatial analysis. Actually these types of questions are relevant to both methods. Some examples help to illustrate how explicit spatial analysis can play a role.
In the case of travel cost models the analysis often seeks to describe how consumers' decisions to use a site are influenced by its quality attributes. These choices are assumed to reveal consumers' preferences for these quality attributes. Meeting this goal requires linking measures of quality characteristics to each recreation site. To illustrate the spatial questions in resolving these issues consider the case of water quality indexes. The water quality measures are usually developed for one set of objectives (e.g., monitoring the sources of pollution) and now must be used for another -- assessing ambient quality that is perceived by users. In the first case the task for the measure involves judging whether an emission rate exceeds a standard. In the other it is to evaluate quality at the locations of a water based recreation site that people use.
How should these interconnections be made and are the processes used important to conclusions derived about recreation site choice and the valuation of amenities? There has been limited research on the importance of such questions in recreation models (see von Haefen [1999] and Smith et al. [1993]).
Hedonic property value models require a parallel question to be addressed in linking measures of ambient concentrations of pollution to home sites -- nearest neighbor, kriging, and other techniques are available. However, few have been evaluated in a comparative framework assessing their impact on the results derived from hedonic models. In unpublished research, I participated in an evaluation of their influence on the estimates from hedonic models but know of few comprehensive evaluations of their importance to the models' results (see Sieg et al. [1999],[2000a],[2000b]).
An important issue that arises in using these models has recently been raised by Irwin and Bockstael [2000]. Their concern parallels issues raised in assessing social interactions (Manski [1993]). More specifically, the parallel arises when the marginal effect of a spatially delineated amenity on property values is affected by the average of amenity values in areas around that location. Open space is the example used by these authors, but any site specific attribute whose impact depends on adjoining values would raise the same issue. The value of a wooded lot depends on the neighborhood, a restored historic home on homes around it, the impact of an undesirable land use on other adjoining land uses, etc.
III. Locational Equilibrium Models
Recently Epple and Sieg [1999] have developed a minimum distance estimator, using the conditions for a locational equilibrium of heterogeneous households selecting residential locations among communities differentiated by local public goods and housing prices. Locational equilibrium models predict a stratification of households by income among communities differentiated by public goods.
These models offer a potentially attractive way of introducing spatial delineation into economic models of markets. They do not require one measure to distinguish location, as in the case of monocentric city models. Instead, a vector of attributes of each community can be used to define the features of locations important to economic agents. Unobserved heterogeneity in households' taste for these features, together with the conditions for an equilibrium, provide predictions about the distributions of households that should be observed. With sufficient restrictions on preferences, it is possible to recover estimates of the preference parameters.
These models appear attractive to spatial applications because they avoid using political or distance related distinctions to include spatial features (through the attribute vectors) into a market model. My recent work with Sieg and graduate students has applied the framework to air pollution problems in Southern California (Sieg et al. [1999],[2000a]) and open space in North Carolina (Walsh [2000]).
A number of research questions have emerged from this activity that require more extensive use of spatial methods. A few examples include: (a) the definition of the spatial unit that defines the community or area that serves as a "commodity" that the models assume is selected for exclusive consumption by each household; (b) the impact of spillovers between communities for the equilibrium; (c) the endogeneity of the public goods sought in locational choice; (d) the modeling of land supply (or housing supply) within locational equilibrium framework; and (e) the linking of these economic location models to spatially delineated ecological models of the impact of land use on the functioning of ecosystems.
IV. Spatial Delineation and Experiments
Recently economists have used the spatial attributes of locational choices of households or firms to estimate the effects of rules delineated in space as a result of political decisions. Holmes [1998] used this logic to evaluate differences in right-to-work laws on firms' location choices and Black [1999] to evaluate whether school quality matters to parents.
In the environmental context, hedonic analyses of locally undesirable land uses perform comparable implicit experiments whenever a measure and a sample of housing sales are selected to evaluate the effect of the use (see Farber [1998] for a review of these studies). When these studies are expressed as spatially delineated experiments it seems natural to ask what do we expect of the spatially matched controls to test the hypothesis involved or recover the valuation estimates? I am currently involved in research that uses a new loop road as the source of an "experiment" changing local lands uses, availability of open space amenities, and the choice set available to households for residential decisions. A spatially defined repeat sales framework is what we plan to use to attempt to evaluate how households evaluate these choices (see Smith et al. [2000]).
|
|
REFERENCES
Anselin, Luc, 2000, "Spatial Effects in Econometric Practice in Environmental and Resource Economics", paper to be presented at ASSA Meetings, Regional Economics Applications Laboratory, University of Illinois, November 15.
Black, Sandra E., 1999, "Do Better Schools Matter? Parental Valuation of Elementary Education", Quarterly Journal of Economics, Vol 114 (May): 577-601.
Chay, Kenneth Y. and Michael Greenstone, 2000, "Does Air Quality Matter? Evidence From the Housing Market", Department of Economics, University of Chicago, October.
Ellickson, Bryan, 1971, "Jurisdictional Fragmentation and Residential Choice", American Economics Review Papers and Proceedings, Vol 61 (May): 334-339.
Epple, Dennis and Glenn J. Platt, 1998, "Equilibrium and Local Redistribution in an Urban economy When Households Differ in Both Preferences and Incomes", Journal of Urban Economics, Vol 43 (January): 23-51.
Epple, Dennis and Holger Sieg, 1999, "Estimating Equilibrium Models of Local Jurisdictions", Journal of Political Economy, Vol 107 (4): 645-681.
Farber, Stephen, 1998, "Undesirable Facilities and Property Values: A Summary of Empirical Studies", Ecological Economics, Vol 24 (1): 1-14.
Holmes, Thomas J., 1998, "The Effects of State Policies on the Location of Industry: Evidence from State Borders", Journal of Political Economy, Vol 106 ( ): 667- . .
Irwin, Elena G. and Nancy E. Bockstael, 2000, "The Problem of Identifying Land Use Spillovers: Measuring the Effects of Open Space on Residential Property Values", paper to be presented at 2001 ASSA Meetings, Department of Agricultural, Environmental and Development Economics, Ohio State University, November.
Kneese, Allen V. and Blair T. Bower, 1979 Environmental Quality and Residuals Management (Washington, D.C.: Resources for the Future).
Manski, Charles, 1993, "Identification of Endogenous Social Effects: The Reflection Problem", Review of Economic Studies, Vol 60 (July): 531-542.
Oates, Wallace, Paul Portney and Albert McGartland, 1989, "The Net Benefits of Incentive-Based Regulation: A Case-Study of Environmental Standard Setting", American Economic Review, Vol 79 (5): 1233-1242.
Schwabe, Kurt A., 1996, Source Heterogeneity, Resource Characteristics and the Performance of Markable Permits, unpublished Ph.D. thesis, North Carolina State University.
Schwabe, Kurt A. and V. Kerry Smith, 1999, "Can Incentive Based Policies Be Oversold?", CEnREP working paper, November.
Sieg, Holger, V. Kerry Smith, H. Spencer Banzhaf, and Randall Walsh, 1999, "The Role of Optimizing Behavior in Willingness to Pay Estimates for Air Quality", American Journal of Agricultural Economics, Vol 81 (5): 112-1117.
Sieg, Holger, V. Kerry Smith, H. Spencer Banzhaf and Randy Walsh, 2000a, "Estimating the General Equilibrium Benefits of Large Policy Changes: The Clean Air Act Revisited", NBER Working Paper No 7744, June.
Sieg, Holger, V. Kerry Smith, H. Spencer Banzhaf and Randy Walsh, 2000b, "Using Locational Equilibrium Models to Evaluate Housing Price Indexes", CEnREP working paper, August.
Smith, V. Kerry, Jin Long Lui and Raymond B. Palmquist, 1993, "Marine Pollution and Sport Fishing Quality: Using Poisson Models as Household Production Functions", Economics Letters, Vol 42 (1): 111-116.
Smith, V. Kerry, Christine Poulos, and Hyun Kim, 2000, "Treating Open Space as an Urban Amenity", CEnREP working paper, September.
Smith, V. Kerry, Kurt A. Schwabe and Carol Mansfield, 1999, "Does Nature Limit Environmental Federalism?" in Environmental and Public Economics: Essays in Honor of Wallace E. Oates edited by A. Panagariya, P.R. Portney and R. M. Schwab (Cheltenham, U.K.: Edward Elgar).
Tietenberg, Thomas H., 1985, Emission Trading: An Exercise in Reforming Pollution Policy (Washington, D.C.: Resources for the Future).
von Haefen, Roger, 1998, "The Welfare Implications of Environmental Integrity for Recreation Site Definition" , Department of Economics, Duke University, November.
Walsh, Randy, 2000, "Analyzing Open Space Policies in Locational Equilibrium Model with Endogenous Landscape Amenities", Department of Economics, Duke University, November.
|