The Costs of Sprawl?
The "anti-sprawl" movement has received much attention in recent years, and has been successful in implementing its "smart growth" policies in some areas. Much of the justification for the current campaign against the low-density (sprawling) urban development that Americans and Western Europeans prefer is based upon assumptions that it is more costly than the more dense development of central cities. A federally financed research project (Costs of Sprawl) concluded that we can no longer afford sprawling development and that failure to force more dense development in the next quarter-century would impose more than $225 billion in additional costs.

Current Urban Planning Assumptions
The urban planning profession generally contends that the following assumptions (called in this paper Current Urban Planning Assumptions) are compelling reasons why greater control should be exercised over land use to fight urban sprawl.

1. Lower spending will be associated with higher population densities.
2. Lower spending will be associated with lower rates of population growth.
3. Lower spending will be associated with older municipalities.

Research to Date
Most of the research on which these assumptions are based is theoretical, projecting standard costs into the future. It makes no attempt to test the actual expenditures of more dense, slower growing, and older municipalities compared to municipalities with the suburban land-use patterns that have developed over the past half-century. The research contained in this paper examines the actual data on municipal expenditures and finds that the Current Urban Planning Assumptions are unreliable and that other factors--principally, variations in employee compensation per capita--explain virtually all of the variation in municipal expenditures.

However, before describing this research, it is important to examine the Costs of Sprawl claims. Although $225 billion in additional costs sounds like a lot (and there are many questions regarding this claim), the cost is actually modest because it is spread over a quarter-century and an average of 115 million households. In fact, in the last 20 years, the average annual increase in local government expenditures in the United States has been 25 times the annual Costs of Sprawl projection.

Econometric Analysis
The source of data for this paper is the United States Bureau of the Census database for 2000. We used this database to conduct an econometric analysis that sought to identify the factors that are most important in explaining the differences in municipal expenditures. Data were available for more than 700 municipalities in the year 2000. We developed three econometric models.

The first, the General Government Model, was used to estimate the impact of factors such as population density, crime rates, and 11 others on municipal expenditures per capita. With respect to the Current Urban Planning Assumptions, no practical relationship was found between municipal expenditures per capita density, population growth rate, or community age. The impact of density on municipal expenditures was found to be statistically significant, but the predicted impact was trivial. Theoretically, if the nation were to reverse 40 years of suburbanization, the annual savings per capita would purchase a dinner for two at a moderately priced restaurant.

Further, the combination of factors that seemed likely to affect municipal spending (both those related to the Current Urban Planning Assumptions and others) explained less than 30 percent of the variation in municipal expenditures per capita. The other two econometric models showed that none of the Current Urban Planning Assumptions bore a statistically significant relationship to the variation in municipal wastewater charges or water charges. This is particularly significant, since these infrastructure functions are among those cited most often in claims that suburbanization imposes additional costs.

Nominal Analysis
A nominal (ranking) analysis of the actual data was also performed. The actual data indicate relationships considerably at variance with the Current Urban Planning Assumptions. The highest density, slowest growing, and oldest municipalities all had higher-than-average expenditures per capita. The oldest municipalities had the highest expenditures.

Employee Compensation
By far the largest expenditure category for municipalities is employee compensation. A further nominal analysis indicated that virtually all of the variation in municipal expenditures per capita could be explained by the variation in employee compensation. For example, the highest density quintile of municipalities spent $68 per capita each year more than the average. Wages and salaries in the same municipalities were $91 higher.

Special Interest Control and Entrenchment?
In short, this analysis indicates that higher payroll costs are associated with larger, older municipalities. Local government employees have a significant, concentrated interest in improving their compensation and working conditions. This could be indicative of a political "entrenchment" that results from special interest control--an influence to which older municipalities would be more susceptible. Other special interests could exert similar influence, although employee compensation alone appears sufficient to account for the variation in municipal spending. It seems much more likely that the differences in municipal expenditures per capita are the result of political, rather than economic, factors--especially the influence of special interests.

Wendell Cox, Principal of the Wendell Cox Consultancy in metropolitan St. Louis, is a Visiting Fellow at The Heritage Foundation and a Visiting Professor at the Conservatoire National des Arts et Metiers in Paris. Joshua Utt is a Ph.D. candidate in Economics at Washington State University and an Adjunct Fellow at the Discovery Institute in Seattle, Washington.