Does Gentrification Harm the Poor?

FOR SEVERAL DECADES, social scientists have tracked the fiscal health of American central cities with some degree of concern. Suburbanization, spawned by technological innovations, consumer preferences, and at least to some extent by government policy, has selectively pulled affluent households out of urban jurisdictions. ¹ The leaders of these jurisdictions are left with the prospect of satisfying more concentrated demands for services with a dwindling tax base, realizing that further increasing the burden they place on residents will simply drive more of them away. In the process, cities have become concentrated centers of poverty, joblessness, crime, and other social pathologies. ²

A detached observer of this cycle might conclude that its reversal—the return of affluent households to the central city, associated increases in property values and the government's tax base—would be welcomed by city leaders. In fact, the response to such a turnaround, if labeled as "gentrification," is quite likely to be negative. In San Francisco, for example, a mayoral candidate in 1999 pledged to declare "war on any and all gentrification" if elected. ³ Though usually associated with larger and faster-growing cities, gentrification and its associated tensions have been noted in cities from Milwaukee to Baton Rouge. ⁴ Scholarly interest in gentrification peaked in the first half of the 1980s, as neighborhood revitalization occurred in a number of U.S. cities. In recent years, as reports of gentrification across the country have accelerated, there are signs of renewed scholarly interest in the process. A number of studies, past and present, have debated whether the process is significant and/or sustainable. ⁵ Still others have addressed the issues that most frequently concern political observers of gentrification: that the process imposes costs on disadvantaged households. ⁶ Does gentrification cause a reduction in well-being among disadvantaged households? This paper examines the question and finds it extraordinarily difficult to answer, for reasons that few previous authors have considered.

The first section begins the examination by illustrating the demographic shifts most commonly associated with gentrification, and then offers two competing explanations for them. The competing explanations motivate two very different views of the distributional effects of gentrification. In the first view, revitalization of urban neighborhoods causes changes in well-being among disadvantaged households. In the second view, gentrification is merely a side effect of other broad economic trends that affect the poor. The analysis also makes clear that residential displacement—the primary focus of most existing literature on the consequences of gentrification—is neither a necessary nor sufficient condition for declines in the living standards of poor households.

The following section considers the general equilibrium effects of gentrification beyond the housing market. Increases in the local tax base might improve the quality of local public goods and services. Employment opportunities in certain industries might improve with the arrival of a more affluent clientele; that is to say, gentrification might partially solve the urban "spatial mismatch" problem. ⁷ Finally, gentrification might decrease the urban concentration of poverty, ameliorating the ills associated with it. ⁸ The subsequent section reviews the literature on the distributional impact of gentrification, and concludes that previous studies are too narrowly focused to fully address the question of whether gentrification harms the poor.

The literature review is followed by a broader analysis of the question, using data from the American Housing Survey (AHS) to consider gentrification in the Boston area between 1970 and 1998. Overall, the data point to no obvious conclusion, which is not surprising considering the difficulty of the task. The greatest empirical difficulty in assessing gentrification is determining what would have happened to individuals had gentrification not occurred. Notwithstanding these difficulties, the empirical work here presents some striking patterns. There is no evidence to suggest that gentrification increases the probability that low-status households exit their housing unit. Poor households are more likely to exit poverty themselves than to be replaced by a nonpoor household. Nonetheless...