Comments

Jacob Vigdor: Unfortunately, race riots are not a thing of the past. From Benton Harbor, Michigan, to Sydney, Australia, violent and destructive outbreaks in minority neighborhoods continue to make headlines in the twenty-first century. As William J. Collins and Robert A. Margo note, a considerable amount of social science research has sought to determine the causes of riots, perhaps in the hope that public policy offers some remedy that can prevent future events. Any such remedy, whether in the guise of governmental reform or redistributive policy, is likely to impose significant costs on society. To determine whether these costs are worth bearing, a complete accounting of the social costs of riots is indispensable. Collins and Margo make an important point: that the social costs of riots are likely to extend far beyond the bricks-and-mortar costs of rebuilding and revitalizing individual neighborhoods. As the authors' analysis and my comments should make clear, the costs identified in this paper are likely to represent only a fraction of the total. As Collins and Margo note, however, this work is but the beginning of an extensive, well-planned, and important research agenda.

Riots, by altering the risk perceptions of firms and households, should make certain neighborhoods less attractive places to live and work. Collins and Margo note the labor demand impact of this change—indeed, it serves as the central motivation for this study—but there are several other noteworthy implications. Demand for residential location in high-risk neighborhoods should decline as well: property destruction is presumably a disamenity to both firms and consumers. It is thus a straightforward prediction that rents will decline in riot-prone neighborhoods.⁵⁵ Owners of property will experience negative wealth effects; renters will witness a reduction in housing costs to offset some or all (or even more than all) of the utility loss associated with increased risk.