Introduction. I. Self-Employment: Escape From Racism? 1. The Status of Self-Employed and Employee African Americans in the New York City Construction Industry; T. Bates, D. Howell. 2. Black Employment Criminal Activity and Entrepreneurship: A Case Study of New Jersey; S. Myers, Jr., W.E. Spriggs. II: Racial Wage Inequality and Discrimination. 3. Measuring Wage Discrimination During Periods of Growing Overall Wage Inequality; W.M. Rodgers, III. III: Health. 4. Will Greater Competition Improve the Markets for Health Care Services and the Health Outcomes for Black Americans; W. Leigh. 5. The Continuing Significance of Race in Meeting the Health Care Needs of Black Elderly; S. White-Means. IV: Race Crime and the Neighborhood. 6. Race and Crime: What is the Connection; K. Gyimah-Brempong. 7. Why Does Race Matter in Housing and Credit Markets? G. Dymski. Contributing Authors. Index.

THE JANUS-FACE OF RACE: REFLEC- TIONS ON ECONOMIC THEORY Patrick L. Mason and Rhonda Williams Many economists are willing to accept that race is a significant factor in US eco­ nomic and social affairs. Yet the professional literature displays a peculiar schizo­ phrenia when faced with the task of actually formulating what race means and how race works in our political economy. On the one hand, race matters when the dis­ cussion is focused on anti-social behavior, social choices, and undesired market outcomes. Inexplicably, African Americans are more likely to prefer welfare, lower labor force participation, and unemployment. On the other hand, race does not matter when the subject of discussion is economically productive or socially accept­ able activities and legal market choices (for example, wages and employment). This Janus-faced construction of race is maintained by economists' stubborn ad­ herence to the market power hypothesis. The market power hypothesis asserts that racial discrimination and market competition are inversely correlated. Discrimina­ tory behavior will persist only in those sectors of society where the competitive forces of the market are least operative. When applied to the labor market, the mar­ ket power hypothesis suggests that pre- and post-labor market decisions represent disjoint sets. On average, members of a disadvantaged social group may accumulate a lower amount of or a lower quality of productive attributes because of discrimina­ tion in marital, residential, or school choice, or because of substantial animosity in day-to-day interpersonal relations with members of a privileged group.