The dramatic reduction in the black–white earnings gap from 1965 to 1975 represents the most significant period of economic progress for African-Americans in the post World War II era. After 25 years of contentious research, economists have arrived at a consensus that Title VII of the 1964 Civil Rights Act, which outlawed employment discrimination, was a major factor underlying the post-1965 trend break in relative earnings (John J. Donahue and James J. Heckman, 1991; David Card and Alan Krueger, 1993). A key piece of supporting evidence is that most gains were concentrated among black workers in the South, where Title VII had its biggest impact. The mortality rate of black infants relative to white infants is another clearly important measure of the relative well-being of AfricanAmericans. Surprisingly, and in contrast to the relative-earnings literature, there are very few studies that focus on long-run trends in the relative health of black infants over time. In addition, the existing research is based on highly aggregated data, both across regions and over time, which provide little information on the precise location and timing of infant mortality changes by race. Consequently, the evidence on the specific factors underlying significant changes in black–white infant health outcomes is sparse and often anecdotal. An examination of infant death rates within a year of birth for whites and nonwhites and the nonwhite–white infant-mortality-rate (IMR) ratio from 1933–1990 in the United States reveals many striking patterns (figure available from authors upon request). Immediately before World War II, about 4.1 percent of white infants and 7.5 percent of black infants died within a year of birth. During World War II, there was a large decline in black infant mortality rates and in the black–white ratio. Since World War II there has been a secular increase in the black– white IMR ratio with one notable exception. In the narrow period of 1965–1970, the black IMR and the black–white ratio declined sharply relative to preexisting trends. Relatively stable during 1961–1965, the black IMR fell 30 percent from 4.0 (per 100 live births) in 1965 to 2.8 in 1971. At the same time, the black–white ratio fell from 1.9 to 1.65, the only prolonged convergence in the post-World War II era. The national patterns suggest that 1965–1970 is the key period for improvements in the relative health of black infants over the past 50 years. This study examines trends in black– white rates of infant death during 1955–1975. To document the location of the improvements, we collected data by race at the state and rural– urban levels, which has not been previously done. Using simple descriptive models, we find † Discussants: David Meltzer, University of Chicago and NBER; Kenneth Chay, University of California–Berkeley and NBER; Jeffrey Grogger, University of California–Los Angeles and NBER; Dan Black, Syracuse University.

We exploit the structure of the Clean Air Act to provide new evidence on the capitalization of total suspended particulates (TSPs) air pollution into housing values. This legislation imposes strict regulations on polluters in “nonattainment” counties, which are defined by concentrations of TSPs that exceed a federally set ceiling. TSPs nonattainment status is associated with large reductions in TSPs pollution and increases in county‐level housing prices. When nonattainment status is used as an instrumental variable for TSPs, we find that the elasticity of housing values with respect to particulates concentrations ranges from −0.20 to −0.35. These estimates of the average marginal willingness to pay for clean air are robust to quasi‐experimental regression discontinuity and matching specification tests. Further, they are far less sensitive to model specification than cross‐sectional and fixed‐effects estimates, which occasionally have the “perverse” sign. We also find modest evidence that the marginal benefit of reductions of TSPs is lower in communities with relatively high pollution levels, which is consistent with preference‐based sorting. Overall, the improvements in air quality induced by the mid‐1970s TSPs nonattainment designation are associated with a $45 billion aggregate increase in housing values in nonattainment counties between 1970 and 1980.

Many programs reward or penalize schools based on students' average performance. Mean reversion is a potentially serious hindrance to the evaluation of such interventions. Chile's 900 Schools Program (P-900) allocated resources based on cutoffs in schools' mean test scores. This paper shows that transitory noise in average scores and mean reversion lead conventional estimation approaches to overstate the impacts of such programs. It further shows how a regressiondiscontinuity design can be used to control for reversion biases. It concludes that P-900 had significant effects on test score gains, albeit much smaller than is widely believed

In the 1980s, many US cities initiated programs reserving a proportion of government contracts for minority-owned businesses. The staggered introduction of these set-aside programs is used to estimate their impacts on the self-employment and employment rates of African American men. Black business ownership rates increased significantly after program initiation, with the black-white gap falling 3 percentage points. The evidence that the racial gap in employment also fell is less clear as it depends on assumptions about the continuation of preexisting trends. The black gains were concentrated in industries heavily affected by set-asides, and they mostly benefited the better educated.