This paper models how migration both influences and responds to differences in disease prevalence between cities and shows how the possibility of migration away from high-prevalence areas affects long-run steady state disease prevalence. We develop a dynamic framework where migration responds to the prevalence of disease, to the costs of migration and to the costs of living. The model explores how pressure for migration in response to differing equilibrium levels of disease prevalence generates differences in city characteristics such as land rents. Competition for scarce housing in low-prevalence areas can create segregation, with disease concentrated in high-prevalence “sinks”. We show that policies affecting migration costs affect the steady-state disease prevalences across cities. In particular, migration can reduce steady-state disease incidence in low-prevalence areas while having no impact on prevalence in high-prevalence areas. This suggests that, in some circumstances, public health measures may need to avoid discouraging migration away from high-disease areas.

This paper produces new estimates for white and black mortality and fertility at the state level from 1800(20)–2000. It produces new estimates of black and white schooling for this same period. Using a calibrated model of black and white parents, we fit the time series of black and white fertility and schooling. We then produce estimates of the benefits of equal education opportunity for blacks over the period 1820–2000. For the better part of US history, blacks have suffered from less access to schooling for their children than whites. This paper quantifies the magnitude of this discrimination. Our estimates of the welfare cost of this discrimination prior to the Civil War range between 0.5 and 20 times black wealth, and between 0.5 and 10 times black wealth prior to 1960. Further we find that the Civil Rights era was valued by blacks in the South by between 1% to 2% of wealth. Outside of the South, we find significant costs of discrimination prior to 1960, ranging from 6% to 150% of black wealth. For these divisions from 1960–2000, blacks have attained rough parity in schooling access. The welfare magnitudes are similar to the hypothetical gains to blacks if they had white mortality rates. We show that the model’s black and white human capital series are strongly, positively correlated with state output measures, black and white permanent incomes and black and white earnings.

Stochastic dominance (SD) is commonly used to rank income distributions and assess social policies. The literature argues that SD is a robust criterion for policy evaluation because it requires minimal knowledge of the social welfare function. We argue that, on the contrary, SD is not a robust criterion. We do this by carefully introducing microfoundations into a model by Chu and Koo (1990) who use SD to provide support to family-planning programs aiming at reducing the fertility of the poor. We show that fertility restrictions are generally detrimental for both individual and social welfare in spite of the fact that SD holds. Our findings are an application of the Lucas’ Critique.