A black teacher argues in the NY Times today that black students gain more from having him as their teacher. This "productivity from own group" exposure was one theme of my Civil War work with Dora Costa. In our 2006 paper, we document the costs and benefits of fighting with diverse peers for black civil war solders. In our 2003 paper, we document the costs of fighting in more diverse war companies.
Based on our research (and common sense), people are more comfortable when they are with members of their own group but they are more likely to learn from others when they are with members of different groups. Thus, there is a tradeoff. Without writing down a math problem there is an optimal level of group diversity and cross-group interaction.
Returning to the NY Times piece, I think he is saying that schools in minority residential areas do not hire enough minority teachers? But , is this due to supply or demand? The same issue arises at research universities. USC would love to hire more minority academics. The competition for hiring such individuals who graduate with a Ph.D. from a top 50 institution (as ranked by REPEC) is fierce. Supply appears to the binding constraint. This raises the issue of how different young people choose a career.
The next generation of dynamic programming labor economists should study this perhaps using the following field experiment research design. Suppose I went to the Harvard (and other elite undergraduate programs) Economics department and interviewed all of the current undergraduate economics majors (some of whom are black and Hispanic) and said, I will give you $X dollars per year extra if you complete a PH.D in Economics. As I vary $X, how would the probability that different Econ majors at Harvard choose to get a PhD change? This would be a supply curve. Are minorities more "price sensitive" (and thus elastic in their response) than whites and Asians?