Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Previous studies estimate that black quit rates are lower than those of whites. This paper suggests that these estimates under-state black quit propensities because they neglect racial differences in quit responses to commuting time and local unemployment rates. Ignoring these differences, the black quit effect appears to be negative. Controlling for them, the residual race effect is positive and sufficiently large to account for all net black quits.