The supply and demand of motivated labor: When should we expect to see nonprofit wage gaps? ¬リニ

• Nonprofit wage gaps emerge in markets with low nonprofit labor demand. • Wages are equal across sectors when nonprofit labor demand is high. • Nonprofit productivity is highest when nonprofit wages are lowest. • This suggests sorting & screening of scarce motivated workers generate wage gaps. a b s t r a c t a r t i c l e i n f o Evidence on whether nonprofit workers earn less than for-profit workers is mixed. I argue that we should only expect wage gaps when labor demand of the nonprofit sector of an industry is low. When labor demand is high, there are not enough " motivated " workers to fulfill demand, so nonprofits must raise wages. I find empirical evidence consistent with these predictions. Penalties for working in a nonprofit are largest in areas where nonprofits require a small share of the labor force. In these same locations, the quality of work is higher than in for-profits. A long literature examines differences in wages across the for-profit and nonprofit sectors. Some empirical work finds clear evidence that nonprofit workers earn less. The most prominent explanation for this finding is the " labor donation hypothesis, " which suggests that some individuals enjoy nonmonetary benefits from working in a nonprofit and as such are willing to work for less. 1 An alternative explanation is that nonprofits tend to locate in lower-paying industries, so composition effects rather than differential preferences of workers may drive the observation of a nonprofit wage differential. However, as demonstrated by Leete (2001) even after carefully accounting for industry and occupation , a nonprofit wage differential exists in some industries but not others — a puzzling result for either of these two prominent explanations. I consider an explanation for the emergence of wage differentials in some settings but not others and, in doing so, revisit the mechanism through which these wage differentials arise more generally. Building on a point made by Preston (1989), I test the hypothesis that a nonprofit wage differential should exist when the share of labor demanded by nonprofits is low relative to for-profits. I start from the assumption that some workers are in fact willing to donate their labor and draw from a recent theoretical literature examining the impact of motivation on labor market outcomes