Abstract
In a field experiment, we find that a mandatory mentorship program raises worker productivity, whereas a voluntary version of the program does not. A significant reason why the mandatory program results in larger gains is that the lowest-productivity employees do not participate when the program is voluntary despite their having the greatest treatment benefits. A nationally representative survey of U.S. workers shows wide variation in human capital development program participation, suggesting that understanding self-selection is important for firms’ returns on these programs across a variety of settings. Our findings have implications for resource allocation, experimental design, productivity dispersion, and inequality. This paper was accepted by Maria Guadalupe, business strategy. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.07524 .

This publication has 87 references indexed in Scilit: