There exists large and persistent variation in not only how, but when employees are paid, a fact unexplained by existing theory. This paper develops a simple model of optimal pay timing for firms. When workers have self-control problems, they under-save and experience volatile consumption between paychecks. Thus, pay whose delivery matches the timing of workers' consumption needs will reduce wage costs. The model also explains why pay timing should be regulated (as it is in practice): although the worker benefits from a timing profile that smoothes her consumption, her lack of self-control induces her to attempt to undo the arrangement, either by renegotiating with her employer or by taking out payday loans. Regulation of pay timing and consumer borrowing is required to counter these efforts, helping the worker help herself.

We analyze a reduced-form framework for understanding the equity loan market's impact on share prices. We show that hard-to-borrow stocks will have distinct return patterns, responding more to shocks in the supply of shares available, and to changes in the heterogeneity of investor beliefs, than other stocks. We conduct two empirical tests in which we find strong support for these equilibrium predictions. In our first test, we take advantage of a tax-driven exogenous shock to share loan supply and find that when supply is reduced around dividend record dates, prices of hard-to-borrow stocks increase 1.1% while prices of easy-to-borrow stocks are unaffected. In our second test, we find that hard-to-borrow stocks have 4.8% lower three-month returns than other stocks, with negative returns concentrated in stocks with high heterogeneity in investor beliefs. Thus, we extend the Diether, Malloy, and Scherbina (2002) result that stocks with a greater dispersion of investor beliefs have lower returns.