We examine the economic determinants of short-sale supply, and its consequences for future stock returns. Lendable supply increases with expected borrowing costs and decreases with financial statement constructs that indicate overvaluation. Although rising loan fees help ease supply constraints, we find shares are still least available when they are most attractive to short sellers. Using a number of firm characteristics, we derive useful instruments for real-time loan supply and demand conditions in the lending market. Further, we show that (1) when lendable supply is binding (non-binding), short-sale supply (demand) is the main predictor of future stock returns, (2) abnormal returns to the short-side of nine well-known market anomalies are attributable solely to “special” stocks, and (3) loan fees significantly reduce the profitability of the short side and several of these anomalies cease to be profitable. Overall our evidence highlights the central role played by the supply of lendable shares in equity price formation and returns prediction.

We investigate whether missing R&D expenditures in financial statements indicates a lack of innovation activity. Patent records reveal that 10.5% of missing R&D firms file and receive patents, which is 14 times greater than zero R&D firms. Pseudo-Blank R&D firms (missing R&D firms with patent activity) demonstrate patent filings analogous to the bottom 90–95% of the positive R&D population. Multivariate difference-in-differences tests indicate that Pseudo-Blank R&D firms are more likely to report R&D after an exogenous auditor change. Finally, we provide simple Monte Carlo simulations to evaluate different methods to handle missing R&D in empirical research.