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.
KOH, P.S. et REEB, D. (2015). Missing R&D. Journal of Accounting and Economics, 60(1), pp. 73-94.