Empirically – trademark protection reduces R&D spending and patents
Davidson Heath and Christopher Mace, 2020 – professors of business at the University of Utah. “The Strategic Effects of Trademark Protection” The Review of Financial Studies 33 (2020) 1848–1877, Oxford University Press, accessed via University of Michigan //DH FTDA = Federal Trademark Dilution Act of 1996, “treated firms” means firms that were affected by the FTDA’s protection
Next we investigate the relationship between trademark protection and innovation. A large theoretical literature predicts that the incentive to innovate depends on firms' market power, specifically the difference between expected preinnovation and postinnovation rents (Arrow 1962; Aghion and Howitt 1992; Aghion et al. 2009). Both the quiet life hypothesis (Bertrand and Mullainathan 2003) and the Chamberlin (1933) rent extraction hypothesis suggest that stronger trademark protection leads to more managerial slack or more preinnovation rents, and thus to less innovation. Alternatively, the higher profits we document might have relaxed financial constraints and been at least partly plowed back into R&D.
Table 4 investigates several measures of firms' innovative activities. Column 1 shows that treated firms reduced R&D spending over assets by 0.32 pp following the FTDA. This is a significant change relative to their pretreatment mean of 3.4%. Column 2 shows that treated firms took out 9% fewer patents post-FTDA. When we weight patents by the adjusted citations they subsequently received, the reduction is greater at —14% (Column 3). However, the estimated dollar value of granted patents for treated firms (from Kogan et al. 2017) actually increased post-treatment by 13% (Column 4). The asymmetry of the citation- and dollar-weighted estimates is striking, and consistent with the FTDA leading treated firms to pursue a less exploratory and more exploitative innovation strategy.
Apart from R&D and patents, a common measure of product-level innovation is announcements of major new product lines gathered from newspapers or trade magazines (Chaney, Devinney, and Winer 1991; Mishra and Bhabra 2001). Table 4 Column 5 shows that treated firms were 13% less likely to announce a major new product in the post-treatment period, with the number of new products announced showing a 24% decrease (Column 6). Taken together, these results appear consistent with both the quiet life and trademarks-foster- monopoly hypotheses, or a model in which the FTDA discouraged innovation by raising preinnovation rents.
Figure 4 shows graphs of the year by year coefficients for the main innovation outcomes, R&D spending and patents. For brevity the graphs span 1982-2005 to include the TLRA, FTDA, and the Moseley decision. Panel A shows that treated firms' R&D spending was unchanged around the TLRA placebo event, fell after the passage of the FTDA, and reversed to pretreatment levels following the Moseley decision. Panel B shows that new patent grants were unchanged around the TLRA placebo event, fell after the passage of the FTDA, and leveled off following the Moseley decision. These year-by-year coefficient estimates are consistent with the estimates in Table 4 and with our hypotheses that the following the FTDA, treated firms moved from an innovative toward an exploitative product market strategy.