How Trade Secrets Hurt Innovation8 February, 2019 / Articles
Firms that don’t protect their innovation risk losing much of the value they create. That’s why intellectual property protection (IPP) is a key component in the strategy of both established enterprises and early-stage startups.
The first form of IPP that typically comes to mind is patenting to claim ownership. While the strength of the patent system varies across geographies, patent protection is well-established and increasingly homogeneous around the world. But studies show that when asked about their IPP strategies, both managers of large enterprises and entrepreneurs often respond that other tools are more important than patents. In particular, trade secrecy, where companies choose not to disclose information about their inventions, is considered an increasingly important defense strategy and source of competitive advantage.
One of the primary ways that employers seek to protect trade secrets is by having employees sign non-compete contracts and non-disclosure agreements – these aim to prevent employees from moving to competitors and from disclosing valuable information to new employers. But while greater trade secrecy may protect existing innovations, it’s not clear how this might affect future innovation. In a recent study, we examine this issue, and find that strengthening employers’ trade secrecy protection can backfire by dampening inventors’ productivity and hurting innovation in the long run.
At least in the academic literature, the effect of stronger protections for trade secrecy is ambiguous. On the one hand, a more employer-friendly trade secrecy regime – meaning more trade secrecy rights are allocated to employers at the expense of employees – would incentivize firms to invest more in their workers, because it is now harder for those employees to abscond with the firm’s trade secrets. Under this theory, better innovation output is the likely result.
But a lot of innovation research suggests that innovation comes from recombining ideas, from different fields, experiences, and organizations. (For example, Ford’s Model T assembly line technologies were said to have originated from meat-packing plant factories.) To the extent that a more employer-friendly trade secrecy regime limits employee mobility across companies, the opportunities for idea circulation in the economy could be curtailed – thus potentially harming innovation.
Another set of theories also suggests that stronger trade secrecy protections may diminish innovation – but for an entirely different reason. This view says that if it’s harder for employees to switch firms, they have less incentive to drive up their market value by demonstrating their productivity – and this reduced incentive could result in dampened innovation output. In other words, placing the brakes on rewarding innovative effort may end up harming innovation outcomes.
So we have one theory that predicts a stronger trade secrecy regime will bolster innovation and two reasons to believe it will dampen it. To find out what actually happens, we focused on a specific change in the legal environment surrounding trade secrecy protection: the adoption of the inevitable disclosure doctrine (IDD) in some U.S. states and not others, starting in 1994. The IDD allows a company to seek an injunction in court to prohibit a former employee from working for a competitor for a certain period of time, if they can show it would not be possible for the employee to perform her job without inevitably disclosing the company’s trade secret.
A landmark case applying the IDD is PepsiCo, Inc. v. Redmond. In this case, William Redmond, a manager at PepsiCo in the early 1990s, accepted a job at a competing sports drink company, Quaker, in 1994. PepsiCo filed suit in the 7th District Court in Illinois, arguing that Redmond had access to trade secrets related to pricing, distribution, packaging, and marketing, and that he could not perform his new job without inevitably disclosing them. PepsiCo won, and in December 1994 the court enjoined Redmond from taking the new position through May 1995. Soon after this ruling, four other U.S. state courts adopted a pro-IDD stance.
We conducted an empirical analysis to understand how IDD affected the innovative productivity of inventors. We compared the patenting productivity of inventors in states that adopted IDD to the productivity of inventors in states that didn’t adopt IDD. In all, our sample spans the 1976-2003 time period, and analyzes the patenting outcomes of over 350,000 distinct inventors over that time span, for a total of over 2.5 million inventor-year observations.
We found that IDD had a negative effect on innovation, and specifically on innovation quality (which we measured by patent counts weighted by patent-forward citations). Inventors based in states where IDD was enforced innovated less compared to before IDD took hold and also compared to comparable inventors in non-IDD states. These results continue to hold after also including a wide range of controls for differences in inventor, invention, and firm characteristics.
The next question was what might be driving this pattern. We first checked whether this is simply the result of firms changing their IPP strategy – from less patenting to more trade secrecy — without necessarily affecting the innovation they produce. We found that this can only explain part of the result.
We then looked at whether this was due to our other two potential explanations: less idea recombination and weaker inventors’ incentives. Our findings are more consistent with the second story: that an employer-friendly trade secrecy regime is associated with diminished innovation outcomes because inventors’ incentives are dampened. It appears that because IDD lets employers sue inventors for joining a competitor, inventors are discouraged from seeking out other employment opportunities and thus perhaps less focused on strengthening their innovation portfolio.
We find that after an inventor is exposed to a stronger trade secrecy regime, that inventor systematically produces inventions which are more general-purpose (applicable to a wider spectrum of uses and industrial contexts) and thus less subject to the IDD (which primarily covers movement within the inventor’s field). We interpret this to mean that if an individual can’t move to direct competitors, they may reduce effort in their current area of focus (leading to the documented innovation slow-down) and find other areas in which they can signal their innovation quality.
Overall, our study suggests that, while firms lobby for a strengthened trade secrecy environment, this may ultimately backfire in the long run by leading to lower innovation. They may be able to retain their talent in the short run, but those employees may be less innovatively productive. More generally, the findings suggest that the non-patent intellectual property environment in which individuals and organizations operate can influence innovation outcomes.
Along with these managerial implications, our findings also contribute to the policy conversation around IPP. A long-standing debate has focused on the role of regional culture versus that of legal infrastructure in driving innovation. Some believe that places such as Silicon Valley are more innovative because of the culture of risk taking and the presence of so many entrepreneurial startups. Others believe that the legal environment, such as the non-enforcement of non-compete clauses (as in the case of California) might be the more important explanation for regional innovation outcomes.
While not necessarily denying the role of regional culture, our study is consistent with the view that regionally-based legal infrastructure can play an important role in the innovativeness of a region. In particular, policy-makers seeking to protect firms from information leakage or talent loss should consider that this may affect them negatively in terms of innovation performance.