Employment Mobility Laws and Competition

Article excerpt

I. INTRODUCTION

Herbert Hovenkamp discusses the relationship between antitrust laws and intellectual property (IP) laws. He points out that the conflict between the two "is explained by deep uncertainty about the optimal amount and scope of IP protection. As long as that uncertainty remains there will always be tension between IP and antitrust."1

In this Article we study a particular sort of IP protection, laws that help firms to impose barriers on employee mobility, such as covenants not to compete (CNC). Our goal is to analyze, in a simple economic model, the role that the scope of these IP laws have on the degree of competition. We see this analysis as a necessary step toward understanding the complicated interactions underlying the optimal scope of IP protection.

In practice, the degree of protection afforded employers suggests that the issue is an important one. States differ significantly in their policies toward the intellectual property of employers, and have seen different histories as a result. In Massachusetts, where the legal system is based on the English system, employment contracts with CNCs are enforceable by the courts. In California, the law, written in the mid-1800s, is codified and does not specifically allow for the enforcement of such contracts.2 There are some states, such as Texas, where CNCs are typically not enforced.3 The differential enforcement of these clauses has been constant for more than a century in Massachusetts and California, and the results are strikingly different.4

There is evidence of hyper-mobility between high-tech firms in California, which is not evident in Massachusetts.5 In the hard drive industry, more than a quarter of the startups in California were spin-outs, firms started by former employees of incumbent firms, while almost all of the start-ups in Massachusetts were de novo firms.6 In addition, the number of firms in Massachusetts in the electronics industry, and more specifically in the hard drive industry, was initially higher. By the mid-1980s, however, the number of firms in California was significantly higher and continued to grow, while the number in Massachusetts was in significant decline.7

We are motivated to study this form of IP law because of the great uncertainty about the optimal scope of such regulations. As a first step toward understanding optimal policy, we seek to understand these laws' implications for competition. History suggests that these policies clearly affect the number of firms in an industry, and therefore are relevant to an antitrust authority concerned with the level of competition. There are at least two ways that these rules might affect the number of firms. On the one hand, to the extent that these policies reduce start-ups by employees, they thwart competition. On the other hand, to the extent that they protect IP, they encourage firms to enter with new ideas when the protection is better.

The purpose of this paper is to put forward a simple model of the two effects, in order to determine under what conditions one or the other might dominate. The model is intended as a starting point for discussing the role of these policies on the level of competition in industries where employee turnover is an important source of new firms. As such, our intent is to provide a guide by including a few key economic ingredients necessary in the construction of such a model.

Our results suggest that, in this case, IP protection and competition need not be in conflict. In fact, in some cases, more competition results from greater IP protection, since it encourages entry by entrepreneurs with new ideas. We find this to be the case when either additional competition reduces profits greatly, or when additional competition has nearly no effect on a firm's profits. The conflict between IP protection and antitrust is most likely to arise in the intermediate cases. There, strong IP protection for employers will tend to foster fewer firms in the long run. …