The surprisingly weak justifications for employee non-competition agreements
posted by Viva Moffat
As I said in my first post, I am working on an article about employee non-competition agreements. As you might infer from the title of this post — * spoiler alert * — I don’t think much of them. When I practiced law, employee non-competes nearly always struck me as unfair and often unnecessary (and I worked on the both the employee and the employer side). Many of the employee-side arguments concerning the problems with non-competition agreements cover fairly well-trod ground: restrictions on employee mobility, unequal bargaining power, little if any negotiation of terms, nominal consideration, difficulties of enforcement, and so on.
What has really struck me in working on the piece is the weakness of the arguments put forth to justify the imposition of non-competition agreements. Those arguments fall roughly into three categories: (1) freedom of contract; (2) general references to “business necessity” or some variation on that; (3) the need to protect intellectual property or IP-like assets, such as trade secrets or “confidential information.” To my mind, each of these justifications is problematic.
The first is problematic for many of the reasons put forth in the “employee-side” arguments mentioned above. “Freedom of contract” is fairly illusory when, for example, the non-compete is presented to the employee after two years of employment and the consideration is one day’s vacation. In addition, non-competes — by their very name — are a form of intervention in the market, putting them at odds to some extent with the freedom of contract principle that is animated by a free market ethos.
The second justification — “business necessity” — is problematic in part because it is so rarely suppported with evidence of the nature of that business necessity. Because the vast majority of non-competes are not challenged, the incentive for employers to carefully craft a non-compete to limit post-employment competition only enough to protect true ”business necessities” or “legitimate interests” (as is often the standard) is minimal. In addition, there is some evidence that, in fact, businesses don’t need non-competes. Overall, firms in states in which non-competes are unenforceable (most notably, California), have not suffered much, if any, disadvantage from the legal rule. There may be some advantages to individual firms in imposing non-competes, but it appears that the overall performance of a region or of an industrial sector is, in fact, hampered by a rule permitting non-compete enforcement. Some observers have concluded that the rule rendering non-competes unenforceable causes “high-velocity” employment, which contributes to knowledge spillovers and consequently greater economic growth and innovation.
The final set of arguments revolves around IP-like justifications: the need to protect trade secrets and other intangible confidential information; and the need for an incentive for firms to invest in the creation and disclosure of information. I argue in the article that these justifications are misplaced. For a variety of reasons, non-competes are simply not a good tool for protecting IP-like assets, and a rule rendering non-competes unenforceable will more properly channel protection to the IP regimes.
January 26, 2010 at 7:56 pm
Posted in: Employment Law, Intellectual Property
Print This Post
Responses (7)
Lawrence Cunningham - January 26, 2010 at 9:06 pm
Very interesting post, which makes me look forward to reading the paper. Two preliminary queries:
1. does the assessment change when the non-compete arises in the context of a sale of business or dissolution of a partnership (exceptions recognized even in states like California)?
2. is the argument concerning recouping an employer’s investment in training and developing an employee different from “business necessity” or subsumed by it and equally problematic?
Ken Rhodes - January 26, 2010 at 10:57 pm
I agree with Professor Cunningham that it sounds interesting, and I too will look forward to reading it. In re Professor Cunningham’s two queries, I would guess:
1. The non-competes in the two special situations mentioned are probably outside the scope of the discussion here, since they are not “employee non-compete” agreements, but rather are agreements made in exchange for payment, specifically by a person becoming a non-employee. When my partner and I sold our software consulting company to a much larger “friendly competitor,” they paid us quite nicely to agree not to compete with them for three years. It would have surprised me to have had it any other way. The non-compete did not take away my ability to work in my field; merely to compete with that company.
2. My guess is that investment in the employee is the weakest argument of all. In my field, the most important justification has always been “trade secrets.” Most specifically, not how the work is done, but rather, how it is sold. A competent engineer can usually figure out how another engineer’s products work, or at least how to develop a reasonable substitute, but figuring out how to sell those products to clients is a craft highly prized, and highly guarded.
A.J. Sutter - January 27, 2010 at 2:27 am
Apropos of Ken’s “agreements made in exchange for payment, specifically by a person becoming a non-employee”: the actual justification in California is closer to that he or she becomes a non-owner (including through certain asset sales) (Business & Professions Code secs. 16600-16602.5). Otherwise, severance packages could be made contingent on non-competes. That would create an awful dilemma for someone who’s getting laid off.
I agree about the weakness of the “investment in the employee” argument; moreover, it’s a two-edged sword. Without non-competes, companies benefit from being able quickly to hire people who’ve already been trained elsewhere.
Even in California, though, ex-employers try to play the trade secret card. Litigation, or the threat of it, can lead to a settlement limiting the scope of the new hire’s duties for some period of time. I seem to recall that when I was in the chip equipment industry, this situation also came up when we hired a guy as a sales exec from one of our biggest customers: they didn’t want him handling other companies’ accounts (i.e., the ex-employer’s competitors) until after 6 or maybe 12 months. It had nothing to do with a contractual non-compete. (We capitulated, heeding one of our in-house mantras, “It’s not nice to sue your customer.”)
John Steele - January 27, 2010 at 5:39 am
I look forward to reading the article. When I transitioned from doing antitrust law to doing IP law in California, I had to look at non-competes from a different angle. I had recalled the arguments that non-competes were necessary for economic investment and development, but, as most people know, in Silicon Valley, the capital of innovation and venture capital in the US, the law rarely enforces non-competes. So the standard argument of necessity has to be over-stated.
By the time a techie or engineer really hits his/her stride, they may have worked at several companies in the industry. The human capital is deep and rich, and it flows to the best opportunities.
Michael Risch - January 27, 2010 at 5:59 am
I too look forward to the article. I tend to agree with all of your criticisms, though I think the IP angle is more nuanced than you let on here based on transactions costs. The argument is that non-compete agreements (shorter, fair ones, at least) are much cheaper to enforce than trade secrets.
Trade secrets are difficult to discern, and proving misappropriation is costly. Defending is also costly. Nonetheless, such cases are brought. I discuss here: http://papers.ssrn.com/abstract=1269039 the potential that more cases are brought where there is weaker non-compete.
The problem is that such cases have a chilling effect, and in fact can chill movement for even longer than a summarily enforced but reasonable time non-compete. If that’s true, then non-competes might provide a social benefit over trade secret misappropriation through reduced litigation and shorter chilling.
While I don’t think the above argument carries the day, I think it is not easily discarded.
TJ - January 27, 2010 at 6:39 am
Not having read the article, this thought is rather preliminary. But your dismissal of “freedom of contract” seems too summary. Not many people argue for freedom of contract for its own sake, nor is it entirely fair to your opponents to characterize them as ignoring disparities of market power when they don’t. Rather, the argument comes down to that if you try to intervene in such contracts, you inevitably create inefficiencies and distortions.
In the non-compete context, employers really like non-competes. So if you make them legally unenforceable, they will turn to extra-legal regimes that can replicate the effect, for example by hiring their cousin instead of some more competent person. Now the extent of this effect is an empirical question, and awfully difficult to study because we are less concerned about large businesses here than smaller businesses where such substitutes are easier to implement. I don’t know whether the data bears out the theory, but I don’t think anybody else does, either.
A.J. Sutter - January 27, 2010 at 7:14 am
TJ, at the Silicon Valley companies where I worked, no cousins were hired. And the level of competence was breath-taking. As John Steele rightly observes, what makes the Valley the Valley is the ability to move freely between jobs — and not only do individuals like, it, companies like being able to hire freely (with the occasional hitch, as I noted above). Start-ups benefit from this at least as much as large companies. Perhaps you’ve been reading too much economics. There aren’t any markets that are free from “inefficiencies and distortions” — eppur si muove.
Leave a Reply