"Embedded Knowledge" and the Trade Secret Gap

For anyone (lawyer or otherwise) interested in the tensions associated with employee mobility, a truly must-read is Talent Wants to Be Free, by Orly Lobel.

Professor Lobel discusses a wide range of issues associated with talent and knowledge flows, and she incorporates thoughts that transcend the law and devolve into economics and sociology. The book is, to say the least, thought-provoking even for those who may choose to disagree with some of her conclusions. There's a particular issue that Professor Lobel discusses that touches upon one of competition law's great tensions.

In my judgment, one of the most difficult areas of the law for lawyers, judges, and clients is drawing a distinction between two concepts: general skill and knowledge (which is not protectable) and a trade secret (which is). Consider the following from an Illinois decision some twenty years ago:

"General skills and knowledge acquired in the course of employment...are things an employee is free to take and to use in later pursuits, especially if they do not take the form of written records, compilations or analyses." Passage quoted from Colson Co. v. Wittel, 569 N.E.2d 1082, 1087 (4th Dist. 1991).

This is somewhat (largely?) unhelpful for two reasons. First, it's so obvious that we shouldn't feel the need to use it as a guidepost or legal marker. Second, the notion of "general skills and knowledge" suggests a disconnect from experiential information or data the employee may have obtained by reason of his or her employment with a business. Put another way, GSK sounds like information an employee may acquire just by virtue of being in a particular field and that generally may be available to anyone, but not of any real use to people who are engaged in other occupations.

This disconnect is troubling when trying to parse out what kinds of information a business can protect by contract or through assertion of a trade secret right. A passage in Chapter 4 of Professor Lobel's book gets closer to solving the disconnect and bridging the ideas of general skill and knowledge, on the one hand, and trade secrets, on the other.

She talks of "embedded knowledge." Here's how she puts it in context:

"Knowledge that is embedded comes from experience, from learning by doing or observing; this kind of knowledge is difficult to codify and write down. Embedded knowledge (also known as tacit knowledge) is learned informally through direct and repeated contacts."

This formulation actually gets us closer to determining what is protectable and what is not, but only if we say embedded knowledge is something only an enforceable non-compete can safeguard. In other words, we should be careful not to lump in embedded knowledge with the amorphous (some might say rudderless) definition of a trade secret.

Examples of embedded knowledge likely include a person's familiarity with key contacts and customer requirements or buying habits. It also may include the odd concept of "negative know-how" (that which isn't effective). It certainly includes pricing patterns or strategies. Perhaps, too, it encompasses knowledge of which particular vendors are reliable or can lead to efficient sales distribution. To be sure, this will be company-specific data; they're not "general" in the fair sense of the word.

Professor Lobel also makes the point that "embedded knowledge" doesn't spread with great accuracy. Again, from Chapter 4:

"As you can imagine by now, the way tacit knowledge spreads depends on the shape of the network and the complexity of the information being diffused. When knowedge is transmitted, it usually does not diffuse accurately and flawlessly across companies."

This is another way of saying that embedded knowledge can be (and likely is) fleeting or ephemeral; it changes, in other words.

Consider this example. An employee is charged with developing a new product and has his hands in various business units that may impact the development of the good before it's rolled out publicly. If that employee leaves, he may have knowledge that his former company still is rolling out the product. But much likely has changed since his departure. Perhaps the company switched vendors, changed chemicals that go into the product formulation, or had to purchase new equipment (thereby decreasing the amount of months it would take to generate a positive return on investment).

It is entirely reasonable that the employee would not know anything about these developments, even if he had day-to-day involvement in the product for a long time while he was there. To Professor Lobel's point, if he then used his prior information about the product (his embedded knowledge) in a new position with a new company, it's likely to be inaccurate. That is to say, the passage of time and fluidity of useful information reshapes the employee's knowledge in such a way as to render it less valuable and even counterproductive to a competitor.

A trade secret, by contrast, shouldn't face the diffusion problem. The central point of a trade secret is that it's a patent substitute. An owner achieves an economic advantage (possibly in perpetuity) by keeping the information secret, rather than gaining a limited monopoly through the patent system. If that's the case, then the trade secret should be something the firm can monetize. Embedded knowledge, by contrast, is that which shifts and is not capable of monetization.

An alternative way for lawyers and judges to look at whether something meets the definition of a trade secret may be to get away from the mutli-factored standard that yields more questions than answers and determine whether, in Professor Lobel's words, it can "diffuse accurately and flawlessly across companies."

The Janitor Analogy, Once Again

"The agreement is so broad that it would prohibit my client from working for his new company even as a janitor!"

- Comment attributed to every defense attorney in every non-compete case, generally accompanied by arms flailing up and down violently.

***


Despite my sarcasm, this isn't a terrible argument.

In fact, it's made repeatedly with great success. And the analogy has managed to work its way into one court opinion after another.

The cases in which you see this generally involve a broad non-compete agreement, as opposed to a more limited non-solicitation of customers restraint. The argument never works for a high-level executive because courts generally assume such executives have pervasive access to confidential information of their former employer, such that a tightly defined job-scope limitation isn't required based on the interest the company is trying to assert.

However, the argument has appeal when a mid-level sales person has a restrictive covenant that goes beyond his or her ability to service or solicit clients and that extends to a broader restriction more appropriate for those that develop products or manage the flow of confidential information. So the argument goes, the broad market-based covenant is not narrowly tailored to protect the interest (customer goodwill) the employee can impair.

Two North Carolina appellate decisions discuss broad non-competes in this context, and both came out in favor of the ex-employee. The cases are Horner Int'l Co. v. McKoy, No. COA13-964 (March 4, 2014), and CopyPro, Inc. v. Musgrove, No. COA13-297 (Feb. 4, 2014). Neither case breaks new ground, but they serve as strong illustrations of how courts examine broader covenants as applied to salespersons. Prohibitions on employment altogether for this class of employees will raise major red flags. The problems easily can be cured with more reasonably drawn non-solicitation of customers provisions.

On this score, though, there's a further problem percolating underneath the surface in restrictive covenant law. Too much, generally, is made of the need for a geographic limitation. While some businesses still are hyper-local, many aren't. As the economy continues to evolve and individual spheres of influence extend beyond particular geographic borders, courts are going to have to reformulate their enforceability tests to move away from a geographic analysis. The short concurrence in the McKoy case makes this point.

Changes to our economy pose a special challenge for lawyers who draft agreements for their employer clients. We have to begin paying more attention to scope limitations in non-competes contracts and tailoring contracts to the particular employee. This is an issue separate and apart from the utility of any geographic restraint. More and more, I am reviewing non-competes that have no real meaningful geographic restriction but which are more carefully defined in the type of role the employee would be restricted from assuming after his or her departure.

It's encouraging that drafting seems to have evolved in this respect. But as the North Carolina cases show, there still frequently is a large disconnect between the employee's activities and the employer's business, on the one hand, and the practical impact of the covenant language, on the other.

Fifield, Federal Style

"The law is an ass. But it's not that big of an ass..."

U.S. District Court Judge - Northern District of Illinois on March 5, 2014, discussing holding of Fifield v. Premier Dealer Services.

***

In the past several weeks, two federal district court judges have declined to follow Fifield's bright-line, 2-year continued employment rule for enforcement of non-compete agreements. As readers of this blog know, Fifield generated enough controversy that - at least among lawyers in Illinois - it's now a noun ("I have a Fifield hearing at 2 pm"), a verb ("The judge just Fifield-ed me."), and an adjective ("That ruling had a crummy, Fifield-ish quality to it.").

Surprisingly, neither judge in the Northern District of Illinois cited my dissenting opinion in Fifield (available here). But the gist of both rulings is the same and, remarkably, somewhat similar to what I laid out last year in my beer-induced missiv:

(1) There has to be some latitude in the consideration rule for a particular employee's position within the company;

(2) The closer the employee bumps against the 2-year window, the less relevance the rule has;

(3) The method of termination - voluntary or involuntary - has to be relevant (if not dispositive).

The Fifield decision has caused a fairly wide rift. It's now a federalism issue. Federal courts, under the Erie doctrine, must determine how the Supreme Court of Illinois would look at the issue of consideration. The court yesterday in my case found that the Supreme Court would not adopt Fifield and that under Erie, the two-year rule was not in fact the law in Illinois.

We continue to see a proliferation of trade secrets and non-compete disputes make their way into federal court under diversity rules or as adjuncts to Computer Fraud and Abuse Act claims. This only will continue in light of recent rulings, and it may spur on plaintiffs to add CFAA claims where they otherwise wouldn't.

It's also only a matter of time until another district within the Appellate Court of Illinois (there are five) confronts a Fifield issue, which may give the Supreme Court a second chance to clarify this unacceptably confusing area of the law.

All the while, lawyers will continue hedging their advice with multiple disclaimers. Which, of course, clients love.