III. Legal Impediments to Endogenous Growth

Any society committed to encouraging the endogenous growth that comes from diffusing nonrivalrous information will probably discover aspects of its laws that impede such growth. This Article argues that a genuinely free market in labor, including easy creation and termination of jobs, is an important institutional support-- perhaps a necessary, though not sufficient, condition--of a society of endogenous economic growth. (Obviously there are poor countries with unregulated labor markets and lots of short-term work that are not encouraging endogenous economic growth or investment in knowledge). In most advanced countries outside the United States, labor laws that discourage short-term work, or make it very expensive to create or end jobs, would be the chief obstacle to the development of a Silicon Valley-style high-velocity labor market with rapid growth due to diffusion of information. However, no United States jurisdiction, not even California, has laws that prohibit the creation or termination of short-term jobs.(1)

It is the law of post-employment, rather, that represents the chief legal impediment to endogenous economic growth in most US jurisdictions. Outside of California, employers have many ways of preventing information from diffusing to rivals. They may be able to enforce a covenant not to compete that will effectively prevent an employee from working for a rival or starting up his or her own competing firm. The first employer may be able to enforce the common or statutory law of trade secrets, or a negotiated trade secrets agreement, that may also be enforced through an injunction prohibiting an employee from taking a job with a competitor, or from working in the area that the employee knows best, or from disclosing any unspecified trade secret of the first employer's. Finally, the first employer may enforce invention assignment agreements that make any idea, that the employee comes up with on subsequent jobs, the property of the first employer.

Presumably the individual employers that draft and enforce these agreements benefit from them individually, although I shall show that it is clear enough that most of that benefit is the supposedly illegitimate interest in suppressing competition, not the supposedly legitimate interest in controlling property. However, the economic and sociological analysis so far suggests that a legal system that impedes employee mobility and information diffusion in this way will also impede economic growth.

Silicon Valley's growth reflects the fact that California law has effectively neutralized these three potential impediments to endogenous economic growth. In this Part, I will briefly review some relevant American post-employment law, showing its potential to retard economic growth. I will also show how California benefited from its nineteenth-century decision to render covenants not to compete unenforceable. This statute assisted the growth of Silicon Valley, but by itself was not sufficient. California also had to neutralize the law of trade secrets as an impediment to growth. That story, a somewhat complicated sociology of law study, will be the subject of Part IV. Invention assignment agreements have not figured much in California jurisprudence. I will postpone consideration of them until Part V, when I will discuss them in the context of Professor Robert Merges' recent discussion.(2)

A. The Threat of Covenants Not To Compete, and How California Avoided It.

It is probably a truism about any aspect of California's economic success that California started with great natural advantages, but then went on to make the most of them in ways that were neither inevitable nor even predictable. So it is with California post-employment law. California started with one great natural advantage, one that it shared with few other states: employee covenants not to compete have been unenforceable in California since 1872. This really is close to an exogenous legal factor. The ban was included in a massive codification of the laws adopted from the Field and other codes, and obviously had nothing to do with policy on high-tech employment.(3) Moreover, only two other states, Colorado and North Dakota, ban enforcement of covenants not to compete (and Colorado's statute has exceptions that California's lacks).(4)

By contrast, in most American states--including all the states along the Atlantic--employee covenants not to compete are enforceable if a court finds that they are "reasonable" in duration and geographic scope. (Because of their potential to restrain trade in socially undesirable ways, covenants not to compete have never been assured of enforcement just as written. Restrictive covenants have always been subject to judicial scrutiny for their substance, even in courts otherwise firmly committed to freedom of contract).(5)

Thus, in Massachusetts, New York, New Jersey, high tech employees may be required to sign promises not to compete with their employer for several years after leaving employment-- whether their leaving was voluntary or involuntary. Effective enforcement of such agreements does not necessarily require judicial action. One common technique outside California is to withhold some aspect of employee compensation, such as unvested bonuses or retirement plan payments. These are forfeited should the former employee work for, or start up, a rival.(6) California's statutory prohibition on covenants not to compete prohibits these simple ways of preventing competition --and, in so doing, gave it a leg up in creating the high-velocity labor market that led to Silicon Valley's growth.

This cannot be the end of the story, however, for, despite the statute, California courts could have--but did not--given employers other, different, effective control over departing employees. First, they might have eviscerated the statute. Indeed, Montana and Oklahoma have statutory bans on covenants not to compete that are identical to California's--they derive from the same nineteenth-century source, the Field Code--but have been interpreted by their courts, contrary to their actual language, to permit "reasonable" covenants not to compete.(7) Why didn't this happen in California?

B.The Threat of Trade Secrets Law

Secondly, the law of trade secrets, in some interpretations, is almost as effective an impediment to employee mobility as a functioning law permitting covenants not to compete. This Section reviews the most pro-plaintiff versions of trade secrets law, and shows how their adoption--which is not squarely prohibited by any California statute or case, and may be underway in Superior Court in Santa Clara County--would have impeded the growth of Silicon Valley. Part IV will show how lawyers, judges, juries, and firms in Silicon Valley have, until now, largely avoided these potential consequences.

It would be helpful at this point if one could refer to an authoritative treatise on the Law of Departing Employees or Intellectual Property Disputes Between Employers and Employees, but such a treatise does not exist and is probably not feasible.(8) As we shall see, trade secrets litigation between employers and employees is quite self-consciously fact-specific and resistant to generalization. Fortunately, it is not necessary for our purposes to form an accurate overall picture of the American law of trade secrets, even assuming such a thing were possible. I want only to identify some aspects of trade secrets law, actually found outside California and sometimes within California, that threaten Silicon Valley's endogenous economic growth. Then, in Part IV, we will see how these potential threats have mostly been neutralized.

For this purpose, the chief features of trade secret law that threaten economic growth are the following. First, it permits employers to enjoin employee mobility, and tie up competitors in litigation, under poorly-defined standards that resist summary judgment. Second, where employees do move among competitors, it may encourage wasteful duplication of research and discourage the growth associated with information spillovers. Finally, in a variant adopted by few courts, trade secret law has the potential to become the functional equivalent of a default covenant not to compete in every employment contract. This variant, the doctrine of "inevitable disclosure," permits Employer 1 to enjoin the departure of Employee to Employer 2 on the ground that Employee would "inevitably disclose" some unspecified trade secret of Employer 1's.

In California, and forty other jurisdictions, the relevant authority, as to trade secrets questions, is the Uniform Trade Secrets Act.(9) In the remaining dozen or so, mostly in the northeast, the relevant authority consists of common law cases or the Restatement (First) of Torts, the Second Restatement having omitted the subject.(10) In my observation, little turns on this distinction. Most Uniform Act courts simply treat the statute as cumulative of their earlier common law. I have not seen any cases that carefully construe the Act's language, as if it were a real statute, or make any effort to rethink earlier state law in light of the statute, or of the new interest of uniformity among jurisdictions.

If an employee takes a job with a competitor of his former employer, are his former employer's customer lists trade secrets? Manufacturing processes? Marketing plans? Product formulas or recipes? The answer in each case is a firm "sometimes." All the above are sometimes trade secrets and sometimes not. The distinction does not rest on close parsing of the statutory or common law definition of trade secrets, which is deliberately broad and excludes little.(11) Nor does it rest on economic analysis, either of the type advocated in this Article or any other. I have never seen a trade secrets case that turns overtly on economic analysis.

Rather, information will be held either to be a "trade secret" or "general knowledge" based, in the words of the Massachusetts Supreme Judicial Court, on "whether the information sought to be protected is, in fact and in law, confidential. In our view, the result in each case depends on the conduct of the parties and the nature of the information.[N]o general and invariable rule can be laid down." In the same passage, the Massachusetts court noted that it had "reached varying results" within such broad categories as "customers' and suppliers' lists, manufacturing processes, merchandising techniques, and other information gained by an employee in the course of his employment and used to the detriment of his former employer."(12) Few honest supreme courts could not similarly characterize the law of their state.

The Massachusetts courts are as good as their word, defining employee information as "trade secret" or "general knowledge" under a whimsical and inarticulate scheme of values. My favorite teaching pair consists of two cases decided four years apart by the intermediate Appeals Court. In the first, a departing engineer worked on highly "technical information retrieval and analysis systems" or TIRAS, an acronym that this very engineer had come up with while at Employer 1. Such systems, used in satellites, gather enormous quantities of data on the performance history of each mechanical component, a gyroscope, for example. They then find ways of analyzing this mountain of data so that it is useful and spots actual trouble. The court found that the design of such systems was discussed in publicly-available literature and that neither the overall package, nor any particular aspect, was a trade secret of Employer 1; it was all "general skill or knowledge."(13) In the second case, an employee left a cookie bakery to start his own. He was permanently enjoined from using his former employer's "trade secret" of sweeping up the "chaff" or "nut dust" from the chopped nuts and adding it to the batter.(14) I had thought that everyone who had ever baked chocolate chip cookies had at one time or another swept the nut dust into the batter. In any case, it looks to me more like "general skill and knowledge" than does the design of satellite information retrieval systems, though the court held otherwise. Similarly instructive pairs could probably be retrieved from most state's decisions.

Under this vague standard for defining "trade secret," it is easy for Employer 1 to threaten Employee and Employer 2 with a suit seeking to enjoin Employee from either working for Employer 2, or working in a specific area, or from disclosing any confidential information of Employer 1's. It is correspondingly difficult for defendants to get a dismissal, or summary judgment, in advance of such a fact-specific inquiry. The main impact of this uncertainty comes, not in litigation, but in normal personnel practices. For example, under the statutory scheme, an employee interviewing for a job, with a competitor of his former employer, knows two kind of information about how to do his present job. Everything the employee knows is either generally known in the industry, in which case it's not a trade secret, but doesn't do the employee much good at the interview. Otherwise, if it's not generally known, if it has any economic value--which it must, or neither the employee nor Employer 2 would care about it--it's a trade secret. (The only other possibility is that the information would be a trade secret but for the fact that Employer 1 has not tried to keep it secret. This may not be clear at our employee's interview, particularly if Employer 1 tells its employees that all information about company procedures is proprietary).

Even ordinary working people performing so-called unskilled labor know a great deal about how to do their particular jobs, secrets for running their particular machines, office routines and protocols, that they learned only on the job and that give some competitive edge to their employer.(15) It is by no means rare for ordinary working people without professional degrees to be caught up in litigation over trade secrets, or other intellectual property, when they attempt to change jobs.(16) In short, the law of trade secrets on the books in every American jurisdiction permits any employer ethically and plausibly to enjoin the departure of just about any employee to a competitor.

The legal standards in trade secrets law are thus favorable to plaintiffs, in this sense that a violation may plausibly be alleged and not dismissed before trial. As we shall see in Part IV when we discuss actual cases, the main advantage to any trade secrets plaintiff is the potential of preventing a competitor from being competitive, at least during litigation.

The main impediments to trade secrets litigation are practical. First, like any litigation, it is expensive and unpredictable. Second, trade secrets law offers a particular practical disadvantage to the plaintiff. The plaintiff must often identify the precise information alleged to constitute a trade secret, and, in so identifying it, secrecy will be lost for practical purposes.

However, the most pro-plaintiff versions of trade secrets law excuse plaintiff from the burden of identifying any particular trade secret. They are thus the functional equivalent of a covenant not to compete, permitting Employer 1 to enjoin Employee from working at Employer 2 on the grounds that Employee would "inevitably disclose" some trade secret or other. While few American courts have adopted the doctrine of "inevitable disclosure", a deviant case illustrates the dangers it could cause for Silicon Valley's high-velocity labor market and associated endogenous economic growth.(17)

In the early 1990s, PepsiCo, Inc., and Quaker Oats Company, each manufactured a sports drink, All Sport and Gatorade respectively. When Quaker wanted a new Vice President--Field Operations for Gatorade, the President of its Gatorade division interviewed and extended an offer to William Redmond, Jr. Redmond was then serving his first weeks as General Manager of Pepsi-Cola's California operations, having just been promoted from General Manager of Northern California operations.

What else was Gatorade supposed to do? Either they promote from within, or they go outside. Promoting from within may not make sense if the brand is underperforming. Even if the brand is doing well, however, it is conventionally supposed these days that promotion from within has lost its former resonance, not merely in Silicon Valley, but throughout the economy.(18) If Gatorade doesn't promote from within, or hire a novice, it is, definitionally, looking at experienced sports drink managers.

Nevertheless, Redmond was preliminarily enjoined under the Uniform Trade Secrets Act, which Illinois has also adopted, from taking any position at Quaker for a year; from ever assuming any duties at Quaker relating to beverage pricing, marketing, and distribution; and from ever divulging PepsiCo trade secrets or confidential information.(19)

The injunction was affirmed even though PepsiCo never identified any specific piece of information that Redmond had and Quaker wanted. Since PepsiCo never identified any specific information, the court never had occasion to decide whether the nonexistent information was a statutory "trade secret." PepsiCo did not show that Redmond knew the recipe for All Sport, or new flavors being worked on secretly, or celebrities approached for endorsements, or anything of that sort.

What did Redmond know? He was far from the key figure in All Sport; he was one of many regional general managers. He had access to Pepsi-Cola's "Strategic Plan" and "Annual Operating Plan" of "financial goals, marketing plans, promotional event calendars, growth expectations, and operational changes..." (The report of the case is silent as to whether he read these doubtless vital documents, or tossed them into the corner of his office). He knew which markets Pepsi-Cola would focus on and something about a new delivery system. In other words, he knew what any manager knows. PepsiCo did not identify one piece of this information that Quaker would even care about, or that Redmond would be likely to be called on to disclose. Rather, the court held that PepsiCo "may prove a claim of trade secret misappropriation by demonstrating that defendant's new employment will inevitably lead him to rely on the plaintiff's trade secrets" and that Redmond "cannot help but rely on [Pepsi-Cola] trade secrets as he helps plot Gatorade and Snapple's new course..."

Additionally, Redmond, like all other PepsiCo managers, had signed its form confidentiality agreement stating that he would not disclose at any time (except to PepsiCo), or make use of, "confidential information relating to the business of [PepsiCo] ... obtained while in the employ of [PepsiCo], which shall not be generally known or available to the public or recognized as standard practices." This gave PepsiCo another arrow in its quiver. Without deciding whether or not this "confidential information" included anything not a statutory "trade secret", the court enjoined any of these disclosures as well. It appears, then, that the form contract added little or nothing to PepsiCo's case. The injunction against disclosing trade secrets rested not on the contract but the statute, and the bulk of the court's opinion discussed the statutory, not the contractual, claim.(20)

The Pepsico decision is unusually pro-plaintiff, not typical. It inhabits a world in which all managers on distribution lists for "strategic plans" and "annual operating plans" serve for life, and may have their departures, at least, to work in the area they know, enjoined at the option of their employer. It seems not to relate at all to today's world of corporate downsizing, managerial layoffs and uncertainty. It does not rest on any sort of economic analysis of the efficiency advantages of letting Redmond go as against letting PepsiCo enjoin him, even though the Court of Appeals for the Seventh Circuit is famous for its law-and-economics approach. Nor does it have much to do with employment-at-will, to which the Seventh Circuit has, in other contexts, paid homage. As explained in Part II, employees with other options, such as managers, may choose not to accept employment at will if, following involuntary termination, they will be unable to work in their areas of greatest expertise.

In particular, the Pepsico decision is unusual in its lonely forthright adoption of the theory of "inevitable disclosure", that is, that employers may enjoin the employment of former employees in particular areas because those former employees would "inevitably disclose" some trade secret, even though no one can now say which it would be, and therefore the employer doesn't have to identify it, or any. Counsel for employers in trade secrets cases against departing employees often argue for the adoption of the doctrine of "inevitable disclosure." (21) However, judicial adoptions (apart from Pepsico) have been quite rare.(22)

Nevertheless, the existence of constructions of the Uniform Trade Secrets Act--the same statute that California and most US jurisdictions have adopted-- like Pepsico's represents a serious potential threat to Silicon Valley's endogenous economic growth. The effective neutralization of this threat is the subject of Part IV.

1. 36 There is irony in the statement in text, although I mean it as a simple statement of fact. Somebody who read only law reviews would know of California employment law only that it recognizes numerous causes of action by which discharged employees can challenge their terminations. Our reader of law reviews might also know that California is a leader among the states in its pro-plaintiff employment laws, that it has a large cohort of lawyers who will represent terminated employees on a contingent fee. Our reader might also know of a RAND Corporation study that concluded, on the basis of highly fanciful assumptions, that these pro-plaintiff court decisions had resulted in levels of employment in California 4.7% below the level anticipated absent the decisions. James N. Dertouzos & Lynn A. Karoly. Labor-Market Responses to Employer Liability 51 (1992). As this Article shows, however, the threat to employers of California's wrongful discharge law is normally exaggerated, often by consultants and lawyers who benefit if employers overestimate their exposure. Lauren B. Edelman, Steven E. Abraham, & Howard S. Erlanger, Professional Construction of Law: The Inflated Threat of Wrongful Discharge, 26 Law & Society Rev. 47 (1992). There are in fact no California cases or statutes that prevent employers from clearly hiring people to serve at the will of the employer , and most Silicon Valley professionals understand and are made to understand, in word and deed, that these are the terms of their employment. California law will create problems for the employer, and properly so in my opinion, if the employer wants it both ways: a boilerplate reservation of employment-at-will along with a course of conduct, informal reassurances, and consistent behavior to others, that lead employees to believe that the true contract is one with more protection for them. See, e.g., Foley v. Interactive Data Corp., 765 P.2d 373, 383-88 (Cal. 1988)(employer's course of conduct may give rise to implied-in-fact contract not to terminate without just cause); Scott v. Pacific Gas & Elec. Co., 904 P.2d 834 (Cal. 1995)(implied-in-fact contract not to demote without cause). The Scott case is a particularly interesting example of an early judicial confrontation with the promises implicit in a high velocity labor market. The employees were engineers, openly maintaining their own consulting business, while employed by the large utility, PG&E. The Supreme Court of California applied findings that PG&E was well aware of the arrangement and had impliedly promised the employees that they would not suffer for it. The arrangement might have been anathema in the heyday of internal labor markets, but seems to fit the 1990s' emphasis that employees plan their careers even while employed by a particular company. The Scott case thus encourages consultantships, information diffusion, and endogenous economic growth, and is thus consistent with the normative and descriptive thesis of this Article, that California law should, and does, encourage such growth.

2. Robert P. Merges, Property Rights Theory and Employee Inventions, Calif.L.Rev. (forthcoming).

3. Cal.Bus. & Prof. Code 16600. Professor Ronald J. Gilson traces the history in a forthcoming article. Ronald J. Gilson, The Legal Infrastructure of High Technology Industrial Districts: Silicon Valley, Route 128,and Covenants Not to Compete (unpub. m.s).

4. Colo.Rev.Stat. 8-2-113; N.D. Cent. Code 8-08-06. See generally Covenants Not to Compete: A State-by-state Survey (Brian M. Malsberger, Ed.)(2d Ed 1996).

5. Michael J. Trebilcock, The Common Law of Restraint of Trade: A Legal and Economic Analysis (1986).

6. See, e.g., Nationwide Mutual Ins. Co. v. Darden, 503 US 318 (1992)(insurance salesman who is independent contractor falls outside Employee Retirement Income Security Act; pension payments do not vest and may be conditioned on noncompetition).

7. 42 Dobbins, DeGuire & Tucker, P.C. v. Rutherford, 708 P.2d 577 (Mont. 1985)(upholding agreement requiring payment to former employer if certain clients obtained within one year after departure despite statutory ban on covenants not to compete); Bayly, Martin & Fay v. Pickard, 780 P.2d 1168 (Ok. 1989)(dictum).

8. 43 Catherine Fisk is undertaking a splendid review of the nineteenth century cases. "A Mortgage on the Worker's Brain": Employer Control of Employee Human Capital in the Nineteenth Century United States, presented to the Joint Meetings, Law and Society Association and Research Committee on the Sociology of Law of the International Sociological Association, Glasgow, Scotland, July 13, 1996. Robert Bone has recently argued that the entire doctrine of trade secrets lacks coherent justification. Robert G. Bone, A New Look at Trade Secret Law: Doctrine in Search of Justification, 86 Cal.L.Rev. 241 (1998).

9. 44 14 ULA 438 (1990). For a list of adoptions, see ULA Trade Secret Refs in Westlaw's ULA database.

10. 45 Kitch, supra n.33, argues rather that the Restatement (3d) of Unfair Competition (1993) is crucial, indeed, that only with its adoption did generalist managerial employees become faced with trade secret restraint. I have not yet seen many cases actually citing to this Restatement, but perhaps its influence is subtle and may grow.

11. 46 For example, in California, a trade secret is information that "(1) derives independent economic value, actual or potential, from not being generally known to the public or to other persons who can obtain economic value from its disclosure or use; and (2) is the subject of efforts that are reasonable under the circumstances to maintain its secrecy." Cal.Civ.Code 3426.1, subd. (d). California's statutory definition of "trade secret" is actually, and ironically, a bit broader than the Uniform Act's. Under Uniform Trade Secrets Act 1(4), the definition of "trade secret" , information is not a trade secret if it is "readily ascertainable by proper means." As adopted in California, the statute omits the quoted language. See Abba Rubber Co. v. Seaquist, 235 Cal.App.3d 1, 21-22 and n.9 (1991)(dictum)(stating that in California, showing that information is "readily ascertainable by proper means" is a defense to misappropriation but does not go to the definition of trade secret). As we shall see, California courts construe the statute as if it contained the language omitted by the California Legislature.

12. 47 Jet Spray Cooler, Inc. v. Crampton, 282 NE2d 921 (Mass. 1972)(citations omitted).

13. 48 Dynamics Research Corp. v. Analytic Sciences Corp., 400 N.E.2d 1274 (Mass.App. 1980).

14. 49 Peggy Lawton Kitchens Inc. v. Hogan, 466 N.E.2d 138 (Mass.App.1984).

15. 50 Tom Juravich, Chaos on the Shop Floor: A Worker's View of Quality, Productivity, and Management (1985)(working knowledge of unskilled machine operators in wire factory); Ken C. Kusterer, Know-How on the Job: The Important Working Knowledge of "Unskilled" Workers (1978).

16. 51 See, e.g., Wommack v. Durham Pecan Co., 715 F.2d 962 (5th Cir. 1983)(general laborer in pecan plant invents method for finding worms in pecans; gets fired; company has implied license to use his method); Danieli v. Braverman, 1994 WL 879855 (Mass.Super. 1994)(conference manager who ran registration, badge printing, etc., using commercially available systems; employer 1 claimed--unsuccessfully--violation of common law trade secrets protection).

17. 52 Pepsico, Inc., v. Redmond, 54 F.3d 1262 (7th Cir. 1995). For a critique from a perspective similar to this Article's, see Hanna Bui-Eve (student author), To Hire or Not to Hire: What Silicon Valley Companies Should Know About Hiring Competitors' Employees, 48 Hast.L.J. 981, 998-1000 (1997).

18. 53 Charles Heckscher, White-Collar Blues: Management Loyalties in an Age of Corporate Restructuring (1995); Peter Cappelli et al., Change at Work 184-85 and passim (1997).

19. 54 On remand, the district court narrowed the injunction, permitting Redmond to take the job at Quaker, but merely enjoining any disclosure of Pepsico confidential information or trade secrets. Pepsico, Inc. v. Redmond, 1996 WL 3965 (N.D. Ill. 1996). A news article six months later identified Redmond as Vice President for sales and field operations of Quaker's Snapple Division. Richard Gibson, Quaker CEO, Led by Snapple Shakeup, Tells Outlets Time's Ripe for Fresh Start, Wall St. J., July 22, 1996, at B5. Quaker has since sold Snapple.

20. 55 Compare Pepsico, Inc. v. Redmond, 54 F.3d at 1267-1271 (four page discussion of statutory claim) with Id. 1271-72 (two-paragraph discussion of confidentiality agreement).

21. 56 Evan Chesler, Esq., whose hypothetical about the Engineer who knows his current employer is about to pour money down a rat hole I used supra TAN 29, offered that hypothetical to me in conversation as an argument for the adoption of the doctrine of inevitable disclosure. As Chesler sees it, the doctrine is necessary in order to prevent Engineer ever from being placed in the impossible position of his hypothetical. This Article argues that there is no social advantage in requiring each manufacturer to duplicate the same production of negative information, to waste the same money trying to improve a production process that cannot be improved in that way. The thesis of this Article is that Employer 1 in Chesler's hypothetical should be allowed to keep its negative information secret only if it can show that the information never would have been produced unless it could effectively have been kept secret. It seems unlikely that Employer 1 in Chesler's hypothetical could meet this burden. It produced the negative information, that trying to improve Process 1 is futile, for its own purposes and would have done so even if it could not control the disposition of that information. In fact, in the real world of employee mobility, employers still produce information about their production processes, knowing full well that this information will not remain secret, for reasons explored at length in Part I of this Article, discussing the work of Paul Romer, Eric von Hipple, Charles Sabel, and others. Chesler argued unsuccessfully for the adoption of the doctrine of "inevitable disclosure" in Seagate Technology, Inc. v. International Business Machines Corp., 962 F.2d 12 (table), 1992 WL 96271 (unpublished text) (8th Cir. 1992)(overturning injunction against departing employee for failure to define the confidential information within its scope), on remand, 941 F.Supp. 98 (D.Minn. 1992)(denying any injunctive relief), discussed more fully in Part IV.

22. Since "inevitable disclosure" has been adopted by few courts, there is no precise definition. This Article will use "inevitable disclosure" to refer to trade secrets litigation in which the plaintiff has prevailed without identifying any specific trade secret. In this strict sense, the PepsiCo decision is one of very few US decisions, none from a state's highest court or federal appeals court, to adopt the theory of inevitable disclosure. No reported California case adopts the theory in this sense. It probably underlies an intermediate New Jersey case, National Starch & Chemical Corp. v. Parker Chemical Corp., 530 A.2d 31 (N.J.App.Div. 1987), and an unreported decision of a New York trial court, Doubleclick, Inc. v. Henderson, 1997 WL 731413 (NY County Ct 1997). Neither New Jersey nor New York has adopted the Uniform Act so both decisions rest on common law. PepsiCo was distinguished in Campbell Soup Co. v. Giles, 47 F.3d 467 (1st Cir. 1995) and APAC Teleservices, Inc. v. McRae, 985 F. Supp. 852 (N.D. Ia. 1997), but neither entirely ruled out arguments based on "inevitable disclosure."

PepsiCo appears to be unique in enjoining the employee from working at all for the new firm, though as mentioned this aspect of the injunction was removed on remand, 1996 WL 3965 (N.D. Ill. 1996).

Sometimes, "inevitable disclosure" is used to refer to a trade secrets case in which the injunction goes beyond enjoining disclosure of named, or any trade secrets, to enjoin defendant from working in particular areas or on particular projects of the new employer's. Such cases do not necessarily raise the concern of "trade secrets litigation without a trade secret." The plaintiff has identified specific trade secrets, and the defendant is enjoined from disclosing them on the new job, but in order to make this relief effective, the court enjoins the defendant from working at all on particular projects. I will reserve the term "inevitable disclosure" for cases, like PepsiCo and the two mentioned above, in which plaintiff identifies no specific secret, and consequently will not use it to refer to this second class of cases involving specific trade secrets and broad relief. Cases in this second group are not new and do not raise the acute problems of PepsiCo, though the general thrust of this Article would counsel against routine use of such broad injunctions. Representative cases in which defendants are enjoined from working in particular areas, lest they disclose a specific and identified trade secret of plaintiff's, include Eastman Kodak Co. v. Powers Film Products, 179 N.Y.S. 325 (App.Div. 1919) and B.F. Goodrich v. Wohlgemuth, 192 N.E.2d 99 (Oh.App. 1963). These cases do not describe themselves as "inevitable disclosure" cases, and I don't think they are, though enthusiastic counsel sometimes so describe them.