April 1999

What Do "Equal Employment" and "Loyalty" Mean If Internal Labor Markets Die? : the Emerging Employment Law of Silicon Valley's High-Velocity Labor Market

By Alan Hyde⁽¹⁾

Economists and others discovered in the 1980s that a significant proportion of the American workforce held implicit contracts for lifetime employment. These contracts had a predictable profile: they lasted a long time; compensation increased over the job cycle; and the contract would terminate with the separation of the employee (usually through retirement or, rarely, discharge for cause), rather than wage reduction.⁽²⁾

This discovery proved to be a powerful lens before the eyes of legal scholars. It brought into focus such disparate problems before the courts as age discrimination, wrongful termination, and other workplace torts and causes of action, and from this focus came specific legal and policy recommendations.⁽³⁾ Contrasts between the American and Japanese economies, it was said, had been exaggerated. Many Americans, professionals, managers, and production workers, could in fact expect stable, lifetime employment, and economic and legal institutions had been built around this reality.

Even as the legal framework was synthesized, its social basis was eroding. While stable careers within firms have by no means disappeared, they do not appear to be growing in importance. Study after study has now made clear that average tenure on the job is growing shorter, with particularly notable drops in very long tenures (over ten years) and particular groups (men with no education past high school).⁽⁴⁾ Particularly in industries facing global competition, wages now respond much more readily to market factors and are much less likely to reflect orderly increases from a given baseline.⁽⁵⁾ With hindsight, the wave of successful lawsuits brought in the 1980s by terminated managers did not reflect a new wave of contracts for stable employment.⁽⁶⁾ Rather, it reflected the repudiation by employers of employment contracts, that had seemed efficient ex ante when entered into in the 1950s and 1960s, but came to seem inefficient ex post in the increased global competition of the 1980s and 1990s. The lawsuits may, if anything, have marginally accelerated a trend already underway to specify that employment would genuinely be at will.⁽⁷⁾

Legal scholars have just begun to consider the legal implications of this decline in internal labor markets and stable employment. If the earlier scholarship was correct, as I believe it was, to observe that embedded in much of our labor and employment law is an unexamined assumption that normal work consists of a lifetime career at a single place of employment, the aspects of labor, employment discrimination, pension, and other employment law that will come in for reconsideration are quite numerous.

My remarks today will focus on just two problems: the implications for the meaning of equal employment opportunity, and of postemployment restrictions such as the law of trade secrets or covenants not to compete, in a labor market of rapid employee turnover and consequent reemployment by other firms. The picture of the labor market that I will sketch will have implications for many other problems in labor and employment law that I cannot pursue today.

Drawing on my research in the labor market of Silicon Valley, California, I will emphasize the way in which today's labor markets are also information markets. Employees moving rapidly among firms can spread information across formal and informal networks that make possible a kind of economic growth, and firm and employee organization, that cannot be achieved in any other way. These networks turn out to be the solution to legal problems, of employment discrimination and trade secrets, that appear quite different when posed against the assumption that employees are spending stable careers inside isolated firms.

Before turning to this question of employment equity and posttermination restrictions in a labor market like Silicon Valley's, I want to make explicit one assumption of these remarks. A high-velocity labor market, with rapid turnover and adjustable wages, and without internal labor markets or implicit contracts, is still a labor market. By that, I mean that the impact of the changes, discussed by Professor Bertrand and others, is not that labor markets have now become ideal neoclassical markets, that, under the First Theorem of Welfare Economics, will reach efficient results so long as property rights are specified and free contract maintained. There are many ways in which labor markets, even with wage adjustment and without implicit contracts, are not simply neoclassical markets, and call for distinctive kinds of regulation. If we legal scholars have been too invested in the problems of implicit careers, we have now to consider other features of labor markets, that economists have discussed, and that will play a part in the next generation of employment law.

Labor markets depart from idealized neoclassical markets in many ways that have been the basis for earlier forms of regulation, and will continue to be important even if internal labor markets and implicit contracts disappear entirely (which of course they are a long way from doing). Supply and demand do not reach equilibrium in a market in which people who live by selling their labor power must entire the market as sellers, no matter how poor a price their labor power may fetch.⁽⁸⁾ The contracting preferences of workers and employers are not exogenously given but are deeply and responsively endogenous to the precise background legal regulation of firms and employee organizations.⁽⁹⁾ Baseline or background property rights are never as unproblematic to lawyers as economists sometimes make them.⁽¹⁰⁾ In the world of employment law, the up-for-grabs quality of baseline property entitlements is particularly acute: the shape of employee free speech, discrimination, and union law is powerfully shaped by whether or not employment is at will.⁽¹¹⁾ Contracting preferences will deviate in many ways from the economists' rationality. They will reflect moral and ideal factors and comparisons with other employees that may be much more salient than market factors in setting wages.⁽¹²⁾ Employees negotiating employment contracts will suffer from particular cognitive biases.⁽¹³⁾ They will probably overestimate the protection the law gives them against unfair treatment and omit self-protective action that they might otherwise have taken.⁽¹⁴⁾ Collective action problems will prevent efficient contracts: terms that a particular employer and its employees would prefer to have will not be negotiated (say, protections against wrongful termination) if because of adverse selection the firm would become a magnet for shirkers, or if employers can impose divisions among the employees that prevent them from enunciating preferences.⁽¹⁵⁾ Some employers will be able to set employment terms as if they were monopsonists; for this reason increasing their legally-mandated minimum wage will not cause them to eliminate jobs.⁽¹⁶⁾ This list could be extended, but I hope I have shown why it would be wrong for a legal scholar to interpret findings, like Professor Bertrand's on wage adjustments, as if they showed only that employment is, and always has been, at will, and law's only role is to reflect this supposed fact and then get out of the way. The fact that the economic work referred to in this paragraph, and other work one might mention, hasn't yet been synthesized into a single unified theory of the law of employment contracts is a challenge to scholars and regulators, not an excuse for lawyers and judges to shirk their duties.

In my remarks today, I will concentrate on some legal and economic implications of the decline of internal labor markets that are less developed in the literature than any of the implications mentioned in the last paragraph. When firms no longer offer stable internal careers, and employees therefore move among firms (but within an industry or geographic district), the employees will spread information among competitors, and protect themselves against market uncertainties through formal and informal network organization.

I. Silicon Valley as a High-Velocity Labor Market

For a few years I have been conducting interviews and gathering information on the labor market in Santa Clara County ("Silicon Valley"), California. I view Silicon Valley as a prototype of tomorrow's high-velocity labor markets, in which job tenures are short and internal labor markets weak. Among many such markets that one might study, Silicon Valley stands out as the successful high-velocity labor market, at least for those in its upper half or so. By studying how some professionals make a high-velocity labor market work for them, we may learn something about how to make it work for the ordinary working people who fabricate the chips or scrub the floors. I have no distinctive interest in the work relations of affluent professionals (though some are in fact my best friends). I am interested in anything that can be used to make life more successful for those who find themselves in high-velocity labor markets not through their own choice. If you believe, as I do, that such labor markets are the wave of the future, that is an additional reason to study Silicon Valley. However, it is a very interesting labor market even if it proves more difficult to reproduce than I believe it to be.

My remarks today will stress how poorly it fits many of the assumptions behind the current law of individual employment rights. Specifically, the law of termination (as already mentioned) often assumes implicit contracts for stable employment that are not found in Silicon Valley. The law of employment discrimination usually assumes internal labor markets and hierarchical organization that are rarely found in Silicon Valley; it also defines "employee" so as to exclude between twenty-five and forty-five percent of the labor force. The law of trade secrets defines employer property in a way that could entirely preclude the common Silicon Valley phenomenon of the "start-up" founded by former employees of a different Valley firm.

My initial research interest was in new forms of employee organization common in Silicon Valley, such as identity caucuses (women's, Black, Asian, gay and lesbian), and formal and informal on-line networks.⁽¹⁷⁾ Subsequently, I became interested in the legal problem of posttermination restrictions and the fascinating economic problem of modeling the information effects of rapid labor turnover.⁽¹⁸⁾ Part II of this paper will discuss the meaning of equal employment opportunity in a labor market with weak internal labor markets and rapid employee turnover. The solution turns out to combine my original interest in new forms of employee organization, here, identity-based networks across firm boundaries, with my later interest in how labor markets function like information markets.⁽¹⁹⁾

Before turning to legal problems, however, I would like to present some basic data on the kinds of jobs held in Silicon Valley. My basic point is that such a high percentage of the workforce is employed at what Europeans like to call "irregular employment" that the irregular has truly become the regular. On the one hand, a strategy of converting these jobs into stable careers is neither feasible nor desirable. Stable careers are not feasible because of the sheer number of people whose work would have to change. They are not desirable for reasons explored in the well-known work by AnnaLee Saxenian, who attributes Silicon Valley's economic triumph, over the similar high technology district along Route 128 around Boston, precisely to Silicon Valley's high-velocity labor market. Silicon Valley's engineers broke off to form start-ups in garages while their counterparts back East pursued orderly, less innovative careers inside the internal labor markets at Digital or Wang.⁽²⁰⁾ Careers in Silicon Valley are rarely defined within the boundaries of a single firm, and individuals report loyalty to teams or technologies rather than to individual companies. On the other hand, the high percentage of self-employed entrepreneurs makes further strategies, of encouraging entrepreneurial careers, quite feasible.

Total 1997 civilian employment, Santa Clara County, CA 933,200

Employed by a temporary agency 33,230

Employed in the business services sector 132,400

Self-employed (from census) 69,900

Total employed persons not statutory "employees" 235,480 = 25.2%

Employed part-time (Calif. state est.) 164,290

Total nonemployees and part-time 399,770 = 42.8%⁽²¹⁾

These estimates should be treated with caution as they incorporate many assumptions. As I say, for the purpose of my remarks today, the main point is to try to dispel any nostalgic dream of turning Silicon Valley into a 1950s-style labor market with implicit contracts for stable careers inside single firms.

Measuring contingent employment: the three figures shown are a rough attempt to measure what Benner calls contingent employment and what I, for this legally-sophisticated audience, will called employed persons who are not statutory employees. The easiest to measure are those employed directly by temporary agencies, figures compiled by the California Economic Development Department. In Santa Clara County, employment in temporary agencies is roughly two or three times the national average. Such individuals are normally employees only of the agency, not of the place they work.

It is much harder to estimate the number of people working as contracted labor. Benner, following others, uses as a proxy the number of people working in the business services sector (advertising, computer and data processing, consumer credit reporting and collection, protective services, building services, and personnel services). As Benner notes, this is not a particularly good proxy. On the one hand, it almost certainly double-counts all those individuals in the first group, employed directly by temp agencies. On the other, it counts none of the individuals employed in subcontracted manufacturing, which accounts for a major portion of manufacturing work in the Valley.⁽²²⁾ It is a rough proxy to be sure, but apparently all we have. Finally, the figure on the self-employed includes those who so described themselves in the U.S. Census. This undoubtedly includes people, like the plaintiff class in Vizcaino v. Microsoft Corp.⁽²³⁾, who had been told that they were "independent contractors" but were held by the Internal Revenue Service and the courts to be statutory "employees" entitled to participate equally in company benefits. For legal purposes, then, one might wish for different survey questions. Still, a great many people working in the Valley are not statutory employees of the establishment where they work. (I have not included the latest media discovery, volunteers and community leaders who bestow substantial uncompensated labor on Internet service providers).⁽²⁴⁾

People who work part-time, by contrast, are normally statutory employees of the place where they work. However, should they complain about their treatment, they will probably be compared only with other part-time employees (unless perhaps their precise complaint is that, because of their sex or race, they were denied full-time status). Benner uses a California state estimate for the percentage of the workforce employed part-time, as apparently figures are not kept at the county level. Surely many of these individuals are also included in such categories already counted as temporary workers or contracted labor.

While it may appear therefore that Benner's figures overstate, perhaps considerably, the size of the contingent labor force in Santa Clara County, I do not believe this to be so. Benner and I both suspect that the so-called regular, permanent, full-time employment, with which all these categories are to be contrasted, is not particularly regular or permanent. Here again, Benner's figures are California state figures, not Santa Clara County figures. Only 21% of Californians in the workforce have been with their current employer for ten years or more, compared to a national figure of 35.4%.⁽²⁵⁾ If separate figures for Silicon Valley were available, I know of no observers who think that they would show anything but even more rapid employee turnover. Interviews confirm that most of the self-employed in Silicon Valley are self-employed by choice, at least if they are at the high end. They see little advantage in so-called regular or permanent employment that is neither regular nor permanent.⁽²⁶⁾

II. Employment Equity in Silicon Valley and the Solution of Network Organizations

What does equal employment opportunity mean in a labor market with such rapid turnover and high proportions of subcontracting? Much of our current apparatus of equal employment law assumes stable careers inside firms, and maps poorly onto Silicon Valley. Plaintiffs' lawyers in discrimination cases ask: "Have any women ever been vice-presidents at this company? How many African-Americans serve in this agency above GS-14?" If the company has rapid turnover at all levels, including managerial; subcontracts out important managerial functions to independent contractors and consultants; and maintains few meaningful promotion ladders, what does plaintiff's counsel even want to ask? Is it meaningful to ask the race or sex of "all persons providing services to the company"? Is it legally relevant, if many of those services are provided by individuals who are independent contractors, not employees? There is no federal or state civil rights statute that governs discriminatory outcomes in the private award of service contracts.

Should we be concerned about this lack of fit between our discrimination law and emerging high-velocity labor markets? Much may depend on our sense of whether Silicon Valley presents particular problems of discrimination. It is no paradox to assert that Silicon Valley is both a highly segregated labor market and simultaneously one of the least discriminatory labor markets yet observed. Moreover, as I will argue, the very technologies and work organization that encourage the economic growth of the Valley show considerable promise in addressing issues of employment equity as well.

Earnings are unquestionably unequal and apparently growing more unequal over time. While the average annual income in Silicon Valley was $46,000 in 1997-quite a bit higher than the national average of $29,900⁽²⁷⁾-inequality is also great. Consider four typical employees in Silicon Valley, each the subject of anthropological studies to which we will refer.

At the top, the top one hundred executives in the Valley's largest companies made almost $7 million each in 1996. Their average compensation grew 390 percent from 1991-1996, and the ratio of their income to that of the average production worker rose during that time from 42:1 in 1991 to 220:1 in 1996.⁽²⁸⁾

Next, consider someone working in the software industry, where the average wage was $90,380 in 1997, or semiconductors, $83,690.⁽²⁹⁾ For reasons to be made clear momentarily, I will assume this individual is an engineer or scientist born in Taiwan or India. In the 1990 census, one-quarter of the engineers and scientists employed in California's technology industries were foreign-born-more than twice that of Massachusetts or Texas. In Silicon Valley, one-third of engineers and scientists were foreign-born in 1990. Most observers anticipate that the 2000 census will reveal an even higher percentage. By most standards, these engineers and scientists have done extremely well. They make more than their white counterparts in either professional or managerial positions. Yet two-thirds of Asian professionals in the Valley believe that their advancement to managerial positions is limited by race, and they remain (in comparison to whites with similar education) more likely to hold professional positions and less likely to hold managerial positions.⁽³⁰⁾

My third and fourth stereotyped employees are also foreign-born.⁽³¹⁾ My third stereotyped employee is a manufacturing worker in Silicon Valley who fabricates chips or assembles communication equipment. She is, to quote the title of the forthcoming book by Karen J. Hossfeld, "Small, Foreign, and Female."⁽³²⁾ The average hourly earnings of a manufacturing worker in Santa Clara County was $16.70 in 1998 which represents a decrease of 2.2% since 1990 ($17.07).⁽³³⁾ Our Filipina or Vietnamese woman working for the manufacturing subcontractor is unlikely to be making that much.⁽³⁴⁾

Fourth, imagine the janitor cleaning the headquarters at night. He or she is likely to be a Mexican immigrant employed by a building services contractor and may make $5.88 an hour.⁽³⁵⁾

Now, as I say, from one standpoint this is a very unequal labor market, full of possible discrimination at every level. But from another, the point of view of employee organization, it is the "most meritocratic labor market anywhere in the world,"⁽³⁶⁾ and has the potential to be even more so. By employee organization I don't necessarily mean a labor union. Ironically, one of many things that (as we shall see) unites the Taiwanese or Indian engineer, the Filipina manufacturing worker and the Mexican janitor is that none of them has a union. Instead, I mean network organization. While the economic success of immigrant Chinese and Indian scientists and engineers reflects their hard work and educational achievements, it also, importantly, reflects their employment of a new organizational form: the identity-based network organization that transcends firm boundaries and helps advance careers in an industry. As we shall see, possible adaptation of this form holds great potential for the Filipina in the chip fab shop or the Mexican polishing the floor.

A. Network Organizations for Chinese and Indian Engineers

I mentioned above that two-thirds of Asian-born professionals in Silicon Valley believe that advancement to managerial positions is limited by race.⁽³⁷⁾ Yet I have never seen a lawsuit under Title VII of the Civil Rights Act of 1964 brought by an Asian-born professional in Silicon Valley.⁽³⁸⁾ Saxenian shows that when Asian-born professionals in the Valley feel their advancement blocked within the company, they leave to start their own -what she terms a response in "typical Silicon Valley fashion." "[B]y 1997 Chinese and Indian engineers were running one-quarter of the region's technology businesses started since 1980, and ... these companies collectively accounted for more than $12.5 billion in sales and 46,290 jobs."⁽³⁹⁾

In starting companies, these professionals were not lonely geeks in garages, but drew on extensive formal and informal networks of other professionals built around ethnicity. Saxenian lists thirteen formal professional associations of Chinese or Indian engineers that foster networking and support along fairly narrow ethnic lines. For example, there are distinct associations of engineers with origins in Taiwan (who speak Mandarin at meetings), Hong Kong (Cantonese), and mainland China. Another association was named The Indus Entrepreneur to include Pakistanis, Bangladeshi, and Nepalese; however its members are in fact almost all Indian. Saxenian carefully discusses the role of these associations in facilitating members' career development. They put professionals in touch with role models and sources of venture capital, they serve as important sources of information about markets for labor or products, and offer formal and informal lessons in basic facts of entrepreneurship, management, and (for Chinese) English communication.⁽⁴⁰⁾

This is an important finding. Earlier writing about Silicon Valley, including Saxenian's book, had emphasized rather the isolation of Silicon Valley engineers, normally portrayed as classic loners who joined no formal associations, not even professional associations, had trouble staying married, and bowled (if at all) alone.⁽⁴¹⁾ This picture, though not totally inaccurate, was perhaps ethnocentric. Formal professional associations are alive and well in Silicon Valley, though not labor unions, and not traditional engineering societies. The successful form is rather the identity-based network organization, that transmits information about markets, and facilitates career development through entrepreneurial opportunity.

Saxenian's paper on these organizations is essential reading for anyone interested in the law or organization of a high-velocity labor market, but rather than quote more extensively from it here, I would like briefly to suggest some possible applications of the model, calling for future research. They reflect my belief that law can do more to address employment equity issues in high-velocity labor markets like Silicon Valley's by encouraging network organizations and ease of starting a business, than it can through employment discrimination laws as we know them.⁽⁴²⁾ Two groups that might make more effective use of the network organizational model than they do at present are women professionals in Silicon Valley, and low-wage workers such as Mexican immigrants cleaning the floors. Let me propose a research agenda for studying these groups.

B. Network Organizations for Women?

My own preliminary and unsystematic interviews with women professionals in Silicon Valley suggest that they do not currently make effective use of network organizations, at least, not anywhere near as effective as the Chinese and Indian engineers who have founded so many successful companies. It is possible, however, that such organizations could be adapted to make entrepreneurial opportunities more available to women professionals.

There does not appear to be much published research on women professionals in the Valley.⁽⁴³⁾ A survey on how women professionals perceive employment equity in a high-velocity labor market is urgently needed, and should include questions, suggested by Saxenian's paper on Asian engineers, on how they plan careers transcending firms, and how they use both networks and formal organizations. My unscientific guesses, based on my unscientific interviews to date, follow. Women professionals in the Valley are still most unlikely to be engineers or scientists. They are more likely to work in law, marketing, finance, or personnel relations. Some women professionals have spent quite a bit of time inside one employer (like Hewlett-Packard or Intel); many have moved among many employers; the proportional breakdown is probably no different than for male professionals. For most of these women, Silicon Valley is an outstanding place to work. Companies are relatively open to new ideas and free of prejudices and stereotypes. More women work in really top jobs, as general counsels or in finance, than would be true in almost any other U.S. industrial sector. Moreover, the acceptance of self-employed contracting creates enormous freedom to custom-design combinations of work and family responsibilities. Finally, the knowledge of a professional in law, finance, or marketing does not date like that of a scientist or engineer.⁽⁴⁴⁾

I would guess further that women professionals employ informal networking arrangements to help learn about openings for consultants or managers at other firms. I would also guess that they do not employ formal organizations in the way that Chinese or Indian engineers do. Formal organizations of women professionals are probably smaller and provide fewer services than the organizations that Saxenian studied. They are probably rarely the source of capital.

I would guess that women professionals would express interest in greater use of interfirm formal networks, as vehicles for portable health insurance and other benefits, training, and industry information. Finally, I would guess that, although these women will attribute some lack of success in attaining managerial positions to discrimination, they, like Chinese and Indian engineers, will find networking for interfirm career advancement, and greater opportunities for self-employment, to be more relevant solutions than traditional lawsuits under Title VII. For example, my women informants tend to describe companies that are "good companies" or "bad companies" for women in terms of organizational culture. "Bad companies" don't let you have a family or private life, make unreasonable time demands, and cut you out of the decisional loop if you weren't in the building at 4 a.m. for the frisbee game. While I do not minimize the impact of this kind of organization (if that is the word) on opportunities for women and others who want to raise children, these are not the kinds of issues with which discrimination laws are equipped to deal.

C. Network Organization for Unskilled Immigrants?

Thinking about how to improve life for the Mexican or Central American immigrant, cleaning bathrooms or assembling boards, is when one really sees the power of network organization. While from some perspectives they face stratified and segregated labor markets, it is impossible to imagine them as victims of any imaginable legally-cognizable discrimination. It is not unlawful discrimination or pretextual to discriminate against the unskilled or poorly-educated. Even if it were, the kinds of remedies for discrimination now part of our law would not get our janitor out of his or her dead-end job. The most able and ambitious cleaner at Sun Microsystems has nowhere to advance at Sun. He or she is probably not a Sun employee. Sun probably does not employ any of its own maintenance staff. Neither his employer nor Sun will train him for more skilled maintenance work. The turnover rate of employees at all levels of Sun does not provide an obvious way to imagine a stable career there for our immigrant. Yet these are the traditional legal remedies for discrimination.

A union solution is excellent, as far as it goes, and I favor all the legal changes discussed elsewhere that would facilitate unionization among low-wage service workers.⁽⁴⁵⁾ The economy of Silicon Valley appears robust enough to pay janitors $7 per hour with health insurance, instead of $5.50 an hour without it. A strong union might also concern itself with educational support for the janitor's children or other social services. However, the only kind of career advancement that a union might facilitate is career advancement within the employer (which doesn't exist here, whether the employer is Intel or Bright Cleaning Services) or within the union itself.

Many of our janitors will stay janitors for their working lives and should be able to do so in dignity, able to support a family. But some might like to establish their own building services contracting firms, employing others, or branching into more challenging and remunerative building or other business services. Increasingly, it is recognized that successful careers for low-wage service workers require ladders from dead-end jobs to better jobs, and that those ladders require training and mobility among firms.⁽⁴⁶⁾ One part of this story will turn out to be precisely the kind of ethnic-based networking organizations already employed by Chinese and Indian immigrant engineers and already employed, much more informally, by Mexican and other immigrants in low-wage jobs.⁽⁴⁷⁾ The best strategy for the economic advancement of those janitors and production workers are organizations that put them in touch with successful contractors of building service or manufacturing labor, train them in new skills, teach basic business skills, including the English language, and transmit information about market opportunities. There is absolutely no reason that unions couldn't take on the functions of preparing members for successful self-employment or business ownership, however untraditional that might be.

III. Postemployment Legal Problems in Silicon Valley⁽⁴⁸⁾

A second area of law, placed under substantial pressure by the decline of implicit employment contracts, concerns posttermination obligations such as covenants not to compete, loyalty, and trade secrets. In a labor market in which careers involve successive stops at multiple employers, these problems increase in importance. In Silicon Valley, nearly everybody is a former employee, or former independent contractor, of a competitor. In such a labor market, termination of employment is normal, expected, and not necessarily a big deal. Termination rarely raises any issues of breach of express or implied contract, breach of implied obligations of good faith and fair dealing, or public policy. Engineers or marketing consultants understand that they will be back on the job market when the current project ends. They want to be sure, however, that when they reenter the job market, they will be able to trade on the experience gained on all their previous jobs.

In most states other than California, the law of posttermination obligations forms a major obstacle to economic growth. Outside of California, employers can and do insert boilerplate covenants not to compete and trade secrets agreements into form employment contracts. The effect is to reinforce the more traditional, internal labor markets associated with East Coast high technology districts such as Boston's Route 128, in which employees pursue predictable careers inside firm hierarchies. There is no moral reason not to use law to create such a stable labor market, if that's what a democratic society (say in Europe) wants to do, although one cost of such a labor market is that it doesn't offer the possibilities for growth and creativity of a Silicon Valley, where information flows freely among firms, carried by mobile employees.⁽⁴⁹⁾ (The bigger cost is the high unemployment rates associated with strong protection for job security, a complex issue that I can't pursue here).⁽⁵⁰⁾

Most Silicon Valley start-ups-the centerpiece of the remarkable entrepreneurial culture that so many immigrant engineers have worked successfully-are founded by former employees of another Valley employer, who leave to develop products that will be compatible with, and may compete with, their former employer's.⁽⁵¹⁾ If California courts enforced covenants not to compete, or agreements not to disclose trade secrets, none of these start-ups could have been founded.

The unenforceability of covenants not to compete in California is a happy historical accident that undoubtedly helped bring about Silicon Valley.⁽⁵²⁾ In 1872, California adopted a massive codification of its laws, adopted from the Field and other codes, that included a provision making covenants not to compete unenforceable.⁽⁵³⁾ Obviously this has nothing to do with policy on high technology. It may be just a happy accident that California law made it impossible for Fairchild Semiconductor to draft enforceable agreements, that would have precluded later competition from its former employees who went on to found Intel and other manufacturers.⁽⁵⁴⁾ However, in some other Western states that adopted the same Field Code prohibition on covenants-not-to-compete, the courts construe the statute as if it enacted only the common law's prohibition of "unreasonable" covenants.⁽⁵⁵⁾ So perhaps a deeper explanation is needed of why California courts didn't eviscerate its statute, but continue to deny enforcement to covenants.

The trade secrets story is even more complicated, for here California has adopted, along with forty other jurisdictions, the Uniform Trade Secrets Act.⁽⁵⁶⁾ Here again, some courts construe this statute in a way that would prohibit the ordinary Silicon Valley start-up. A particularly horrible example is Pepsico, Inc. v. Redmond, construing the Uniform Act to enjoin a regional manager from Pepsico from taking a job as Vice President for Field Operations of what was then the Gatorade division of Quaker Oats. Although Pepsico was unable to identify any particular trade secret of its that was jeopardized, the Seventh Circuit granted the injunction on the grounds that he would "inevitably disclose" some trade secret or other on the new job.⁽⁵⁷⁾ This is an unusually pro-plaintiff decision. Trade secrets plaintiffs normally have to identify the precise secret that they fear will be disclosed-that is, courts have mostly refused to adopt plaintiffs' arguments that there is a doctrine of "inevitable disclosure"-and few, if any, cases actually enjoin defendants from taking a job.⁽⁵⁸⁾ Still, if it were generally adopted as the correct construction of the Uniform Trade Secrets Act, there could hardly be a start-up in the Valley, most of which involve at least some former employees or contractors of competing firms who might be called on to disclose something they learned at the last job.

California law is rather unsettled on these points, as there are no important decisions of its Supreme Court, and few of its Courts of Appeals, construing the Trade Secrets Act. Until recently, the interesting story was that relief was unavailable to Silicon Valley employers even when employees went out the door with diskettes or files. Courts and juries tended to find that the relevant information wasn't a trade secret, was general knowledge in the industry, or hardly different from information that had been published, or information that the plaintiff had not kept secret.⁽⁵⁹⁾ The effect was to permit the well-known Silicon Valley phenomenon, already discussed, in which employees leave their employer and obtain venture capital for a new venture, manufacturing software or hardware designed to be compatible, or compete, with their old employer's. More recently, however, some California courts, in unreported decisions, have issued injunctions prohibiting departing employees from working in particular areas at competitors.⁽⁶⁰⁾

I don't think that employers can have it both ways on this one. Obligations of loyalty and confidentiality may have meant something in the days when employers implicitly promised a lifetime career, and kept that promise. They necessarily mean something different, if they mean anything at all, when the employer makes no reciprocal promise to the employee, express, implied, or implicit. Along with this argument from fairness, however, is the economic argument that I have been developing. There is a kind of economic growth that can only be achieved when internal labor markets erode; when employees move freely among firms, spreading information as they go. As I argued in Part II of these remarks, there is also a kind of equal employment opportunity that can only be achieved in such a labor market. So other jurisdictions, that seek to emulate Silicon Valley by renaming some random geographical feature "Silicon," might be better advised to examine the extent to which their law of covenants-not-to-compete and trade secrets impede the formation of a high-velocity labor market and growth district. The law of posttermination intellectual property will be among the many aspects of labor and employment law that will come under pressure if current trends, towards shorter-term work and more adjustment of labor contracts, continue.

1. Professor and Sidney Reitman Scholar, Rutgers. The State University School of Law, Newark, NJ 07102-9173. Phone 973-353-5463, fax 973-353-1445, email hyde@andromeda.rutgers.edu. Prepared for delivery to the New York University 52^nd Conference on Labor, May 20, 1999. Revised drafts of these remarks will be published periodically on my web page, <http://andromeda.rutgers.edu/~hyde>. An earlier version of Parts I and II was presented at Intell4 (International Network for Transformative Employment and Labour Law), Equity, Minimum Standards and the Pressures of Globalization, Cape Town, South Africa, March 18-20, 1999.

2. The discovery of the reality of stable employment is generally attributed to Robert E. Hall, The Importance of Lifetime Jobs in the U.S. Economy, 72 Am.Econ.Rev. 716 (1982). Forerunners include earlier work on internal labor markets, such as Peter Doeringer and Michael Piore, Internal Labor Markets (1971). The idea that such contracts normally precluded downward wage adjustment and thus were subject to termination, whether or not such termination rights had been reserved ex ante, was developed by Edward P. Lazear, Why Is There Mandatory Retirement?, 87 J.Pol.Econ. 1261 (1979). See generally Sherwin Rosen, Implicit Contract: A Survey, 23 J.Econ.Lit. 1144 (1985).

3. Michael L. Wachter & George M. Cohen, The Law and Economics of Collective Bargaining: An Introduction and Application to the Problems of Subcontracting, Partial Closure, and Relocation, 136 U.Pa.L.Rev. 1349 (1988) (duty to bargain under National Labor Relations Act 8(a)(5) should extend only to managerial decisions likely to reflect opportunism left open by implicit employment contract); Katherine Van Wezel Stone, Employees as Stakeholders Under State Nonshareholder Constituency Statutes, 21 Stet. L.Rev. 45 (1991) (labor law implications); Marleen A. O'Connor, Restructuring the Corporation's Nexus of Contracts: Recognizing a Fiduciary Duty to Protect Displaced Workers, 69 N.C. L.Rev. 1189 (1991)(wrongful termination); Stewart J. Schwab, Life-Cycle Justice: Accommodating Just Cause and Employment At Will, 92 Mich.L.Rev. 8 (1993)(wrongful termination); Paul C. Weiler, Governing the Workplace: The Future of Labor and Employment Law 48-104 (1990) (wrongful termination); Alan Hyde, In Defense of Employee Ownership, 67 Chi.-Kent L.Rev. 159, 179-95 (1991)(employee ownership as remedy for strategic nondisclosure of information and high costs of monitoring associated with implicit employment contracts); Robert Howse & Michael J. Trebilcock, Protecting the Employment Bargain, 43 U.Toronto L.J. 751 (1993)(worker participation institutions evaluated as enforcement devices for implicit employment contracts); David Charny, The Employee Welfare State in Transition, 74 Tex.L.Rev. 1601 (1996)(evolution of employee insurance against risk presupposing implicit contracts; discussion of alternatives given decline in those contracts); Samuel Issacharoff & Erica Worth, Is Age Discrimination Really Age Discrimination?: The ADEA's Unnatural Solution, 72 N.Y.U.L.Rev. 780 (1997); Erica Worth, Note, In Defense of Targeted ERIPs: Understanding the Interaction of Life-Cycle Employment and Early Retirement Incentive Plans, 74 Tex.L.Rev. 411 (1995); Steven L. Willborn et al, Employment Law: Cases and Materials, passim, e.g 8-10, 88-103, 174-75 (2d Ed. 1998)(explaining numerous workplace contract and tort actions as protection of employees in lifetime employment contracts). The assumption of a stable career at single employer was not gender-neutral but reflected employment patterns much more typical of men than of women. Joanne Conaghan, The Invisibility of Women in Labour Law: Gender-neutrality in Model-building, 14 Int.J.Soc.L. 377 (1986).

4. Henry Farber, Trends in Long Term Employment in the United States 1979-96, Princeton University Industrial Relations Section, Working Paper No. 384 (1997); Henry Farber, The Changing Face of Job Loss in the United States 1981-1993, National Bureau of Economic Research Working Paper No. 5596 (1996), http://www.nber.org/papers/w5596; Kenneth A. Swinnerton & Howard Wial, Is Job Stability Declining in the U.S. Economy?, 48 Indus.& Lab.Rel.Rev. 293 (1995); Stephen J. Rose, Declining Job Security and the Professionalization of Opportunity, Res.Rep. No. 95-04, Nat'l Comm. for Emp. Pol., May 1995. Peter Cappelli et al, Change at Work 173-93 (1997), review the evidence. Scholars who study only tenures of a year or two sometimes question whether there has been an overall drop in job tenure, though all agree that some identifiable groups, notably men with no education beyond high school, have seen such drops. David A. Jaeger & Ann Huff Stevens, Is Job Stability in the US Falling? Reconciling Trends in the Current Population Survey and Panel Study of Income Dynamics, NBER Working Paper No. 6650 (July 1, 1998); David Neumark, Daniel Polsky & Daniel Hansen, Has Job Stability Declined Yet? New Evidence for the 1990s, NBER Working Paper No. 6330 (Dec. 1, 1997). A comprehensive discussion of contingency is Gillian Lester, Careers and Contingency, 51 Stan.L.Rev. 73 (1998).

5. Marianne Bertrand, From the Invisible Handshake to the Invisible Hand?: How Import Competition Changes the Employment Relationship, National Bureau of Economic Research Working Paper 6900, http://www.nber.org/papers/w6900, January 1999. This is a particularly remarkable finding. As noted, a staple of the analysis of implicit employment contracts was that wages could almost never be decreased; consequently, if the contract had become ex post inefficient, there was no alternative to terminating it. Lazear, supra n.1; Truman F. Bewley, A Depressed Labor Market as Explained by Participants, 85 Am.Econ.Assoc.Papers& Proc. 250 (1995); William R. Johnson, The Social Efficiency of Fixed Wages, 100 Q.J.Econ. 101 (1985). On this model, some wage inflation was necessary component of economic stability. George A. Akerlof, William T. Dickens, & George L. Perry, The Macroeconomics of Low Inflation, 1996 Brookings Papers on Econ.Activity 1.

6. Famous examples include Pugh v. See's Candies, Inc., 171 Cal.Rptr. 917 (1981)(Grodin, J.)(termination of vice-president after thirty-two years of employment may violate implied-in-fact promise). See generally Schwab, supra n.2.

7. The survey of Michigan employers in J. Hoult Verkerke, An Empirical Perspective on Indefinite Term Employment Contracts, 1995 Wis.L.Rev. 837, 868, reports such communication, although the data are somewhat ambiguous as to whether this reflects any real appreciation by the employers of a change in their legal liability.

8. Claus Offe, The Political Economy of the Labour Market, in Disorganized Capitalism: Contemporary Transformations of Work and Politics 16-20 (1985).

9. Mark Barenberg, The Political Economy of the Wagner Act: Power, Symbol, and Workplace Cooperation, 106 Harv.L.Rev. 1379, 1456-96 (1993).

10. Robert Hale, Coercion and Distribution in a Supposedly Non-Coercive State, 38 Polit.Sci.Q. 470 (1923). On the right to terminate the contract as a property right that can be specified as a baseline or reallocated by the parties, see Shubha Ghosh, Property Rules, Liability Rules, and Termination Rights: A Fresh Look at the Employment at Will Debate with Applications to Franchising and Family Law, 75 Ore.L.Rev. 969 (1996).

11. Cynthia L. Estlund, Wrongful Discharge Protections in an At-Will World, 74 Tex.L.Rev. 1655 (1996)(allocation of termination rights in employment contract affects exercise of all other rights in employment law).

12. George A. Akerlof & Janet L. Yellen, The Fair Wage-Effort Hypothesis and Unemployment, 105 Q.J.Econ. 255 (1990); Richard H. Thaler, Interindustry Wage Differentials, in The Winner's Curse: Paradoxes and Anomalies of Economic Life 36-49 (1992); Daniel Kahneman, Jack L. Knetsch, and Richard H. Thaler, Fairness as a Constraint on Profit Seeking: Entitlements in the Market, 76 Am.Econ.Rev. 728 (1986).

13. Samuel Issacharoff, Contracting for Employment: The Limited Return of the Common Law, 74 Tex.L.Rev. 1783,1800-03 (1996)(cognitive disabilities in employment contracts). See generally Stan.L.Rev.

14. Pauline T. Kim, Bargaining with Imperfect Information: A Study of Worker Perceptions of Legal Protection in an At-Will World, 83 Corn.L.Rev. 105 (1997).

15. Adverse selection: David I. Levine, Just-Cause Employment Policies in the Presence of Worker Adverse Selection, 9 J.Lab.Econ. 294 (1991). Collective action: Keith N. Hylton, A Theory of Minimum Contract Terms, with Implications for Labor Law, 74 Tex.L.Rev. 1741 (1996).

16. David Card & Alan B. Krueger, Myth and Measurement: the New Economics of the Minimum Wage (1995).

17. Alan Hyde, Employee Caucus: A Key Institution in the Emerging System of Workplace Representation, 69 Chi.-Kent L.Rev. 149 (1993), reprinted in The Legal Future of Employee Representation 146 (Matthew W. Finkin ed. 1994); Alan Hyde, Employee Organization in High- Velocity Labor Markets, 50 Proc.NYU Conf.Lab. 209 (1997).

18. I'll discuss these issues briefly in Part III of this paper. Paul Romer and I will be modeling some of the economic aspects in forthcoming work.

19. I believe this demonstrates the importance of these related concepts of network organization and the information aspects of labor markets. However, I recognize that this might also bring to mind the aphorism that if your only tool is a hammer, the world is full of nails.

20. AnnaLee Saxenian, Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (1994)

21. These figures and the explanations in the succeeding paragraphs are adapted from Chris Benner, Silicon Valley Labor Markets: Overview of structure, dynamics and outcomes for workers, prepared for conference on Work, Labor Organizations, and Labor Market Institutions in the New Economy: Lessons from Silicon Valley, San Jose, CA, Jan. 29-30, 1999.

22. Most manufacturing by Silicon Valley companies is performed overseas, but production workers physically in Silicon Valley are still needed for prototypic and other short-term projects that must be completed rapidly. "Although manufacturing done on-site in Silicon Valley involves 'higher-tech' stages of the production process, the assembly work that immigrant women engage in closely resembles the same 'low-tech' labor done by their 'sisters' overseas." Karen J. Hossfeld, "Their Logic Against Them": Contradictions in Sex, Race, and Class in Silicon Valley, in Women Workers and Global Restructuring 153 (Kathryn Ward ed. 1990). Increasingly, the local manufacturing employees are employed by subcontractors. Benner, supra n.20, quotes (at 13) an estimate that, for a typical manufacturer of personal computers or related equipment (like Hewlett-Packard or Sun), the cost of components, software, and services purchased from outside has increased over the last decade from less than 60 to more than 80 percent of total production costs. Dieter Ernst, From Partial to Systemic Globalization: International Production Networks in the Electronics Industry. Berkeley Roundtable on the International Economy Working Paper 98 (1997).

23. 120 F.3d 1006 (9^th Cir. 1997), cert. denied 118 S.Ct. 899 (1998).

24. Lisa Napoli, Despite a Passion for the Net, Many On-Line Volunteers Want Pay, New York Times, April 19, 1999, at C1, col. 2; Janelle Brown, Must AOL pay "community leaders"?, Salon, April 16, 1999 (available on-line at http://www.salon.com/tech/feature/1999/04/16/aol_community/index.html). Apparently ten thousand individuals around the country volunteer just for America Online. Other internet service providers like Prodigy and Geocities similarly use volunteers.

25. The national figure for employees with the same employer for ten years or more has been dropping rapidly, though not as rapidly as California's. As recently as 1979, 50% of men between the ages of 35 to 64 had been with their current employer for ten years or more. Today, nationally, that figure is down to 40%. Henry S. Farber, Trends in Long-term Employment in the United States, 1979-96, Working Paper #384, Industrial Relations Section, Department of Economics, Princeton University, July 1997.

26. In one set of interviews of 52 technical contractors, only four expressed a desire to return to permanent employment. Most had never experienced significant time between contracts, although some anticipated such problems in the future. Nearly all made substantially more money as contractors than they had as permanent employees. Gideon Kunda, Stephen R. Barley, & James Evans, Why Do Contractors Contract? The Theory and Reality of High End Contingent Labor 29- 36 (unpub. ms., January 26, 1999)(available from Center for Work Technology and Organization, Department of Industrial Engineering and Engineering Management, Stanford University). By contrast, most in a small sample of low-end temps wanted more permanent employment, at least at points in their lives when health benefits were particularly important. C.N.Darrah, Temping at the Lower End: An Incomplete View from Silicon Valley, presented at the conference Work, Labor Organizations and Labor Market Institutions in the New Economy: Lessons from Silicon Valley, San Jose, CA, Jan. 29-30, 1999. These figures are roughly comparable to those in Sharon R. Cohaney, Workers in Alternative Employment Arrangements: A Second Look, Monthly Lab.Rev. (Nov. 1998)(84 percent of independent contractors prefer that arrangement to a traditional job; only one in three workers employed by temporary agencies preferred that arrangement).

27. Benner, supra n.20, at 18, citing Joint Venture: Silicon Valley Network 1998 Index, http://www.jointventure.org

28. Benner, supra n.20, at 20.

29. Id. 18, citing Joint Venture, supra n.25.

30. AnnaLee Saxenian, Silicon Valley's New Immigrant Entrepreneurs: Skills, Networks, and Careers, presented at the conference Labor Market Institutions in the New Economy: Lessons from Silicon Valley, San Jose, CA, Jan. 29-30, 1999. The survey revealing the belief that racial prejudice limits Asian advancement to management is Asian Americans for Community Involvement, Qualified, But...A Report on Glass Ceiling Issues Facing Asian Americans in Silicon Valley (1993). Those surveyed attributed this less to "racial prejudice and stereotypes" than to the perception of an "old boys network that excludes Asians"and "lack of role models." See also Marilyn Fernandez, Asian Indian Americans in the Bay Area and the Glass Ceiling, 42 Sociological Perspectives 119 (1998).

31. Twenty-three percent of the population of Santa Clara County was foreign-born in 1990, surpassing San Francisco County. Id. 6, citing Rafael Alarcon, From Servants to Engineers: Mexican Immigration and Labor Markets in the San Francisco Bay Area, University of California at Berkeley, Chicano/Latino Policy Project Working Paper, Jan. 1997.

32. An excerpt is Hossfeld, supra n.21.

33. Benner, supra n.20, at 22, citing Bureau of Labor Statistics, Annual Report on Employment, Hours and Earnings.

34. When Hossfeld, supra n.21 at 154-55, conducted her interviews in 1982-86, Silicon Valley manufacturing operatives were 90 percent female, 80 percent foreign, and made entry wages of $4-$5.50 an hour, rising to $5.50-$8 an hour after some experience.

35. Christian Zlolniski, The Informal Economy in an Advanced Industrialized Society: Mexican Immigrant Labor in Silicon Valley, 103 Yale L.J. 2305, 2312-16 (1994), divides Silicon Valley's janitors into three groups: janitors who work for unionized companies under contract with Service Employees International Union Local 1877 and normally make around $7 per hour with health insurance and some other fringe benefits; janitors who work for nonunion contractors and make less than $5.50 an hour with no medical or fringe benefits; and women (mostly) working in small crews for self-employed contractors, paid in cash and making around $3.30 an hour, violating legal standards for wages and also health and safety conditions. Zlolniski does not estimate the proportion of the janitorial workforce in each of the three groups. The figure in text is a median average for janitors' wages, from Benner, supra n.20, at 10, citing California Economic Development Dep't.

36. AnnaLee Saxenian, personal communication, March 9, 1999.

37. Supra n.29

38. The statement in text is literally true. There may well be such lawsuits; I haven't finished looking for them every place I intend to. Still, I mean mainly to contrast the thousands of such professionals and their widespread belief in discrimination with the paucity of lawsuits challenging it; the point would still be valid if the number of those lawsuits turned out to be very small, as opposed to zero.

39. Saxenian, supra n.29 at 2.

40. Saxenian, supra n.29 at 21-43.

41. Saxenian, supra n.19, discusses (at 30-37) informal networks among engineers, and (at 47-49 and 163-64) formal associations among company executives; conspicuous by their absence are formal professional or other associations of the engineers and professionals. The image of the Silicon Valley manager or engineer, who belongs to no formal organizations and has trouble with marital commitment, is central to such popular books as Robert N. Bellah et al, Habits of the Heart: Individualism and Commitment in American Life 3-8 (1985). The "bowling alone"synecdoche, for the supposed general decline in Americans' voluntary affiliations, is from Robert D. Putnam, Bowling Alone, Revisited, 5 Responsive Community (No. 2, Spring 1995) at 18.

42. I am not suggesting for a second that any employment discrimination laws be repealed. I think that some of them could be usefully amended in ways that this paper will not address. On the basic issue of the effect of globalization on employment equity, I think that access to networks, job mobility, and entrepreneurial options are simply more relevant responses than any conceivable changes in the legal definition of discrimination. My entire discussion, however, presupposes the continued presence, not repeal, of existing statutes on discrimination.

43. Saxenian, supra n.29 does not discuss gender at all, but all the individuals mentioned in her paper are men. I think that there are very few women engineering and scientific professionals in the Valley and almost none of Chinese or Indian origin.

44. According to a survey conducted by the National Science Foundation and the Census Bureau, six years after finishing college, 57 percent of computer science graduates are working as programmers; at 15 years the figure drops to 34 percent, and at 20 years-when most are still only in their early 40s-it is down to 19 percent. By contrast, the figures for civil engineering are 61 percent, 52 percent, and 52 percent respectively. Norman Matloff, Now Hiring! If You're Young, New York Times, Jan. 26, 1998, at A19, col. 1.

45. See generally Howard Wial, The Emerging Organizational Structure of Unionism in Low-Wage Services, 45 Rutgers L.Rev. 671 (1993). This may involve more scope for secondary action by labor organizations, see Zlolniski, supra n.34 at 2334-35.

46. See generally Stephen A. Herzenberg, John A.. Alic, & Howard Wial, New Rules for a New Economy: Employment and Opportunity in Postindustrial America (1998).

47. See generally Zlolniski, supra n.34.

48. A much fuller treatment of this subject is my unpublished paper The Wealth of Shared Information: Silicon Valley's High-Velocity Labor Market, Endogenous Economic Growth, and the Law of Trade Secrets, available on my web page <http://andromeda.rutgers.edu/~hyde>

49. The underappreciated role of information in economic growth, specifically, information that is shared with others and from which competitors cannot legally be excluded, is modeled in Paul M. Romer, Endogenous Technological Change, 98 J.Polit.Econ. S71 (1990); Paul M. Romer, The Origins of Endogenous Growth, 8 J.Econ.Persp. 3 (1994); and Paul M. Romer, Capital, Labor, and Productivity, 1990 Brookings Papers on Econ.Activity: Microeconomics 337. I don't mean to suggest that employee mobility is the only mechanism of what economists sometimes call "information spillover." For example, in biotechnology, knowledge of patentable microorganisms doesn't "spill over" when employees change jobs; that kind of knowledge, from which others can be excluded, is purchased in market exchanges with scientists. Lynne G. Zucker, Michael R. Darby, & Jeff Armstrong, Geographically Localized Knowledge: Spillovers or Markets?, 36 Econ.Inq. 65 (1998).

50. See Alan B. Krueger & Jörn-Steffen Pischke, Observations and Conjectures on the U.S. Employment Miracle, National Bureau of Econ. Res. Working Paper No. 6146 (1997); Alan Hyde, Employment Law After the Death of Employment, 1 U.Pa.J.Lab.& Emp.L.99, 106-109 (1998).

51. Saxenian, supra n. 19, at 25-27; Joseph Bankman, The Structure of Silicon Valley Start-Ups, 41 UCLA L.Rev. 1737, 1739 (1994); Mark Charles Suchman, On Advice of Counsel: Law Firms and Venture Capital Funds as Information Intermediaries in the Structuration of Silicon Valley (PhD. Dissert., Stanford Univ., Sociology, 1994).

52. This thesis is developed in more depth in Ronald J. Gilson, The Legal Infrastructure of High Technology Industrial Districts: Silicon Valley, Route 128, and Covenants not to Compete (unpub.ms. 1998).

53. Now Cal.Bus.& Prof.Code 16600. The statute permits covenants-not-to-compete when a business is sold, but forbids them in most employment settings. The only two other states that appear similarly to ban covenants-not-to-compete are Colorado and North Dakota, and Colorado's statute has some exceptions that California's lacks. 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).

54. See Tim Jackson, Inside Intel (1997).

55. See, e.g., Dobbins, DeGuire & Tucker, P.C. v. Rutherford, 708 P.2d 577 (Mont. 1985); Bayly, Martin & Fay v. Pickard, 780 P.2d 1168 (Okla. 1989)(dictum). Both Montana and Oklahoma have statutory bans on covenants not to compete that are identical to California's but are construed by the courts to permit "reasonable" restriction. On those common law restrictions, see generally Michael J. Trebilcock, The Common Law of Restraint of Trade: A Legal and Economic Analysis (1986).

56. 14 ULA 438 (1990). The California version is Cal.Civ.Code 3426.

57. Pepsico, Inc., v. Redmond, 54 F.3d 1262 (7^th Cir. 1995). 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, 1996 WL 3965 (N.D. Ill. 1996). The decision has been criticized. 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).

58. I have found only two cases that appear to adopt "inevitable disclosure" in the Pepsico sense, that is, plaintiff gets relief in a trade secrets case, without identifying any specific secret, on the grounds that defendant will inevitably disclose something. National Starch & Chem.Corp. v. Parker, 530 A.2d 31 (N.J.App.Div. 1987); 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. I have not located any cases, other than Pepsico, that enjoin a defendant from taking a new job, though as I say this aspect of the injunction was later removed by the district court. There are a number of cases, many old, that enjoin defendants from working in particular areas on a new job, e.g. Eastman Kodak Co. v. Powers Film Prods., 179 N.Y.S. 325 (App.Div. 1919). These cases have nothing to do with "inevitable disclosure", as I am using that term. In them plaintiffs have identified a specific secret that they own and don't want disclosed, but, in order to make the relief effective, the court goes beyond enjoining disclosure of the secret, to prohibit defendant from working on particular projects. However, since "inevitable disclosure" is not a part of the law, having been adopted by few courts, there is no precise definition of it, and sometimes enthusiastic plaintiff's counsel describe cases of the "broad injunction" type as cases adopting the doctrine of "inevitable disclosure."

59. One that received a great deal of publicity was Intel's completely unsuccessful litigation against ULSI Technology, Inc., both criminal and civil. See Jackson, supra n.52, at 284-93.

60. Advanced Micro Devices, Inc. v. Hyundai Electronics America, No. CV752679 (Cal.Super.Ct., Santa Clara County, Apr. 5, 1996), discussed in Dan Goodin, AMD, Hyundai Resolve Litigation Over Trade Secrets, The Recorder, Nov. 19, 1996, at 1 (avail. in LEXIS, News data base), and Crista Hardie, AMD, Hyundai Settle Flash Squabble, 42 Electronic News, Nov. 25, 1996, at 16 (avail. 1996 WL 13933861); Ian C. Ballon, Keeping Secrets, IP Magazine, March 1998,avail. on-line at <http://www.ipmag.com/98-mar/ballon.html>, reporting on GTE Government Systems Corp v. Patrick, No. 770111; Ian N. Feinberg, Inevitable Disclosure of Trade Secrets: A New Problem for Companies Hiring Experienced Technical Employees, avail. on-line at <http://www.gcwf.com/articles/interest/interest_5.html>, reporting on IBM v. Read-Rite, Inc.