Aside from the usual consortium of Zamboni drivers and Don Cherry fanatics, few people mourned when the National Hockey League cancelled its season, a few weeks ago, after the owners and the players failed to come to a new labor agreement. But the fact that people are uninterested doesn’t mean they don’t have an interest. Businessmen should be paying attention to the N.H.L., because its troubles could soon be theirs. The impasse is less about hockey than it is about history—and being on the wrong side of it.

Traditionally, owners haven’t had such a hard time. In the struggle between capital and labor, more often than not capital has won, because the real source of value for most companies has historically been the hard assets that they owned and controlled. Toyota owes its success to its machines, its assembly lines, and its system of production. For Wal-Mart, it’s primarily store location, technological efficiency, and product selection. For Coca-Cola, it’s carbonated beverages and exceptional distribution. Workers for these companies are, for the most part, interchangeable, so their bargaining power is limited.

But in a host of industries—most notably in what we now call the knowledge economy—the arrangement is different. In Hollywood, in Silicon Valley, on Wall Street, and in professional sports, hard assets matter far less than people. The employees—the so-called knowledge workers—make the difference between success and failure. (Difficult though it may be to think of pro hockey players as knowledge workers, that is essentially what they have become.) Capital is plentiful; it is skilled people who are scarce. The salient struggle, in this realm, is no longer capital versus labor but, in the words of the business professors Roger L. Martin and Mihnea Moldoveanu, capital versus talent. So when N.H.L. owners speak wistfully of old-time hockey, what they really mean is old-time economics—when the boys were labor, not talent.

The upshot is that in many knowledge businesses the employees often do better than the shareholders. Investors in the Hollywood studios have historically earned small returns, and yet directors and actors make tens of millions of dollars. In the N.H.L. the past two seasons, players reportedly took home seventy-five per cent of the league’s total revenue. Even in Silicon Valley, land of the inflated stock price, companies are so desperate to attract and keep the best and the brightest that workers often prosper at the expense of the capitalists. In 2000, according to a Business Week estimate, Cisco Systems employees earned between five and eight billion dollars in option profits alone—in a year when the company made only $4.6 billion. And, according to the 2003 book “In the Company of Owners,” during the tech boom a few years ago employees at the top hundred New Economy companies pocketed almost eighty billion dollars in compensation.

Talented workers were always in demand, but only recently did they recognize how much they could get for their services. In 1957, Mickey Mantle earned sixty thousand dollars. This year, Carlos Beltran (who is no Mickey Mantle) will make fifteen million dollars. Adjusted for inflation, that’s about forty times as much. In the forties, Clark Gable made a hundred thousand dollars a picture—about eight hundred thousand dollars in today’s money. Tom Hanks makes closer to twenty million dollars a film. According to Martin and Moldoveanu, C.E.O.s now earn, on average, eight times as much per dollar of corporate profit as they did two decades ago. And on Wall Street traders and bankers now walk away with a much bigger cut than they ever have before. Things may be getting harder for traditional labor—real wages for most workers actually fell last year—but they’re getting better for the talent.

Businesses are still figuring out what to do about all this. Closing shop, as the N.H.L. has done, is hardly sustainable. In hockey, as in other sports, the owners have stayed in the game by coöperating to hold down salaries—by forming a kind of union of their own—but in most industries you can’t do that without violating antitrust law. Many companies have tried giving workers a greater stake, either through revenue-sharing or through stock options, but there’s little evidence that what’s good for the talent is good for the shareholder.

Some companies have succeeded at the talent game by increasing revenues. A bigger pie makes it easier to give out a bigger slice. Others have chosen to change the game—to call into question the central premise that talent is scarce. For instance, one reason that C.E.O.s are paid so much is the belief that there aren’t many potential good ones. But, as the Harvard Business School professor Rakesh Khurana has shown, a company that broadens its definition of what constitutes a C.E.O. can find high-level talent at bargain prices. The same goes for companies that look abroad (to India, to Russia) for programmers and engineers; they’re finding that the talent pool is bigger than they thought.

Still, the trend is unmistakable; in the words of Martin and Moldoveanu, the “talent class has declared war on shareholder capitalists.” There are worse things. You don’t have to be a Marxist to think that if employees are creating most of the value then they should reap most of the rewards. The real problem arises when the rewards extracted vastly exceed the value created. Consider Ak Bars Kazan, a pro hockey team in the Russian republic of Tatarstan. Ak Bars fans pay five dollars for a seat and twenty cents for a cup of tea. The team will be lucky to gross a million dollars this year. And yet it shelled out roughly fifteen million dollars to recruit a handful of idle N.H.L. stars. Ain’t payback a bitch?