We’ve created a new economic aristocracy in the United States: CEOs. That’s a fair reading of recent corporate pay surveys. A study by compensation consultant Equilar for the New York Times finds that Charif Souki of Cheniere Energy was the highest-paid chief executive officer in 2013 at $141.9 million. Oracle’s Larry Ellison clocked in at $78.4 million. But more interesting than these individual totals are typical amounts. In 2013, CEO pay for the top 200 companies averaged $20.7 million, consisting of $6.8 million in cash and the rest in stocks and options.

Any CEO of a major company is virtually guaranteed to become a multimillionaire. In the Equilar survey, the median holding of company stock was $83 million. CEO compensation has vastly outstripped average wage gains. In 1980, CEO compensation for the top 350 firms was roughly 30 times typical worker pay, estimates the Economic Policy Institute, a left-leaning think tank. Now, that ratio is almost 300 to 1. (The peak was nearly 400 to 1 in 2000.)

What does society get from this lavish pay? It’s unclear how much, if at all, economic growth has improved. CEOs’ dependence on stocks might even have hurt the recovery if firms squeezed hiring and investment to maximize short-term profits and share prices. Rising executive compensation has fueled growing economic inequality. Executives and top managers represent nearly one-third of the richest 1 percent of Americans by income, reports a study by economists Jon Bakija, Adam Cole and Bradley Heim.

Let me be clear: I’m not against CEOs. Over the years, I’ve met many. They have usually struck me as intelligent, well-informed and down-to-earth. Few have seemed to be the stuffed shirts of stereotypes. Of course, they promote their corporate self-interest. That’s what they’re paid to do.

It’s also true that some CEOs are transformative. They decisively shape — or save — companies. In this category, I’d put Lee Iacocca of Chrysler in the 1980s; Lou Gerstner of IBM in the 1990s; and, more recently, Steve Jobs of Apple and Alan Mulally of Ford. There are others, including many CEOs who are founders of firms. They often deserve hefty payouts.

Still, most CEOs are not so heroic or influential. Most seem overpaid by two common-sense tests: You could pay them less, and most would take the job anyway; and many — if fired tomorrow — couldn’t get work near their present pay.

The history of this overpayment is instructive. Until the 1980s, most CEOs were appointed from within. They were “organization men” who spent most of their careers at one firm and identified with it. “CEO pay was designed with reference to the rest of the organization,” says finance professor Charles Elson of the University of Delaware. CEO pay couldn’t get too far ahead without antagonizing others in the company.

This system presumed the superiority of U.S. management. By the late 1970s, this premise was untenable. Japanese competition threatened many U.S. firms. Profits suffered. Stocks lagged. “Entrenched” executives were often blamed. New management ideas emerged. Companies should not confine CEO searches to inside the firm. Outsiders were often more qualified. Executive compensation should be aligned with “shareholders’ interests” — more pay should be in stock so that CEOs would pursue higher prices.

Up to a point, this was a constructive response. Many firms were complacent and uncompetitive. They needed to change, even at the cost of disruption. Otherwise, they would not survive.

But the new theories also produced accidental excesses. From 1980 to 2000, stock prices increased by a factor of 12. Little of the gains reflected better management. The most important causes were declining inflation (and interest rates) and long economic expansions. But with more executive pay tied to stocks, many CEOs enjoyed huge windfalls. All this affected norms and expectations. Executive pay catapulted to a much higher plateau.

We’re left with a system that follows its own peculiar logic. As Elson, the Delaware professor, notes, it has an upward bias. In evaluating executives, company directors — often CEOs themselves — rely on compensation consultants who compare pay levels with those at “peer” companies. But if most CEOs expect to be paid at least the average of their peers, then the average will tend to rise. Another justification for lucrative pay packages is that there’s a “competition for talent.” Companies must pay to keep their stars. This is a simplification. CEOs successful at one firm cannot automatically duplicate their success elsewhere because company differences are too great. In reality, most CEOs are paid so well they can leave anytime they please.

Americans dislike aristocracies. Unless companies can find a more restrained pay system, they risk an anti-capitalist public backlash. This is the ultimate danger. For all the flaws of today’s system, government regulation of pay — responding to political needs and pandering to popular prejudices — would be much worse.

Read more from Robert Samuelson’s archive.