Executive Compensation: What Worked?

Steven A. Bank is Paul Hastings Professor of Business Law at UCLA School of Law. This post is based on a forthcoming article by Professor Bank; Brian Cheffins, Professor of Corporate Law at the University of Cambridge; and Harwell Wells, Associate Professor of Law at Temple University Beasley School of Law. Related research from the Program on Corporate Governance includes the book Pay without Performance: The Unfulfilled Promise of Executive Compensation by Lucian Bebchuk and Jesse Fried.

There is a widespread consensus that something is seriously wrong with executive pay. Managerial compensation has generated controversy and criticism for at least a quarter-century, but various reforms aimed at curbing compensation have enjoyed little success. For those perplexed or frustrated that these reforms have not checked top pay, history can provide valuable lessons. American business enjoyed unparalleled success from the 1940s to the end of the 1960s, yet CEO pay at the time was comparatively modest. According to one study, in 1949 the median top executive at a large public manufacturing firm made 17 times the pay of the average worker. Most studies of CEO compensation today find that comparable multiples are now in the hundreds. In our new Article, forthcoming in the Journal of Corporation Law, we ask “what worked?” to constrain executive pay during this period, and speculate about what changed to end this regime of (relatively) moderate pay.

We begin with tax. Some present-day observers, most notably the economist Thomas Piketty, credit the mid-20th century’s high marginal income tax rates for low executive compensation. The argument seems plausible; if tax rates meant the government wound up with most of what executives were paid, those executives might be less driven to seek high salaries. But we find that tax rates are not sufficient to explain low executive compensation. Certainly, top marginal rates in the 1950s and 1960s were extraordinarily high by present-day standards, but empirical studies have failed to find a meaningful link between changes in tax rates and changes in pay levels over this period. Even with high marginal rates, furthermore, executives had reasons to value higher pay, such as the need to avoid “compression” of compensation for senior managers and the psychological reward of a high pretax salary. Hence, while tax likely had some effect on compensation, such as creating a bias in favor of “perks” that were relatively lightly taxed, it does not suffice to explain why overall compensation was comparatively low.

Corporate governance mechanisms cited by present-day reformers also did little to hold pay down in the mid-20th century. Today’s critics of executive compensation, for instance, often blame lackadaisical boards of directors for falling under CEOs’ sway and paying whatever is asked for. But boards in the 1950s were, if anything, even less independent from top executives than they are now, as they were frequently stocked with “inside” directors prone to backing the firm’s chief executive. Nor did shareholders at mid-century have any real say in executive pay. A few “gadflies” gained attention protesting executive pay but attempts to limit pay through shareholder resolutions almost always failed, as did shareholder litigation challenging allegedly excessive compensation. Finally, direct government regulation did little to keep executive pay down, even in the case of short-lived controls on pay implemented during World War II, the Korean War, and the Vietnam War. Federal requirements that pay at public companies be regularly disclosed may have contributed to lowering pay, but disclosure alone was insufficient to affect pay—it was what others did with that information that mattered.

So what did hold down executive pay? We identify three “external” variables that appear to have contributed to limiting pay. The first is union power. Mid-century was the heyday of unionism in the United States, with up to 35% of non-agricultural workers belonging to unions. Unionized firms likely were reluctant to pay executives vast sums lest they encourage their unionized workers to demand generous pay as well. Specialists in executive compensation warned that lucrative executive pay could be “provocative of labor problems,” and statistical studies have found a negative correlation between executive compensation and unionization. A substantial decline in union membership in the 1980s and 1990s also coincided with a sharp rise in executive pay.

Second, the market for managerial talent operated differently. Corporations in the 1950s rarely hired top management from outside. Instead, those at the very top were almost always “company men” who had joined the firm in their 20s and expected to, and did, spend their careers there. This lack of external hiring reflected a widespread belief that a corporation’s success did not hinge on having a visionary leader at the helm. By the 1980s, this was changing, with the belief that a dynamic CEO was critical to corporate success becoming increasingly prevalent. This produced a more robust market for senior talent, which in turn helped drive pay upwards. Changing attitudes may have reflected economic reality; one careful empirical study found that in the 1950s a CEO could be credited with just below 10% of a company’s performance, but by the end of the 1990s this increased to between 15% and 17%.

Finally, we conclude that CEO pay was likely held down at mid-century by social norms. Norms—social rules not dependent on the government for enforcement—are difficult to measure but impossible to ignore. However, observers from the 1950s to today repeatedly noted that, at one time, social rules curbed CEO greed. A 1955 study of 50 leading firms, for example, cautioned that “[m]oneygrubbing of substantial proportions is no longer possible for corporate managers,” warning of employee and customer backlash if CEOs maximized their pay. By the 1980s, in contrast, personal gain was no longer so looked down on.

We began our Article by asking what kept pay down at mid-century, in part because we wished to know whether there were any lessons from that time that could be applied today. We conclude, though, that the conditions that helped keep pay down—union power, a different market for managerial talent, norms which discouraged too-overt greed—may be very difficult to restore due to broad based changes in the environment in which public companies operate.

The full Article is available for download here.

Both comments and trackbacks are currently closed.