Why Has CEO Pay Grown So Much Faster Than the Average Worker’s?

But it was not always that way, says Carola Frydman, an associate professor of finance at the Kellogg School. From the 1940s through the 1970s, wage differences between high and average earners actually fell. They then started skyrocketing in the 1980s, Frydman says, peaking in the early 2000s. Pay disparities among top executives in different firms also rose during this period.

Less clear, though, is why these differences in pay have changed over time. “We really don’t have a good model for what determines these disparities in pay. And so we thought it was important to take that step,” Frydman says. “Income inequality is at the heart of the current political debate. Yet without understanding its root cause, it is difficult to judge its magnitude.”

So what were CEOs and other senior leaders doing during the late 20^th century to merit pay hikes that so vastly outstripped their employees?

That question is at the core of new research from Frydman and Dimitris Papanikolaou, also an associate professor of finance at Kellogg. Their research focuses on the role of technological innovation. Specifically, they wondered how much of the growth in executive pay could be tied to a CEO’s ability to recognize promising innovations and steer a firm to invest in them.

While executives only make up a tiny fraction of the labor force, Frydman says that their outsize influence on pay disparity demonstrates precisely why it is so important to better understand how executives are compensated.

“They’re important. They are the main decision-makers in large firms, and therefore their decisions can have a big impact in the economy,” she says.

Executives make routine day-to-day decisions that affect the productivity of their firms, but they also make big investment decisions that can alter the course of their firms, Papanikolaou adds.

“It is the skill to make the right investment decisions that differentiates a good from a bad executive,” he says.

In their study, Frydman and Papanikolaou construct a mathematical model that replicates how an executive’s compensation might change based on investments in new technologies. They find that technological innovation is a significant factor in pay disparity, explaining about half of all fluctuations in the ratio of executive pay to worker pay. Their model implies that periods of rapid technological change will be accompanied by increases in differences in pay, both between executives and workers, as well as among different top executives.

Executive Functions

In their model, workers receive a wage that is determined by the overall output of the firm: when output goes up or down, so do wages.

Payment for executives, however, is more complex. Many of their day-to-day duties, like managing personnel, contribute directly to the firm’s output, and they are compensated accordingly. Unlike other workers, however, executives also spend part of their time planning for the future by identifying new ventures for the company to invest in—and they’re compensated for this as well.

In the model, game-changing technologies appear randomly and can increase the firm’s returns on certain capital-intensive projects. (Think Toyota choosing to develop the Prius just as electric batteries became more efficient and cheaper to produce.) Those returns then increase the executive’s paycheck—if the executive had the good sense to invest in a venture that took advantage of the technology, that is.

When these technologies appear, “it’s really, really important to have talent within your firm that is good at finding new products and valuable investment opportunities,” Frydman says. In other words, this is when executives earn their keep.

Frydman and Papanikolaou then turned to historical data on worker pay and firm output to see how closely the model replicated the real world. They drew from two databases, one of which included firm information and executive pay for hundreds of companies going back to 1992, and the second of which had been gathered by Frydman for an earlier study, with data going back to 1936.

In order to calculate the total added value that an executive brings to a firm via new projects, the researchers measured how much the stock-market value of a firm increased whenever the firm issued a new patent. Then, by totaling all of a firm’s post-patent value increases in a given year, they could approximate the added value of innovation.

“It’s a very neat way to measure it, and it works quite well within our model,” Frydman says of the technique, which Papanikolaou had devised in earlier research.

Real-World Comparisons

Frydman and Papanikolaou’s model was able to accurately account for about half of the fluctuations in the ratio of executive to worker pay seen since the 1930s.

Specifically, the model managed to accurately reproduce the increase in disparity between executive and worker pay that began in the 1980s. But it did not do as well with other eras.

“In the data, the 1950s, the 1960s, were a pretty equal period with low disparities in pay,” says Frydman. “But our model would predict more inequality given the pace of technological advancement. And what that’s telling us is, there were other factors—that it wasn’t all about technological innovation.”

For instance, the model did not consider whether the bargaining power that executives had relative to their workers changed over time. Thus, the unexpectedly low pay disparity in the real world could have been due to factors like unionization, which helped keep executive pay in check for much of the 20th century, as show by Frydman in an earlier study.

The findings shed light not only on the disparity between workers and executives, but on the disparities among different executives as well. As the authors expected, there was a correlation between executive pay and how quickly a firm grew. And the introduction of new technologies that increased firms’ productivity tended to provide the biggest boost in pay to the executives who best took advantage of those technologies.

Overall, finding new investment opportunities accounted for 63% of the average executive’s pay, with normal work duties accounting for the remainder.

The research comes at a time when executive pay is under increased scrutiny—and not just from scholars and politicians. The SEC recently started requiring publicly traded firms to disclose the pay of their executives relative to their workers within the firm, Frydman explains.

Despite the limitations of the model, the authors think it provides an important step in understanding income inequality in the United States.

“We wanted to see how far a model of purely technological innovation and growth can go in explaining these trends,” Frydman says. “And what we find is, it can go quite far.”