Did The Solution Make The Problem Worse?

Thoughts on pay incentives and eating what you kill.



Thoughts On Pay Incentives and Eating What You Kill

Frederick Winslow Taylor was more than just the father of scientific management. He was the father of management consulting as a performance art. During his glory years in the early 20th century, he liked to invite businessmen to his estate in Chestnut Hill, Philadelphia, seat them around his fireplace and, with his pet cat Put Mut perched on his shoulder, regale them with stories about his time working for Bethlehem Steel. His accounts were populated with colorful characters — immigrant laborers who mispronounced English in various amusing ways but could carry enormous hods of pig iron, “college men” who could speak English perfectly but knew little about life on the factory floor. But his stories all had two points: that he had discovered a new principle of scientific management and that the essence of that science lay in the proper use of incentives. Offer employees more financial incentives and they will work harder.

Taylor’s idea quickly won him a huge following. Acolytes compared him with Luther or the Messiah or even God himself: “Taylor in this moment is comparable to the Almighty,” said one convert with a straight face. Businessmen paid top dollar to sit by his fireside. Companies redesigned their incentive structure to exploit his ideas. No one blinked when his gravestone proclaimed him “the father of scientific management.”

The scientific management craze inevitably produced a backlash. Social scientists questioned its intellectual underpinnings. Elton Mayo, of Harvard Business School, and Mary Parker Follett argued that workers were often motivated by love of the job rather than lust for rewards, and that the pursuit of individual gain could undermine collective activities. In 1960, Douglas McGregor of the Massachusetts Institute of Technology published “The Human Side of Enterprise,” which argued that rising prosperity meant companies should pay less attention to “external coercion and control” and more to getting people excited about their work. Eight years later Frederick Herszberg wrote a career-crowning article for the Harvard Business Review which argued that money was nothing more than a “hygiene factor”: not enough of it causes distress but by itself it has little to do with job satisfaction.

But in the 1980s, the pendulum swung again. Michael Jensen and Kevin Murphy mounted a full-blown assault on the idea that CEOs should be paid “like bureaucrats” and argued instead that they should be given stock incentives to perform. This would solve what they called the “agency problem”: the fact that managers sometimes acted in their own interests (feathering their nests or sleeping on the job) rather than acting in the interests of their employers, the shareholders. It would also update scientific management for a more sophisticated era. Taylor had focused on using calibrated rewards to motivate shop floor workers. Jensen and Murphy used them to motivate senior managers.

Jensen’s thesis electrified the business world. Businesspeople had grown sick of the industrial militancy and sluggish growth that characterised the 1970s. They worried that America had lost its edge, particularly to Japan. The Jensen-Murphy thesis offered them a chance to bounce back. Jensen’s acolytes poured out of Harvard Business School and preached the gospel of performance-related pay the length and breadth of corporate America. Jensen’s article in which he outlined his conclusions soon became the most cited article in management literature. Upstarts such as Bain & Company and Bain Capital used highly leveraged incentive schemes to restructure their clients and launch new companies such as Staples, the giant chain of office stores. Established companies adjusted their payment systems to encourage senior managers to behave like entrepreneurs. Corporate America had once looked askance at bonuses (in the 1920s, bonuses were much more common in Europe than in the United States). But by 2006, more than 60 percent of the compensation received by the bosses of America’s biggest companies was tied to performance.

The past 20 years have provided a damning verdict on Jensen’s theory. A new approach that was supposed to solve the agency problem made it worse. Managers got far richer thanks to a combination of a long boom and stock prices that were in some ways responsive to known stimulus. A theory that was supposed to ensure that capitalism worked more smoothly has produced a succession of corporate disasters. Michael Jensen has been reborn as a critic of what he contends is the misapplication of his ideas. He has published a series of articles with titles such as “CEO Incentives: It’s Not How Much You Pay, But How” and “Integrity: A Positive Model That Incorporates the Normative Phenomena of Morality, Ethics and Legality.”

The 2000s saw an epidemic of corporate fraud. The decade started with a succession of collapses and near-collapses on Main Street — Enron, WorldCom and Arthur Andersen disappeared, and Tyco almost followed them. It expanded into a full-blown financial crisis in 2007–2008, with Lehman Brothers collapsing, a dozen other banks shivering on the edge and governments around the world forced to inject trillions of dollars into the economy.

The overwhelming lesson from all this is that incentives can be very dangerous indeed. The companies at the heart of these disasters were all enthusiastic supporters of performance-related pay and highly leveraged bonuses. Jeff Skilling, the boss of Enron, earned $100 million in a single year. He was also a devotee of the “rank and yank” principle:  Every year he gave his best-performing employees generous bonuses and sacked the worst performers. Richard Fuld, the chairman and chief executive of Lehman Brothers, earned $40.5 million in 2006 and $34 million in 2007. He sat on top of an “eat what you kill” culture in which employees were paid strictly according to the business they generated.

Incentives can distort lives as well as ruin companies. In “How Will You Measure Your Life?” Clayton Christensen argues that a striking proportion of his contemporaries as students at Harvard Business School or Rhodes Scholars at Oxford— the elite of the elite — had made a mess of their lives. Some made a mess of their personal lives: There were numerous stories of divorces or unhappy marriages. Some like Skilling ended up in prison (for 24 years, later reduced to 14).

It is not unusual for high-fliers to fall to earth: The Greeks warned us that when you fly near the sun the wax that holds your wings together tends to melt. But the pursuit of outside rewards probably played a role in bringing out the darker sides of some executives’ personalities. Christensen notes that the Jeff Skilling he knew as a Harvard business student was a brilliant and idealistic man. It is not unreasonable to surmise that the prospect of earning $100 million a year changed him.

More generally, the incentive culture has eroded faith in the capitalist system. In 1960, CEO pay was 30 times an average worker’s salary. Today it is about 200 times. This pay inflation led to widespread resentment even before the 2007 financial crisis. A growing number of people worried that the “One Percenters” were pulling too far ahead of the 99 percent — and that the meritocratic dream of upward mobility was being destroyed.

The financial crisis turned resentment into fury. Some observers worried that highly incentivised CEOs were willing to endanger the entire capitalist system in order to win their bonuses. They fumed when taxpayer money was used to bail out too-big-to-fail companies or to pay pensions to the likes of Fred Goodwin, the man who brought down RBS, the big Scottish bank. In the United States, the capitalist country with the highest tolerance for inequality, the proportion of Americans who believe that rich and poor are in conflict has increased from 47 percent in 1990 to 66 percent in 2012.

This resentment is inevitably producing a reassessment of the merits of performance-related pay, not just among an angry public, but also among anxious scholars and policy-makers. Congress has passed two sweeping corporate reforms: the Sarbanes-Oxley Act of 2002 and the Dodd-Frank Act of 2008. Sarbanes-Oxley tried to force boards to keep a more vigilant eye on CEOs. Dodd-Frank obliged companies to disclose the ratio of the CEO’s pay to that of the median worker in the company. In Switzerland, voters granted shareholders a binding veto over executive pay (though they also voted against capping executive pay at 12 times the salary of a company’s lowest-paid worker). In Israel, legislation is pending to fix CEO pay as a multiple of the pay of shop floor workers.

Management gurus are competing with each other to condemn the system. Roger Martin of the University of Toronto argues that the era of outsized rewards for talent needs to come to an end for the good of the system. Michael Dorff of Southwestern Law School argues that “the best solution would be to abandon the experiment with performance pay and go back to compensating CEOs as companies did before the 1970s: primarily using cash salaries.” But the most wide-ranging condemnation of pay-for-performance is Daniel Pink’s “Drive: The Surprising Truth About What Motivates Us.”

Pink does not limit himself to condemning outsized executive salaries but also launches a frontal assault on performance-related pay in general as incompatible with modern scientific knowledge and repugnant to sensible managerial practices.

Pink argues that performance-related pay might have had a role in the world of manual labor: moving bits of pig iron, for example. But in an economy where human beings focus on brainwork, it can be counterproductive. Performance-related pay is more likely to dampen workers’ motivations and kill their creativity than it is to force them to put their noses to the grindstone.

So how should firms motivate employees? Pink argues that companies need to return to the wisdom of the likes of Elton Mayo and Mary Parker Follet: give people more control over their own lives and thus allow them to draw on their deep inner wells of diligence and drive.

The problem with Pink’s argument is that it goes too far in the opposite direction. It is one thing to point to the downside of Lehman Brothers’ incentive packages, quite another to discredit pay-for-performance in general. Pink argues that “the scientific literature” is hostile to the use of performance-related pay. But four reviews of research on the subject from the 1980s onward have all come to the same conclusion: that pay-for-performance can increase productivity dramatically.

John Abowd examined data on more than 16,000 managers at 250 large corporations to determine whether conditioning pay on performance enhances corporate results. He found that a performance bonus of 10 percent was associated with an extra 4- to 12-percent growth in share price.

The Lehman example is in fact decisive proof that extrinsic incentives have a dramatic impact on employee behavior — though not necessarily a positive impact. The problem is not that external incentives don’t work but that they are so powerful that we have to make sure that we get them right.

Pink paints an idealistic portrait of companies like Google and other tech enterprises that supposedly rely on “extrinsic motivation.” But in fact tech companies are some of the most ruthless capitalists in business: They not only pay high salaries to recruit stars but also obsessively measure their employees to match pay with performance.

Pink’s argument is particularly weak when it comes to recruitment and retention. Linking pay to performance is not just a short-term tactic to boost motivation in any given year. It is also a long-term strategy to recruit a pool of high-quality workers and then to keep them hanging around in hope of promotion. Edward Lazear and Sherwin Rosen have demonstrated that one of the most important functions of high CEO pay is to create a “tournament” between high-powered employees who see themselves as future CEOs and are willing to stay around for years, taking on demanding assignments, in order to end up at the top.

A few organizations are skilled at using intrinsic rewards to attract and motivate employees: N.G.O.s such as Doctors Without Borders or Heifer International are master motivators. But in general the record of organizations that eschew pay for performance is dismal. The U.S. teaching profession has failed for decades to compete with other professions in attracting high-quality employees because it insists on a lock-step system in which everybody advances by seniority and nobody can be fired.

In the 1980s, British universities were dead set against the Thatcher government’s plans to introduce performance-related pay. Surely university teaching was a vocation, they argued. And surely it was impossible to measure the quality of scholarship.

But this resistance proved disastrous. High-flying professors abandoned British universities for better-paying American ones. British universities were in danger of becoming rest homes for academics who could not get jobs elsewhere until they began importing America’s more ruthless  methods. Britain now competes well with the U.S. in university league tables.

The world’s brightest students gravitate to organizations that make extensive use of performance-related rewards, such as partnerships and share options. This is just as true of the creative industries, which are supposed to be strongholds of intrinsic rewards, as it is of finance and management consulting. Creative clusters such as Hollywood and London’s Soho are hotbeds of payment by results. Writers employ agents to get the highest possible advances. Actors negotiate profit shares.

There are good reasons why the likes of Jack Welch should be paid their billions: He generated huge returns for General Electric shareholders and kept G.E. growing at a time when its biggest rival, Westinghouse, went from disaster to disaster. There are also good reasons why the corporate sector should be free to pay talented employees whatever it takes to recruit them. If governments impose restrictions on corporate pay, then ambitious people may simply gravitate to other activities that pay more, such as venture capital, private equity, private companies or consulting.

But none of this means that we should throw up our hands and simply accept the practices that were prevalent in Enron and Lehman Brothers. It is precisely because performance-related rewards are such powerful motivators that we need to be careful to make sure that the rewards do not encourage short-term or otherwise destructive behavior. One solution is to reward employees with stock options that can only be traded in after a period of time—say 10 years after they leave their jobs. Another is to balance collective with individual rewards to ensure that managers look after the good of the organization.

It is perhaps time to invoke an old-fashioned principle in the fashion-driven world of incentive-theory: the Golden Mean. The problem with theories of incentives is that they tend to sacrifice common sense on the altar of philosophical rigor. People who think that human beings are basically selfish embrace highly leveraged financial incentives. People who think that they are basically altruistic embrace softer ideas. But human beings are a mixture of both — and incentive schemes should be calibrated to reflect that mixture. Incentives need to be geared carefully to circumstances: People who do boring jobs need more financial incentives than people who do more interesting ones. People who work in teams need to have more collective rewards than lone wolves. Incentives also need to take into account cultural circumstances: Americans are notoriously motivated by money, for example, but Indians are driven more by status and impressive-sounding corporate titles.

Frederick Taylor was guilty of all sorts of mistakes during his fireside chats, from racism — all those jokes about mispronounced words —  to his mania for material incentives. But his biggest error was to argue that there is “one best system” for motivating people. The time for management monomania has long passed. We need to allow a thousand incentive systems to bloom.

Download the PDF