Do bankers deserve big bonuses? Economists will tell you that bonuses improve employee productivity by rewarding good work. But did the large performance-based payments given to Wall Street securities traders, for example, really steer them to better choices during the run-up to the recent financial crisis? What about financial advisors who base their fees on a percentage of the assets they manage? Does the incentive to grow AUM make them perform any better?
The answer to all these questions is “probably not,” argues Duke behavioral economics professor Dan Ariely, the author of a new book, The Upside of Irrationality,which expands upon themes he first laid out in his 2008 book, Predictably Irrational.
In fact, Ariely writes, studies suggest that instead of improving performance by increasing motivation, bonuses and other forms of performance-based pay – AUM-based fees included – can actually hurt performance by taking employees' minds off their work. In essence, Ariely argues that instead of focusing on the task at hand, workers faced with the prospect of a big performance-based payoff daydream about how they will spend it. They also fret over the possibility of losing it if they screw up, which, thanks to the distraction of the looming pay day, they sometimes actually do.
The big bonuses paid out to securities traders in the lead-up to the financial crisis, in other words, may have made their financial decision-making worse unnecessarily distracting them from day-to-day trading decisions. Similarly, Ariely noted in an interview with melast week, financial advisors may actually perform better if they charge a fixed fee – say in the range of $5,000 to $10,000 per year – than if they are paid as a percentage of AUM.
Ariely's arguments are rooted in psychological experiments he has conducted. In one experiment, Ariely asked participants living in rural villages outside the city of Madurai, India to play a number of simple games, including darts and an electronic memory toy called Simon, and he promised to pay the participants a variable amount of money based on their performance. Participants who were promised up to 240 rupees based on their ability to successfully complete the games – about two week's pay for an average worker in that part of India – performed fairly well, achieving high levels of performance about 40 percent of the time. Participants who were promised a small fortune of up to 10 times that amount, however, performed considerably worse, reaching higher levels less than 20 percent of the time. They even earned less in absolute terms than their counterparts who had less money at stake.
The prospect of more money, Ariely argues, distracted the latter group. While participants who played for just two weeks' pay were calm as they carried out their assigned tasks, the participants with more money at stake trembled with fear as they tried to win the simple games.
The bonus experiment illustrates Ariely's larger point: that the rationality assumptions that characterize mainstream economic theory – or "rational economics," to use Ariely's term – don't always apply in the real world.
Indeed, people are not only irrational; they are predictably so. Beyond simply making mistakes, we make the same mistakes over and over again. As Ariely said in his interview with me, consider this: If people were perfectly rational, there would be no need for financial advisors. People would already know the optimal mix of stocks, bonds and other investments needed, for example, to maximize their retirement savings, save up for a new house or accomplish other financial goals. But people, of course, are not perfectly rational when it comes to their finances, and that's where the expertise of advisors comes in.
Furthermore, the best way to identify our foolish tendencies, according to Ariely, is through controlled experiments. Doctors, misled by archaic medical theories that were not scientifically vetted, spent hundreds of years using bloodletting as a treatment for a range of illnesses. It was not until the widespread medical experimentation of the 19th century that doctors let go of their old, erroneous practices. It is time for economists and economic policymakers, Ariely writes, to let go of their field’s own long-held, unproven assumptions.
There are problems with these arguments, however. The first is the ambiguous results that often flow from the sorts of experiments that Ariely advocates. All too often, Ariely's policy advice seems to boil down to finding some happy medium. On the topic of bonuses, for instance, Ariely posits that the effects on performance of financial rewards actually follow a "reverse-U" pattern (roughly, a bell curve) While a big bonus can hurt productivity, in other words, a midsized bonus still works better than a very small bonus or no bonus at all.
The reverse-U, unfortunately, is an ambiguous guide to policy. It is one thing to say "pay your employees some money at bonus time, but not too much," but it is something entirely different to act on that advice if you are a bank executive who must decide how much money to distribute to your securities traders at the end of the year. Ariely urges executives to experiment in order to find the best payment scheme for their companies: to try paying employees on a straight salary basis, for example, or to offer smaller and more frequent bonuses. Ariely also suggests basing bonus payments on performance measures that have been averaged over five years, not just one year, in order to minimize nonproductive stress.
The practical problem with experimenting with these schemes, however, is that unlike some of the tests carried out in Ariely's psychology lab, these decisions have real-world consequences. Say an executive decides to follow Ariely's advice and eliminates employee bonuses entirely, or severely limits them. While that executive could indeed see productivity gains due to reduced stress, if payments are decoupled from productivity too much, workers may think they have nothing to gain by working harder, and productivity may deteriorate as a result. Some of the most productive workers may even decide to leave for another firm; a compensation system that rewards all employees the same amount, regardless of how much (or how little) they produce, after all, will only benefit the least productive among the workforce, while hurting the best performers.
Another problem is the legal and institutional barriers to the kinds of policy experiments that Ariely advocates. In his interview with me, Ariely claimed that lawyers may actually be the biggest barriers to experimentation at corporations. Things like compensation plans, after all, are often protected by contracts, and a change in compensation would often entail a breach of contract, which comes with the threat of a lawsuit. Indeed, Ariely noted, the legal hazards of policy experimentation lead many executives to conclude that finding the most rational possible compensation policy is more trouble than it's worth.
This, in turn, raises a question Ariely never really explores in his work: At what point do the costs of experimentation render the pursuit of the most rational possible solution irrational? Yale political scientist Charles Lindblom addressed this question in his classic 1959 article, "The Science of Muddling Through." While the rational calculation of ideal means to meet one's ends is an indeed a good goal to work towards, Lindblom argued, rationality has its costs. The costs of the constant research, experimentation and development needed to improve policy are often so high they outweigh the benefits of finding a better solution. In these cases, it is actually more efficient to pursue our current course of action, even if a more efficient solution may still lurk undiscovered.
Policy in the real world is thus often a process of "muddling through," of refining what we already have, and of altering our course only occasionally as conditions change.
Lindblom's argument leads one to wonder: Is Dan Ariely missing the point? Is the widespread "irrationality" he observes in the real world not, in fact, failure to find the perfect policy solution, but a rational determination that finding that perfect solution is more trouble than it's worth? As the British World War II-era aphorism that inspired Lindblom goes, "We'll muddle through somehow." Indeed, we may have no choice.
Read more articles by Charlie Curnow
