Some years ago, Men's Wearhouse founder George Zimmer came to a class I taught to discuss a case study I had written on his company.
Zimmer commented on a bonus program I had written about where each store employee (except the store manager) would receive $20 if the store met its "good" sales target for the month, and $40 if it met its "excellent" sales goal. My students thought these amounts were quite small, but Zimmer thought the incentives were perfectly sized. They were large enough to provide some recognition of store achievement, he said, but more importantly, the payouts gave people a chance to celebrate success together without being large enough to distort people's behaviors.
Zimmer's insight that, in the case of incentives, less is often better than more is too infrequently embraced by leaders who instead seek to use substantial rewards to fundamentally channel behavior.
HR Managers and c-suite executives would do well to learn from Zimmer's wisdom. While most employees today assume incentives will be part of their job, how large they are and how they are presented can substantially impact an organization.
Incentives Can Undermine or Crowd Out Intrinsic Motivation
Beginning in the 1970s, studies in psychology found that providing people rewards—extrinsic incentives—could undermine intrinsic motivation for engaging in inherently interesting tasks. One theory suggested that people found incentives controlling, and rebelled against attempts to control their behavior. Another perspective suggested that people interpreted incentives as signaling that a task was inherently unpleasant, reducing their interest in doing it. The takeaway? Incentives have the potential to reduce people's motivation and interest in tasks.
Even economists, who have traditionally looked more favorably on incentives, have also argued that incentives can backfire. They argue that providing extrinsic incentives "crowds out" intrinsic interests in doing something. Consequently, incentives can backfire, and make it less likely that people will do what the incentives want them to do. For instance, one study observed that parents were more often late in picking up their children from a day care center when a fine was imposed, while yet other researchers observed that volunteers who were paid a small amount worked fewer hours than volunteers who were not paid. Here, the evidence suggests that using incentives to drive desired behavior may not work.
Consistent with the idea that smaller is better, smaller incentives will be less likely to crowd out or reduce intrinsic motivation because smaller incentives are less psychologically prominent and salient. For organizations concerned about maintaining intrinsic task motivation—which is probably most workplaces—the crowding out and undermining research provides one more reason to be cautious in the use of incentives.
Incentives Drive Behavior, But in Often Unanticipated—and Counterproductive—Ways
As Bob Sutton and I pointed out in our book on evidence-based management, a huge problem with incentives is that they are too effective at influencing behaviors. And most people and companies aren't great at anticipating how behaviors will change in response to incentives. There are enough examples of this to fill a book—or maybe several.
In 2018, William Dudley, CEO of the Federal Reserve Bank of New York, noted that "misaligned incentives contributed greatly to the 2008 financial crisis." In the scenario that Dudley is referring to, many mortgage brokers were compensated for the number of loans they made—not necessarily for making sound loans that would be repaid. And many of the incentives for mortgage brokers for senior financial industry executives were short-term rewards. Meanwhile, the assets being created (the loans) and the financial results were inherently longer-term. The time horizon on incentives needs to match the time horizon of the results being affective. Simply put, short-term incentives aren't going to be very good for creating long-term results.
How to Make Incentives Work
Based on extensive empirical evidence, there are some simple but important implications for implementing incentives in ways that aren't likely to cause misbehavior that adversely affects organizations.
First, and most importantly, keep incentives small enough to not overly influence behavior. That may seem counterintuitive—many workplaces implement incentives precisely to influence behavior—but, as noted, people are often quite bad at predicting the ways in which incentives may drive behavior.
Second, spend time trying to anticipate how people could achieve the goals signaled by incentives in ways that are harmful to the organization's interests and try to put up various guardrails to detect and deter such behavior. One way to do this, is to monitor how rewards are being received and the behaviors associated with them based on consistent conversations with employees who are benefiting from the rewards program.
And third, if incentives are driving bad behavior, don't do what many workplaces do, which is to try and solve an incentive-based problem by implementing even more incentives. Many companies try to use incentives to substitute for leadership (coaching and feedback) or a strong, positive organizational culture. As research going back decades from places like shows, leader behavior matters a lot in motivating performance and reducing turnover. Incentives are a poor substitute.
In the case of incentives, the inescapable conclusion is that less—less reliance, less use, less magnitude—is most often better than more.
Vill du lära dig mer? Utforska våra produkter, kundberättelser och det senaste i branschen.
RPS vill “göra det komplexa enkelt”
RPS är ett globalt tjänsteföretag som specialiserar sig på naturliga och byggda miljöer med närvaro i fler än 125 länder.Företaget har en bred medarbetarbas, från konsulter till akademiker, forskare, arkitekter och ingenjörer och alla kategorier däremellan. Fram till 2018 var HR decentraliserat utan någon globalt övergripande funktion eller People Director och man arbetade med manuella system och processer.