When employees take the initiative at work—whether it’s to advance breakthrough innovation or simply find a better way get a job done—organizations of all sizes can benefit. But taking initiative also elevates an employee’s chance of failure, and many hesitate to accept that risk.
What, then, is the best way to motivate employees not only to work hard but also take the kinds of risks that could really move the needle for an organization?
That question was the launching pad for new research by Jeroen Swinkels, a professor of management and strategy at the Kellogg School, and coauthor Hector Chade, a professor of economics at Arizona State University.
Through mathematical modeling and analysis, they found that the ideal approach favors rewarding high-value results in grand fashion, while also being gracious with failure.
“The way you get people to work hard is by having big payoffs for big outcomes,” Swinkels says. “And firms that are also good at managing the failure process so that it is graceful and survivable, and good at taking advantage of the projects that then succeed, are firms that are going to thrive.”
The moral-hazard problem
Employers have much to consider when designing compensation contracts with their employees.
After all, companies generally want their employees to work hard, while employees may prefer to shirk. A contract that simply pays employees for showing up presents what economists refer to as moral hazard, where employees aren’t exposed to the consequences of a job poorly done. Hence the allure of redirecting some of this risk back to employees by rewarding their success and punishing their failure.
But this raises a problem from the employee’s perspective, because there are times when their projects fail even when they work really hard.
In light of this risk, companies that choose to punish failure often need to pay quite well to draw in employees.
“The classic model of moral hazard, which has been hugely influential and has had enormous impact, is one where the firm faces this big trade-off,” Swinkels says, “that, if I as an employer put you under risk, I may get you to work hard, but your average compensation has to be very high.”
With this classic moral-hazard model in mind, companies “are going to punish low output, be reasonably nice to employees with middle output, and be very nice to employees with high output,” he says.
A more complex scenario
The real world, however, can be more complex than the classic moral-hazard problem allows.
Take, for instance, the case of Mary Barra, CEO of General Motors. If the only thing GM wants from her is to work hard, then it will punish or possibly fire her if the company stock is low, be nice to her if the stock is at an acceptable level, and make her rich if the stock is high.
“The problem is that for neither Mary Barra nor for many employees is that quite the right problem that the organization is trying to solve,” Swinkels says.
That is, many companies today want their employees not just to work hard but also to take initiative on high-stakes work that has the potential to deliver significant value to the company. In Barra’s case, that might mean transitioning GM from gas to electric vehicles or unlocking a new market—projects that could catapult the company’s stock price to new highs. Yet these same projects, if not successful, could cause the stock price to crater and potentially cost Barra her job.
“The way you get people to work hard is by having big payoffs for big outcomes.”
Jeroen Swinkels
Reward the extremes
To better understand how companies can motivate employees to both work hard and take initiative on high-risk, high-reward projects, Swinkels and Chade turned to mathematical modeling.
They set up a model that simulates this multidimensional version of the moral-hazard problem. But unlike past research, their new model also takes into account the possibility that employees can “play it safe” by choosing middle-of-the-road projects that are unlikely to flop but also unlikely to have a big impact on the organization.
Through a formal analysis of the model, the economists determined that the key to an optimal employer–employee contract is an incentive structure that—compared with the standard prediction—gives steeper rewards for high-value outcomes but less punishment for low-value outcomes.
In other words, an organization hoping to induce both hard work and initiative should reward employees generously when they succeed and be forgiving when they don’t. And the incentive for high-value outcomes should be so compelling that it lures employees away from mediocre work.
This is different from the incentive patterns that emerge under a classic moral-hazard model, where failure is punished harshly, and mediocrity is more handsomely rewarded.
“Compared with the standard model, you want to reward the extremes,” Swinkels says. “You punish failure less than you used to. You reward big successes more than you used to. But you also have to make it so that just-adequate performance is a little less comfortable and not as rewarding for employees as it used to be.”
Impact matters
These dynamics are reflected in a number of work environments where success is rewarded steeply while failure is barely punished, including a scenario very close to home for Swinkels.
“If all the university was concerned about is how hard I work, then they can very effectively get me to work hard by punishing failure,” Swinkels explains. “But the university doesn’t want just that; it wants me to pursue important, high-risk research projects and work very hard on them.”
The university holds big successes in very high regard. The sort of work that wins Nobel Prizes or gains admittance to prestigious societies has a serious positive impact for the university. And it can be worth an extraordinary amount—billions of dollars in the case of Lyrica, a blockbuster drug that came out of Northwestern research, Swinkels says. If the researcher behind Lyrica had written several mediocre papers instead, that would have been of little value to the university and society more broadly.
“And so from the institution’s point of view, the difference between writing nothing this year and writing something mediocre doesn’t matter that much,” he adds. “But the difference between writing something mediocre and writing something that has serious intellectual impact matters a lot.”
And the institution’s incentive structure reflects these values. For junior faculty, low research output and mediocre output are rewarded similarly—and neither leads to tenure. The difference between mediocre output and writing important work, however, is that you get tenure. For tenured faculty, who are somewhat protected from the consequences of unsuccessful research, high-impact work can lead to substantial career success and rewards.
