I recently read Drive by Dan Pink, and the most surprising thing for me was how rewards can lead to decreased performance and intrinsic motivation. This is counterintuitive - usually, you think of rewards as something that should increase people’s motivation. However, Pink recites a couple of fascinating studies and findings.
One example that made a big impression on me is the experiment from the 1970s where psychologist Edward Deci and colleagues rewarded pre-school children for drawing pictures. First, they identified children who liked drawing pictures in their free-play time. Then, they divided them into three groups to test how rewards would affect their motivation:
- The first group was told that, if they drew a nice picture during their next free-study time, they would get a reward (“expected reward group”).
- The second group was told nothing, but received a reward after they had drawn a picture (“unexpected reward group”).
- The third group was a control group. They were told nothing, and received no reward (“no reward group”).
The reward was not much, by the way. It was a nice-looking “Good Player” certificate featuring the child’s name.
Two weeks later…
Two weeks later, the children were observed again during free-play time, and the result was quite surprising. The kids from the “unexpected reward” and “no reward” groups showed the same joy and enthusiasm for drawing, and drew for just as long as before. However, children from the “expected reward” group showed significantly less interest, and spent less time drawing. Even after two weeks, the “expected reward” children had less intrinsic motivation for drawing than before.
When these results were published - and later confirmed with adults -, they were very controversial, because they went against the predominant and common belief that rewards increase motivation and performance. Researchers, especially economists, had so far largely believed in carrots and sticks: Give the horse a carrot, and it will run. Give it a bigger carrot, and it will run faster. Hit it with a stick to keep it from doing something that is undesirable.
However, the explanation for the phenomenon observed in the drawing experiment is quite logical: The reward agreement - “If you draw a picture, then you will receive a certificate” - took away the children’s autonomy. They were no longer masters of their own free play time, but now had an obligation: They had agreed to draw a picture in exchange for a reward. They could not back out and change their mind mid-way. They were no longer in control, but they were being controlled by somebody else. In other words: The agreement had turned play into work.
Note that it was not the reward itself that came with such high cost, since the children from the “unexpected reward” group kept drawing pictures as readily as before. It was the contingent reward that proved so lethal to intrinsic motivation. Pink calls it an “if-then” reward, as opposed to the unexpected “now-that” reward which leaves intrinsic motivation intact.
I wrote about things that managers can learn from raising children, and I could well add this one: Reward with extreme care. The study described above shows the potentially disastrous effects on intrinsic motivation. Moreover, as education experts Rudolf Dreikurs and Vicki Stolz illustrate in their classic
Children: The Challenge
Over the years, reality will kick in, the rewards will not keep coming as readily as the now-teenager would wish, and she will grow discontent and resentful, constantly feeling she does not get what she deserves. By paying them or otherwise rewarding children to help in the household, “if-then” style, parents do great damage to their children, the relationship between themselves and the children, and to society as a whole.
Less intrinsic motivation
Back to the scientists: Deci and colleagues were able to confirm their findings across a wide range of institutions, situations, and age groups, so that they were able to state:
“Careful considerations of reward effects reported in 128 experiments lead to the conclusion that tangible rewards tend to have a substantially negative effect on intrinsic motivation.”
Rewards focus on short-term control, and this can be justified in exceptional situations. However, if applied regularly, they can do considerable long-term damage. The negative effect on intrinsic motivation might be the most severe consequence of contingent rewards, but it is not the only one.
Less creativity
Another interesting scientific finding is that contingent rewards reduce people’s creativity. This has been demonstrated in the lab as well as in real life, with artists producing better, more creative work for themselves - with no financial or other reward in sight - than for clients who were willing to pay for the work.
When a reward is promised, people’s focus partly shifts from the activity itself to the reward, which narrows their thinking. In today’s workplaces in knowledge intensive industries, where creativity is a requirement, this can lead to lower quality of work. Rewards make us race ahead, but keep us from looking left and right.
Lower performance
Ok, let’s forget about creativity, and focus on performance instead. But even here, scientific experiments question the effectiveness of financial incentives. A research team asked people in rural India to play games like hitting a target with a tennis ball (motor skills), or recalling a string of digits (concentration). Different groups were promised and - if successful - paid different rewards for the same tasks, with the highest rewards amounting to nearly five months’ pay. Guess what: The candidates who had the highest financial incentives came up with the worst results in almost every discipline.
Scholars at the London School of Economics second this finding after having reviewed 51 studies of corporate pay-for-performance plans:
“We find that financial incentives can result in a negative impact on overall performance.”
Less good behaviour
If people do something good out of their own motivation and free will, rewards should get even more people to follow their example, right? Not necessarily. In an experiment in Sweden, prospective blood donors were, again, divided into different groups. Some were told that, if they decided to donate blood, they would do so voluntarily. Others were promised a moderate amount of money for their donation.
As you suspect by now, the “non-profit” group turned out to be the more diligent blood donors. Donating blood is something that you do because you want to do something good and leave a mark in the world. It is a selfless act that makes you feel good about yourself. However, if you add money to the equation, a lot of this goes away. Suddenly, it is not a selfless act any more. Putting it in the worst light possible, you could even say that you profit from the emergency situation that others are in, or that you sell a piece of yourself. These thoughts are not nearly as nice as thinking of yourself as a person doing something good and potentially saving somebody’s life, so a lot of people would rather not be paid for donating blood.
Conclusion for the workplace: If somebody is ready to take one for the team, don’t attach a contingent reward to it. They might be less willing to do it, because it makes them look like a mercenary instead of a team player.
Conclusion
If an organization hands out bonuses, it should do so with extreme care. Avoid “if-then” rewards, and rather use “now-that” rewards that are unexpected. The best way to avoid damage to people’s intrinsic motivation is to take the issue of money off the table by paying them fairly and sufficiently in the first place. Money works mainly as a hygiene factor: If there is too little of it, this can result in dissatisfaction. However, once there is enough of it, handing out more will not result in a significant increase in satisfaction or motivation.
If this requirement is met, and you want to reward somebody, remember that the important, lasting thing about a reward is not the money. It is the recognition and the appreciation expressed by it, along with valuable feedback that will help the person grow.
Time investment
It took me about 4.5h in total to compose this, including going back and forth from my writing to re-reading things and finding the best quotes.
I'm Tom Bartel, Germany-based software developer, engineering manager, speaker, and human communication geek. More