Every quarter a familiar ritual plays out in offices and classrooms alike. A manager announces a spot bonus for hitting a target. A teacher promises pizza to the class that reads the most books. A corporate learning platform hands out badges and streaks for finishing modules. The logic is intuitive and almost never questioned: pay people for the behavior you want, and you will get more of it. The evidence says the opposite is often true, and it has said so for a quarter of a century.
What 128 experiments actually showed
In 1999 Edward Deci, Richard Koestner and Richard Ryan published a meta-analysis of 128 controlled experiments on how extrinsic rewards affect intrinsic motivation, the drive to do something because it is interesting or satisfying in itself. Their conclusion was stark. Tangible rewards, whether offered for merely engaging with a task, for completing it, or for reaching a performance standard, significantly undermined intrinsic motivation as measured by whether people freely chose to keep doing the activity once the reward was withdrawn (Deci, Koestner and Ryan, 1999). The effect was strongest with children and on tasks people already found interesting, exactly the situations in which you would least want to erode enthusiasm.
There was one important exception. Verbal rewards and positive feedback, the kind that carry information about competence without controlling behavior, tended to strengthen intrinsic motivation rather than weaken it (Deci, Koestner and Ryan, 1999). The mechanism matters more than the money. A reward that feels like control shifts a person's sense of why they are acting, from "I do this because I want to" toward "I do this because I am paid to." Remove the payment and the reason leaves with it.
Honesty requires a caveat the authors themselves would accept. These were mostly laboratory experiments run between the 1970s and the 1990s, and the finding drew a well-known methodological challenge from Cameron and Pierce. The undermining effect is not a universal law of behavior, and its size depends on the type of reward and the population studied. But it is robust enough to have become one of the empirical foundations of self-determination theory, the framework Ryan and Deci later carried into education and the workplace. Whatever the boundary conditions, the direction of the effect for tangible, controlling rewards has held up across decades of replication and debate.
Why incentive design keeps backfiring
The practical trap is that rewards work brilliantly in the short run and quietly corrode the source of effort in the long run. A gamified sales dashboard lifts activity this month. Then the leaderboard becomes the point, people optimize for the metric rather than the customer, and the intrinsic interest that once made the work absorbing has been trained out of them. The same arc plays out in classrooms that pay children to read. Reading rates rise while the reward lasts, and the appetite for reading, the thing the program was supposedly protecting, comes out thinner than it went in. Ryan and Deci (2020) argue that durable motivation comes not from bribing behavior but from supporting three basic psychological needs: autonomy, competence and relatedness. Well internalized motivation, the kind where a person genuinely endorses what they are doing, predicts better outcomes across every level of education and across cultures. Coercive rewards push in the wrong direction because they undermine the sense of autonomy that internalization depends on.
The constructive version of this evidence is not "stop paying people." It is "stop using pay to manufacture engagement." A meta-analysis by Wang and colleagues (2024) synthesized 36 intervention studies covering 11,976 participants and found that programs built to support all three psychological needs improved student motivation. Notably, this was the first systematic review in more than a decade to test interventions across autonomy, competence and relatedness together rather than autonomy alone. The lesson transfers directly to work. The interventions that raise real motivation are structural, giving people meaningful choice, useful feedback, and a sense of belonging, not transactional.
The read for anyone who manages effort
Most incentive schemes are answering a question the research settled long ago. If a task is dull and nobody would do it without payment, a reward is honest and harmless, because there is no intrinsic motivation to protect. The danger sits precisely where organizations least expect it: on the interesting, high-judgment work you most want people to care about. Bolt a bonus onto that, and you risk teaching your best people that the work is only worth doing for the payout.
The alternative is unglamorous and cheap. Deci, Koestner and Ryan (1999) found that informational feedback, praise that tells someone their work is good and why, enhances rather than erodes the drive to continue. Ryan and Deci (2020) would add autonomy over how the work gets done and a team worth belonging to. None of that appears on a compensation spreadsheet, which is part of why it is so consistently overlooked. Firms that ration recognition and lean on cash incentives are buying compliance and, without noticing, selling off the curiosity that made the work good in the first place.
Sources
- Deci, Koestner and Ryan (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Synthesis of 128 experiments showing tangible rewards undermine intrinsic motivation while verbal feedback enhances it, the empirical core of the argument. https://doi.org/10.1037/0033-2909.125.6.627
- Ryan and Deci (2020). Intrinsic and extrinsic motivation from a self-determination theory perspective. Framework arguing durable motivation rests on autonomy, competence and relatedness rather than coercive rewards. https://doi.org/10.1016/j.cedpsych.2020.101860
- Wang et al. (2024). SDT-based educational interventions: a meta-analysis. 36 studies and 11,976 participants showing need-supportive interventions raise motivation, supporting the structural alternative to incentives. https://www.sciencedirect.com/science/article/abs/pii/S0023969024000572