Notes on Why Incentive Plans Cannot Work (Alfie Kohn, HBR 1993)
Overview
Alfie Kohn argues that incentive plans are fundamentally flawed as a means to motivate employees. Despite widespread belief in rewards, a large and growing body of evidence shows that rewards typically undermine the very processes they aim to enhance. The problem is not just the details of how incentives are implemented but the underlying psychological assumptions that ground all such plans.
Rewards tend to produce temporary compliance rather than lasting change in attitudes or behavior. When rewards run out, old behaviors tend to return; incentives rarely foster enduring commitment to values or long-term goals.
The article surveys laboratory and real-world studies across settings (workplaces, classrooms, etc.) and highlights that according to these studies, extrinsic motivators often fail to improve, or even harm, performance, especially on tasks requiring higher cognitive effort and creativity.
What "Work" Means in Incentives
Two meanings of "work":
Short-term, observable compliance with a rewarded behavior.
Long-term, enduring attitude and value alignment that sustains behavior without ongoing external prompts.
Key distinction: extrinsic motivators (rewards) do not create lasting commitment to a value or action; they merely change what people do temporarily.
Empirical pattern: many studies find that people who expect a reward for a task underperform relative to those who expect no reward, across a range of tasks from memory to creative problem solving to design.
When cognitive demands are higher or tasks are open-ended, the negative impact of rewards on performance tends to be stronger.
The researchers themselves are often surprised by these findings; rewards do not reliably improve outcomes and can hinder genuine engagement.
Some scholars caution that linking pay to performance may raise questions about whether the performance measure is appropriate or whether compensation truly drives the intended behavior.
A key implication for managers: the mere existence of financial incentives does not guarantee higher performance and may mask deeper organizational issues.
Empirical Evidence and Key Studies
General finding: rewards produce temporary compliance, not durable change.
Example domains include weight loss, smoking cessation, seat belt use, and prosocial behavior in children; incentives for these behaviors are often less effective than alternative approaches and can be worse than doing nothing.
Meta-analytic and large-scale reviews:
A major meta-analysis examined roughly 330 comparisons across 98 studies. It found no significant overall effect of financial incentives on productivity when data were pooled and varied across studies. Additionally, incentives were not clearly related to employee absenteeism or turnover. By contrast, training and goal-setting programs showed stronger positive effects on productivity.
Welders case (Rothe):
When an incentive system was removed for welders at a Midwestern plant, production initially declined but then rose to, and eventually exceeded, prior levels. This long-term data set suggests that incentives were not causal in sustaining higher output; their removal did not lead to permanent losses.
Rich and Larson (1982) study on shareholder value:
Interviews and proxy statements across 90 major U.S. companies showed no difference in shareholder returns between firms with incentive plans for top executives and those without.
Jenkins (1986) synthesis on financial incentives and performance:
Tracked 28 studies measuring the impact of financial incentives on performance. Found that 16 (≈ 57 ext{%}) showed a positive effect on performance.
However, all of the positive effects were on quantitative performance (more output, faster production). Only 5 of the studies examined the quality of performance, and none showed benefits for quality.
Overall implication from these analyses: while some incentives appear to boost short-run quantity, there is little evidence that pay-for-performance reliably improves meaningful, high-quality work over the long term.
The Six-Point Framework: The True Costs of Incentive Programs
Kohn outlines six core costs of relying on incentive plans. Each point includes implications for motivation, relationships, and organizational learning.
1. "Pay is not a motivator."
Deming’s famous claim is invoked: money buys goods, and people say they care about money, but when asked directly, pay ranks only around the 5^{th} or 6^{th} in what people truly care about.
Even if pay matters, higher pay does not automatically translate into better work or sustained motivation. Herzberg argued that money is not a reliable or sufficient driver of motivation; increasing pay does not necessarily increase satisfaction or motivation in the long run.
Key implication: the assumption that more pay will yield proportionally better performance is not supported by the evidence.
2. Rewards punish. (Rewards = control)
Rewards function as coercive instruments, similar to punishment, by signaling that behavior is contingent on a bribe.
Herzberg’s concept of a KITA (kick in the pants) is cited to illustrate that such approaches may produce movement but not genuine motivation.
The central claim: rewards and punishments are two sides of the same coin; both are manipulative and can reduce intrinsic interest in the task.
When a reward is contingent on behavior, subordinates may perceive increased control by the manager, which over time becomes punitive in effect.
The absence of a reward can feel like punishment, magnifying demotivation when rewards are expected but not received.
3. Rewards rupture relationships. (Competition and surveillance harms cooperation)
Incentives and related performance appraisal systems tend to undermine cooperation and teamwork, which are essential for quality and organizational learning.
Peter Scholtes warned that incentive systems “pressurize the system for individual gain” and hinder collective improvement.
Publicizing rewards can make coworkers view one another as barriers to personal gain, eroding collaboration.
Supervisor-subordinate dynamics can deteriorate: rewards may lead employees to withhold information, feign competence, or flatter management to secure bonuses rather than seek help or report problems.
The broader risk: a culture of competition over collaboration damages organizational performance, particularly in contexts that rely on joint problem-solving.
4. Rewards ignore reasons. (Incentives address symptoms, not causes)
Incentives do not diagnose or fix underlying problems in the workplace.
Root causes to consider include: inadequate training, insufficient feedback and social support, insufficient opportunities for self-determination, rigid hierarchies, and burnout.
Managers may substitute pay-for-performance for genuine good management (e.g., providing feedback, enabling collaboration, offering resources).
Evidence from Rothe and others suggests that leadership and supportive management decline in the presence of incentive systems, reducing overall effectiveness.
Surveys (e.g., American Productivity Center) indicate little evidence of increased active employee involvement in organizations that rely on small-group incentive plans.
A number of scholars argue that pay-for-performance can impede managers’ ability to lead and develop teams.
5. Rewards discourage risk-taking. (Creativity is the first casualty)
When rewarded, people focus on what is explicitly rewarded and often avoid risk, exploration, or innovative approaches.
Neurological and behavioral findings: people tend to select easier tasks as the payment for success increases.
The phenomenon is sometimes described as a regression to the mean or a narrowing of task scope to what is easily measurable.
This risk-averse shift undermines innovation and long-term organizational learning.
Related observations: leaders and workers may adjust task strategies to ensure predictability and simple metrics, rather than pushing into uncertain, exploratory work.
6. Rewards undermine interest. (Intrinsic motivation is lost)
Intrinsic motivation often drives high-quality, passionate work; extrinsic incentives can crowd out this inner drive.
Deci and Ryan’s foundational work showed that contingent payments tend to undermine intrinsic motivation, especially for interesting or challenging tasks.
Freedman et al. (1992) found that the larger the incentive, the more negatively people viewed the activity for which the bonus was received.
The broader mechanism: incentives convey the message that the activity is something one would only do for a reward, reducing self-direction and interest.
Conclusion: any incentive or pay-for-performance system tends to reduce enthusiasm and commitment to excellence over time.
Dangerous Assumptions and Misconceptions (Beyond the Six Points)
Intrinsic vs extrinsic motivation is often misunderstood outside psychology; many managers assume they can simply add intrinsic and extrinsic motivators and get the best of both worlds.
In practice, incentives can create a self-fulfilling prophecy: reliance on rewards signals that motivation is externally driven and should be rewarded, which reduces intrinsic interest and engagement.
Barry Schwartz notes that behaviorist explanations may describe workplace dynamics but also help transform work into a context dominated by behaviorism when incentives are heavily used.
More broadly, managers who insist that a job cannot be done without rewards have failed to justify the behavioral manipulation with a compelling argument for why the reward drive is necessary.
The article ends with a strong caution against bribes in the workplace as an effective long-term strategy.
Connections to Theory and Real-World Relevance
The discussion links to classic and contemporary theories of motivation:
Intrinsic vs extrinsic motivation (Deci & Ryan).
Herzberg’s two-factor theory and the critique of monetary satisfiers.
Behaviorist ideas (Skinner) and their limits in complex, creative, and knowledge-work tasks.
Real-world relevance includes:
The popularity of executive compensation schemes (stock options, performance bonuses) and their mixed or negative associations with performance quality.
The use of team-based or profit-sharing plans, and the potential harm to teamwork and learning.
The emphasis on training, feedback, goal-setting, and employees’ sense of autonomy as more robust drivers of performance.
Implications for Practice (Alternative Approaches)
Focus on intrinsic motivators by enabling autonomy, competence, and relatedness.
Invest in meaningful feedback, coaching, social support, and opportunities for employee development.
Emphasize teamwork and shared goals over individual rankings and bonuses.
Use training and goal-setting as more effective, longer-lasting levers of performance than pay-for-performance schemes.
When incentives are used, design them with caution, ensuring they do not undermine intrinsic motivation or social cooperation and that they address root causes of performance gaps.
Notable Takeaways and Quotations to Remember
Rewards typically foster temporary compliance rather than lasting change: revenue in the short term does not guarantee long-term growth or learning.
The larger the incentive, the more negatively people may view the activity itself. ext{larger incentive}
ightarrow ext{more negative evaluation}The six-point framework highlights the core costs: manipulation, damaged relationships, ignored root causes, stifled risk-taking, and eroded intrinsic interest.
Rothe’s welders case: removing incentives can, over time, sustain or even improve output relative to the incentive period.
Meta-analytic findings: 330 comparisons across 98 studies showed no robust overall effect of financial incentives on productivity; training and goal-setting are often more effective.
Quick References (Key Numbers)
Meta-analysis scope: 330 comparisons, across 98 studies.
Positive effects on performance: 16 of the 98 studies (≈ 57 ext{%}).
Quality-related findings: only 5 studies addressed quality; none showed quality benefits.
Welders case: data from a long-term observation of production following incentive removal.
Executive compensation comparisons: 90 companies studied for shareholder return differences.
Two decades of research cited in the discussion of intrinsic motivation and rewards (Deci & Ryan; Freedman et al.).
Overall time frame: discussions span roughly 30 years of research up to the 1990s, with some references dating back to the 1960s–1970s.
Summary
The central claim is that incentive plans are not a reliable, durable method for improving performance or fostering genuine engagement. They tend to produce short-term compliance, undermine intrinsic motivation, damage collaboration and leadership quality, and fail to address root causes of performance problems. Alternative approaches that emphasize intrinsic motivation, meaningful feedback, training, and supportive management are more likely to yield enduring improvements in performance and innovation.