Systems produce the outcomes they are structured to reward, making the architecture of incentives a more reliable predictor of behavior than stated intentions.
Why systems reliably produce the outcomes they are structured to reward, regardless of stated intentions.
Introduction
If you want to understand why a system behaves the way it does, examine what it rewards. Organizations, markets, and institutions produce outcomes that align with their incentive structures, often regardless of their stated goals, official policies, or the intentions of the individuals within them. This is not a claim about human nature. It is an observation about systems: the incentive structure is a more reliable predictor of behavior than any other single factor.
Incentives are not limited to financial compensation. They include what gets measured, what gets rewarded, what gets punished, what gets attention, and what gets ignored. A company that measures quarterly revenue growth creates incentives to prioritize short-term revenue even if its stated strategy emphasizes long-term value creation. A market that rewards share price appreciation creates incentives to optimize for price rather than for underlying business quality. The formal incentive system and the informal one, which encompasses what actually gets noticed and valued, jointly shape behavior.
Understanding incentive structures as a primary driver of system behavior helps explain why organizations sometimes act against their own stated interests, why well-intentioned policies produce unintended consequences, and why simply changing goals without changing incentive structures rarely changes outcomes.
Core Concept
Incentive structures operate through a straightforward mechanism: people allocate their effort, attention, and creativity toward activities that produce outcomes they value or that the system rewards. When the rewarded activities align with the desired outcomes, the system functions as intended. When they diverge, the system produces what it rewards rather than what it intends. The divergence between intended outcomes and incentivized outcomes is one of the most common sources of organizational dysfunction.
Measurement is the backbone of incentive structures. What gets measured becomes visible; what is visible gets managed; what gets managed gets optimized. This creates a systematic tendency to optimize measured quantities at the expense of unmeasured ones. If customer satisfaction is measured but employee turnover is not, the system will tend to optimize customer satisfaction through means that may increase employee turnover. If short-term revenue is measured but long-term customer relationships are not, short-term revenue will be prioritized.
Incentive misalignment occurs when different participants in the same system are rewarded for different or conflicting outcomes. A salesperson compensated on volume and a credit department rewarded for loan quality face structural tension. A manager evaluated on annual results and a researcher whose work requires multi-year horizons operate under conflicting timeframes. These misalignments are not personality conflicts; they are structural features of the incentive system that produce predictable friction.
Second-order effects of incentive structures are often more significant than first-order effects. A bonus structure designed to reward sales performance may, as a first-order effect, increase sales. As a second-order effect, it may cause salespeople to prioritize easy sales over strategic accounts, to discount heavily to hit targets, or to time transactions to maximize their bonus period rather than to serve customer needs. The second-order effects follow logically from the structure but are often unintended and unexamined.
Structural Patterns
- Metric Optimization — Systems optimize what they measure. When the measured metric is a good proxy for the desired outcome, this produces beneficial behavior. When the metric is a poor proxy, the system optimizes the metric while the desired outcome deteriorates.
- Time Horizon Mismatch — When incentives reward short-term results in contexts where long-term outcomes matter, participants rationally sacrifice long-term value for near-term performance. This is not short-sightedness; it is a rational response to the incentive structure.
- Risk Shifting — Incentive structures that reward upside without penalizing downside encourage risk-taking. Participants who capture gains but can externalize losses have structural motivation to take larger risks than the system's overall interests would suggest.
- Gaming and Goodhart's Law — When a measure becomes a target, it ceases to be a good measure. Participants learn to optimize the metric through means that may not produce the intended outcome. Teaching to the test, hitting quarterly targets through channel stuffing, and meeting safety metrics through selective reporting all follow this pattern.
- Principal-Agent Divergence — When those who make decisions bear different consequences than those affected by the decisions, the incentive structures diverge. Managers whose compensation differs from shareholder returns, fund managers whose fees differ from investor returns, and politicians whose incentives differ from constituent outcomes all exhibit this structural divergence.
- Cascade Effects — Incentive structures at the top of an organization cascade downward, shaping behavior at every level. A CEO evaluated on stock price creates incentives for division heads to prioritize projects with visible near-term results, which creates incentives for managers to defer maintenance and cut research. The cascade amplifies and sometimes distorts the original incentive at each level.
Examples
Healthcare payment structures illustrate incentive effects on systemic behavior. When healthcare providers are paid per procedure, the system produces more procedures. When they are paid per patient outcome, the system produces different behaviors focused on prevention and efficiency. The medical knowledge, technology, and personnel may be identical; the incentive structure changes what the system produces. Shifting from fee-for-service to value-based payment is fundamentally an incentive redesign, not a medical innovation.
Executive compensation structures demonstrate the link between incentive design and corporate behavior. When executives are compensated primarily through stock options, the incentive is to increase stock price over the option period. This can be achieved through genuine value creation, but also through share buybacks, accounting optimization, cost-cutting that sacrifices long-term capability, or risk-taking that happens to produce favorable short-term results. The compensation structure does not dictate which path is taken, but it creates conditions under which all of these paths are rewarded equivalently.
Academic publishing incentives shape what research gets done. Researchers are evaluated primarily on publications in high-impact journals. These journals preferentially publish novel, positive results. This creates structural incentives to pursue novel topics over replication, to report positive findings over negative ones, and to frame results dramatically rather than cautiously. The resulting body of published research systematically overrepresents novelty and positive results, not because researchers are dishonest but because the incentive structure rewards these outcomes.
Risks and Misunderstandings
The most common misunderstanding is attributing systemic behavior to individual character rather than incentive structure. When a system consistently produces outcomes that differ from stated intentions, the structural explanation is usually more accurate than the moral one. Individuals within well-designed incentive structures generally produce the intended outcomes; individuals within poorly designed structures generally do not, regardless of their personal qualities.
Another mistake is believing that awareness of incentive effects is sufficient to override them. People who understand that their incentives may lead them toward suboptimal behavior still respond to those incentives. Institutional investors who understand the limitations of quarterly reporting still manage to quarterly expectations because the incentive structure demands it. Changing behavior requires changing the structure, not just understanding its effects.
It is also tempting to design simple incentive structures to avoid complexity. But simple structures applied to complex systems often produce unintended consequences because they reward a small number of dimensions in a multi-dimensional context. The dimensions that are not incentivized tend to deteriorate. More thoughtful incentive design considers the full range of desired behaviors, not just the most easily measured ones.
What Investors Can Learn
- Examine what the system rewards — Understanding how individuals within a company are compensated, measured, and promoted reveals what the system is optimized to produce, which may differ from stated strategy.
- Watch for metric-outcome divergence — When the metrics a company reports improve while the business reality seems to deteriorate, the incentive structure may be optimizing the metrics rather than the underlying outcomes.
- Consider time horizon alignment — Whether management incentives are aligned with the time horizon relevant to the investor's assessment reveals potential divergence between what management is optimizing for and what creates long-term value.
- Assess principal-agent alignment — The degree to which decision-makers' incentives align with stakeholder outcomes indicates how reliably the system will produce outcomes that serve those stakeholders.
- Expect second-order effects — The direct effects of incentive structures are usually intentional. The second-order effects, how participants adapt their behavior to optimize within the structure, are often unintended and may be more consequential.
Inside CompanyGraph
View this perspective inside CompanyGraph.
CompanyGraph