Moral Hazard

Definition

Moral hazard (also called hidden action) occurs when an agent can take actions after a contract is signed that the principal cannot directly observe or verify, and these actions affect the principal's payoff. The agent may exert less effort, take excessive risk, or misuse assets entrusted to them—particularly when the agent's payoff does not fully reflect the consequences of their actions. Unlike adverse selection, where hidden information exists before contracting, moral hazard involves hidden actions taken by the agent after the contract is in place.

In the Book

Chapter 4 develops moral hazard as the second core problem of incentive theory. The basic model assumes the principal cannot observe the agent's effort level, only a noisy output signal that depends on both effort and a random state of nature. The first-best contract would condition on effort directly, but since effort is unobservable, the contract must rely on output alone—which creates noise and inefficiency. The agent bears risk on output to create incentives for effort, but if the agent is risk-averse, the cost of bearing this risk may be high.

The book shows that even with risk-neutral agents, moral hazard creates trade-offs. With limited liability constraints (Ch. 4.3), the agent cannot be penalized beyond their assets, so extreme contracts are infeasible. With risk aversion, the principal must balance insurance against incentives: paying the agent more when output is high incentivizes effort, but exposes the risk-averse agent to income variability.

Adam Smith recognized this in his discussion of piece-rate work (Ch. 1), noting that workers "when liberally paid by the piece, are very apt to overwork themselves, and to ruin their health." The book also discusses the multitask incentive problem (Ch. 5.2): when an agent can allocate effort across multiple tasks but the principal can only measure some of them, incentive pay on measured tasks may cause neglect of unmeasured ones.

Why It Matters

Moral hazard explains the structure of real compensation and monitoring systems. Performance pay, bonuses, equity stakes, and threat-of-termination all exist as solutions to hidden action. It also explains why monitoring, auditing, and punishment exist despite their costs. The framework reveals that some inefficiency is necessary—perfect incentive alignment under uncertainty is impossible without bearing costs. Understanding moral hazard is essential for organizational design, particularly in contexts where actions are difficult to verify.