Skip to content

Incentive Compatibility

Why It Matters

Most institutional failures stem from incentive incompatibility - situations where doing what the designer wants actively harms the agent. You can't trust, educate, or exhort your way out of misaligned incentives. If the structure rewards bad behavior, you'll get bad behavior. Incentive compatibility makes you diagnose whether your proposed system fights human nature or harnesses it.

The Key Move

For any proposed rule, policy, or mechanism, ask: "If I were subject to this rule and acting purely in my self-interest, what would I actually do?" Compare that answer to what the rule intends. If they diverge, you've found an incentive compatibility failure. Then either redesign the incentives or accept that you won't get the intended behavior.

Classic Application

Second-price sealed-bid auctions (Vickrey auctions). Bidders pay the second-highest bid, not their own. This makes truthful bidding a dominant strategy - you never benefit from lying about your valuation. Bidding your true value maximizes your chance of winning when you should and avoids winning when you shouldn't. The mechanism's structure makes honesty optimal.

Surprising Application

Code review processes in software teams. If promotions depend on individual output metrics (lines of code, features shipped), reviewing others' code is incentive incompatible - it takes time away from your own metrics. Predictable result: superficial reviews or review avoidance. Fixing this requires making review quality visible and rewarded, not hoping developers will "do the right thing."

Failure Modes

Over-application to settings where intrinsic motivation dominates (some people actually care about goals beyond narrow self-interest). Ignoring enforcement costs (sometimes incentive compatibility is too expensive to achieve). Forgetting that common knowledge of rationality is unrealistic - mechanisms that require everyone to know everyone is rational often fail. Assuming agents understand complex incentive structures (many real mechanisms are too opaque for participants to calculate optimal strategies).

Go Deeper

Mas-Colell, Whinston, and Green, "Microeconomic Theory," Chapter 23. Hurwicz (1973), "The Design of Mechanisms for Resource Allocation" - the foundational paper. Myerson (1989), "Mechanism Design" in The New Palgrave for accessible overview.