The Principal–Agent Problem

Almost nothing important gets done by one person acting alone. You hire a contractor to build the extension, a fund manager to grow your savings, a surgeon to fix your knee, a mechanic to sort the car; shareholders hire a CEO, citizens hire politicians, a company hires ten thousand employees. Each time, you hand your goal to somebody else and say, in effect, act for me. And each time you run into the same wall: the person acting for you has their own interests, and knows things you don’t. The contractor would rather bill more hours; the fund manager would rather gather assets than beat the market; the surgeon knows whether the operation was really needed and you don’t. That gap — between what you want and what your stand-in is actually driven to do, across a wall of things you can’t see — is the principal–agent problem, and it is one of the deepest and most universal problems in all of economics.

The principal is the party who wants an outcome and delegates it. The agent is the party who acts on their behalf. The problem is that their interests diverge and the principal cannot fully observe what the agent does or knows — so the agent, chasing their own payoff exactly as any of us would, drifts away from the principal’s goal, and the principal often can’t even tell. This isn’t a story about villains. It is the default friction that appears the instant one person delegates to another whose incentives don’t perfectly match and whose actions or knowledge are hidden. Name it once and you will see it everywhere: in every contract you sign, every professional you rely on, every organisation you’re part of.

This is an expert course sitting at the meeting point of economics and psychology, and it leans on two models you’ve already built. From incentives you bring the iron law that people respond to the rewards they actually face, not the ones you wish they’d honour — the engine that makes an agent’s interests pull against yours in the first place. From mechanism design you bring the designer’s move: don’t beg people to behave — build the rules so that the behaviour you want is their own best response, which is exactly the craft we’ll turn on the agency problem. If those two feel shaky, shore them up first; everything here composes them.

From there we build the whole model, floor to ceiling. We start with the setup — principal, agent, and the gap between them — and see how astonishingly wide the model reaches. Then we split the problem along its two great fault lines: hidden action (moral hazard — you can’t watch the agent’s effort or risk-taking, so they shirk or gamble with your stake) and hidden information (adverse selection — the agent knows things you don’t before you even sign, the trap behind the used-car “market for lemons”). We put a price on all of it with agency costs — the value bled away by misalignment plus everything you spend monitoring and bonding against it — and then open the alignment toolkit: pay-for-performance and equity, monitoring and audits, screening and signalling, reputation and repeated dealing, efficiency wages — and the ways each one backfires. We prove why the perfect contract is impossible (the unavoidable trade-off between risk and incentives, incomplete contracts, the multitasking trap), and we stay ruthlessly honest about where the model lies to you: when money is not the only motive and paying more gets you less, when strong incentives just amplify measurement error and invite gaming, and when treating trust and monitoring as free is its own expensive mistake.

Throughout, you’ll have an incentive-contract designer to play with — slide an agent’s pay between flat salary and pure commission, dial their appetite for risk, and watch effort, gaming, the agency cost, and your own net payoff move until you feel the punchline in your hands: the best contract is almost never pure salary and almost never pure commission, and pushing incentives too hard makes everything worse. By the end you’ll stop asking the helpless question — how do I get this person to just do what I want? — and start asking the sharp one: given that they’ll chase their own payoff through a wall I can’t see past, what deal makes acting in my interest their best move too?