The Principal-Agent Problem: When the Person You Hired Has Different Goals

You hire a contractor to renovate a kitchen while you are away for a month. You want quality work at a fair price. The contractor wants to be paid as much as possible for as little effort as possible, and you are not there to watch. You have just created, in miniature, the single most important relationship problem in all of organizational economics — and it scales all the way up to the relationship between a corporation's millions of shareholders and the executive they pay to run it.

What the problem actually is

A principal-agent problem arises whenever one party — the principal — hires another party — the agent — to act on its behalf, but cannot perfectly observe or control what the agent does. Two conditions have to be present. First, the goals of principal and agent must diverge at least somewhat. Second, there must be asymmetric information: the agent knows more about its own effort and choices than the principal can see. Remove either condition and the problem dissolves. If goals aligned perfectly, the agent would do what you want anyway. If you could observe everything, you could simply require the right behavior. It is the combination — conflicting interests plus hidden action — that creates the friction.

This is the same hidden-action problem that produces moral hazard, applied to delegation. The 2001 Nobel Prize work of Akerlof, Spence, and Stiglitz on asymmetric information laid the groundwork for the formal theory of how to write contracts under exactly these conditions (NobelPrize.org, 2001).