The Contract’s Blind Spot

On 10 October 2016, the Royal Swedish Academy of Sciences announced that Oliver Hart and Bengt Holmström would share the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel “for their contributions to contract theory.” Their work addresses a surprisingly productive starting point: agreements must operate around things people cannot fully observe or specify.

The achievement is not a contract that finally anticipates everything. It is a clearer account of how cooperation can be organized when that ambition fails.

What the result cannot tell you

Consider someone hired to run a company. Shareholders can observe results, but not every action that produced them. The Academy describes Holmström’s late-1970s work as a way to design contracts for this principal-agent problem: one party, the principal, relies on another, the agent, whose actions are partly hidden.

Performance-related pay might seem the obvious answer. Reward success; await success. A wonderfully economical theory, especially if one leaves out the world.

But results and effort are not identical. Imagine a manager working diligently while circumstances outside their control worsen. Linking pay to an outcome must contend with both incentives and risk. According to the Academy, Holmström’s informativeness principle specifies how performance-relevant information should enter that arrangement. The question is not merely how large to make a reward, but what the available evidence actually reveals about the work.

His later research extended the framework to settings involving promotion, teams and multiple tasks, including cases where only some dimensions of performance are observed.

That last problem is easy to picture. Suppose a job involves both completing transactions and helping people understand them, but the employer counts only transactions. The count makes one part of the job visible. It has not made the rest unimportant. This is an illustration of the measurement problem, not evidence that every particular incentive scheme fails.

Who decides when the document runs out?

Hart’s contributions address a different limit. Even if everyone could inspect everyone else’s work, a contract could not specify every eventuality.

The Academy dates his fundamental contributions to incomplete-contract theory to the mid-1980s. Instead of requiring an exhaustive script, this approach examines the allocation of control rights: which party gets to decide, and under what circumstances?

The gap is not just missing prose. It is a question about authority.

The Academy identifies applications to business ownership, mergers, financing and public versus private ownership of institutions such as schools and prisons. These are tools for analyzing choices, not an announcement that one ownership model has won every argument. The prize release provides an overview of the research; it does not establish a verdict on any particular institution.

A workable agreement is not the whole good

There is a useful discipline here: design around the limits of knowledge rather than disguising them with confident paperwork. Better information and clearer decision rights can make a relationship more intelligible without making its participants omniscient.

But intelligibility is not legitimacy. A model can help examine incentives while leaving us to ask whose interests define success, who bears the risks, and whether people affected by an agreement have a meaningful way to challenge it. Those are ethical questions this account raises, not conclusions supplied by the prize announcement.

The aim should not be an arrangement so orderly that nobody can object. It should be one capable of hearing what its measurements and clauses leave out.

A contract cannot contain a future. It can help people meet one together—with fewer avoidable misunderstandings, and somewhere for the unexpected to go besides onto the shoulders of whoever has the least power.