Hurwicz, Maskin & Myerson (2007): Engineering Institutions That Make People Tell the Truth

Game Theory in Reverse

Most of economics asks: given the rules of the game, what will players do? Mechanism design asks the opposite question: given the outcome we want, what rules should we design?

This is economics as engineering. Instead of analyzing existing markets and institutions, mechanism design theory provides tools to build new ones — auctions, voting systems, regulatory schemes, contracts — that achieve desired social goals even when participants are self-interested and hold private information.

Three economists built this field. Leonid Hurwicz, who was 90 when he received the prize (the oldest Nobel laureate in economics ever), posed the fundamental question and defined the key concepts. Eric Maskin developed the theory of when desired outcomes can be implemented. Roger Myerson proved the revelation principle — the master theorem that simplifies the entire field.

Hurwicz: The Founding Question

In the 1960s and 1970s, Leonid Hurwicz asked a question that no one had formalized before: how do you design economic institutions when you can’t trust people to tell the truth?

The problem is universal. A government wants to allocate radio spectrum efficiently, but doesn’t know how much each company values it. A regulator wants to set fair prices for a monopoly, but the monopolist knows its costs and the regulator doesn’t. A group wants to divide a shared resource, but each person exaggerates their need.

Hurwicz defined the key concepts:

• Mechanism: A set of rules that specifies what messages participants can send and how those messages determine outcomes. An auction is a mechanism. A voting system is a mechanism. A tax code is a mechanism
• Incentive compatibility: A mechanism is incentive-compatible if every participant’s best strategy is to act truthfully — to reveal their genuine preferences, costs, or valuations. If lying is profitable, the mechanism is flawed
• The impossibility result: Hurwicz showed that in many settings, no mechanism can simultaneously achieve efficiency (the best possible allocation), incentive compatibility (truthful behavior), and budget balance (no outside subsidy needed). Something must give — and the art of mechanism design is choosing what to sacrifice

This framing transformed economics. Instead of assuming institutions exist and analyzing their properties, economists could now ask: what is the best possible institution for this problem?

Myerson: The Revelation Principle

Roger Myerson proved the theorem that makes mechanism design tractable — the revelation principle:

Any outcome achievable by any mechanism can also be achieved by a direct mechanism in which participants simply report their private information truthfully.

This sounds abstract, but it’s enormously powerful. It means that when searching for the best mechanism, you don’t need to consider every possible complex scheme — every elaborate auction format, every intricate contract structure. You only need to consider simple mechanisms where people report their types honestly. If the best direct mechanism achieves a certain outcome, no cleverer mechanism can do better.

Myerson applied this to design optimal auctions — proving that the revenue-maximizing auction for a seller depends on the distribution of buyer valuations, and sometimes involves setting a reserve price below which the item isn’t sold (even if a buyer is willing to pay something).

He also developed the theory of incentive-compatible mechanisms for settings with both private information and moral hazard — where people can take hidden actions as well as hold hidden knowledge. This unified the treatment of adverse selection and moral hazard under a single framework.

Maskin: Implementation Theory

Eric Maskin tackled a different question: when can a social choice rule be implemented by some mechanism?

A social choice rule specifies the desired outcome for every possible configuration of people’s preferences. For example: “allocate the good to whoever values it most” or “choose the policy that maximizes total welfare.” The question is whether there exists any mechanism — any set of rules — that produces this outcome when self-interested people play strategically.

Maskin’s implementation theorem (1977, published 1999) provided necessary and sufficient conditions:

• Maskin monotonicity: If an outcome is chosen when preferences are one way, and preferences change so that the chosen outcome doesn’t fall in anyone’s ranking relative to alternatives, then it should still be chosen. This is a necessary condition for implementation in Nash equilibrium
• No veto power: If all but one person agree on the best outcome, that outcome should be chosen regardless of the dissenter’s preferences

These conditions tell us exactly which social goals are achievable through institutional design and which are not — a fundamental map of what’s possible.

Real-World Applications

Mechanism design isn’t just theory — it has shaped major real-world institutions:

• Spectrum auctions: Governments worldwide use mechanism design principles to auction radio frequencies worth billions of dollars. The design ensures efficient allocation while maximizing revenue
• Matching markets: The design of school choice systems, kidney exchange programs, and medical residency matching all draw on mechanism design
• Regulation: How should a regulator set prices for a monopoly when it doesn’t know the monopoly’s costs? Mechanism design provides the answer — offer a menu of contracts that induces the monopolist to reveal its true costs
• Public goods provision: How do you get people to honestly reveal how much they value a public park or clean air? The Vickrey-Clarke-Groves mechanism provides incentive-compatible solutions
• Online platforms: Auction design for Google ads, Uber pricing, and marketplace rules all use mechanism design principles

Their 2007 Nobel Prize was awarded “for having laid the foundations of mechanism design theory.”

Explain It to a Child

Imagine you and your sister both want the last piece of cake. Your mom could just cut it in half, but maybe one of you wants it more. How does she find out? If she asks, you’ll both say “I want it MORE!” — because exaggerating costs nothing. So your mom invents a clever rule: “One of you cuts the cake, and the other picks first.” Now the cutter has every reason to cut exactly in half — because the picker will take the bigger piece. That’s mechanism design: inventing rules so that selfish people end up doing the fair thing automatically. Hurwicz asked “can we always find such rules?” Myerson proved that we only need to check simple, honest rules. And Maskin figured out exactly which goals can be achieved this way and which can’t.