Chapter 11 asked: given preferences and endowments, do competitive markets produce efficient outcomes? The answer — yes, under the welfare theorem conditions — takes the market mechanism as given. This chapter inverts the question: given a desired outcome, can we design a mechanism to achieve it?
Mechanism design is often called "reverse game theory." Instead of predicting the outcome of a game, we design the game to produce a desired outcome. Market design applies these ideas to real institutions — auctions, matching markets, spectrum allocation, kidney exchange.
Prerequisites: Chapters 7 (game theory basics, Nash equilibrium) and 10 (welfare theorems, general equilibrium).
Named literature: Myerson (1981); Vickrey (1961); Clarke (1971); Groves (1973); Gale & Shapley (1962); Roth (2002); Milgrom (2004).
The challenge: agents' types are private. How do we get them to reveal their types truthfully?
Figure 12.1. Mechanism design timeline.
The mechanism designer chooses the rules (message space and outcome function) to achieve a desired social choice function.
A direct mechanism asks each agent to simply report their type (their private information). It is incentive compatible (IC) if truthful reporting is an equilibrium strategy — no agent benefits from lying.
This is the most powerful simplification in mechanism design — arguably the most powerful simplification in all of economic theory. In principle, the space of possible mechanisms is infinitely large. An auction could have any number of rounds, any bidding rules, any payment formula. A matching algorithm could work in any conceivable way. Searching over all possible mechanisms for the best one seems hopeless.
The revelation principle says: you don't have to search. Whatever outcome any mechanism can achieve, a direct mechanism (just ask everyone to report truthfully) can achieve the same outcome. So the mechanism design problem reduces to: find the best allocation rule and payment rule as functions of reported types, subject to the constraint that truth-telling is optimal. This transforms an impossibly broad search into a well-defined optimization problem.
DSIC is stronger but harder to achieve. BIC is weaker but allows more mechanisms.
You now have mechanism design tools — the revelation principle, incentive compatibility, and the distinction between DSIC and BIC. These tools formalize what a government can and cannot achieve when it can't observe people's types directly.
Mechanism design formalizes the redistribution problem with startling clarity. The government wants to transfer from high-ability to low-ability agents but can't observe ability directly — only income, which is a choice variable. The revelation principle says any redistribution scheme can be analyzed as a direct mechanism where agents report their type. The binding constraint is incentive compatibility: high-ability agents must not find it profitable to mimic low-ability agents by working less. A tax-and-transfer system is literally a mechanism — it maps reported incomes to after-tax incomes — and the revelation principle tells you that if any scheme can achieve a redistributive goal, a truthful direct mechanism can too. This is the conceptual foundation of optimal income taxation (Mirrlees 1971): the tax schedule is a mechanism designed to maximize social welfare subject to incentive compatibility.
Incentive compatibility creates an irreducible tradeoff between redistribution and efficiency — and it's worse than the intuitive version. The Myerson-Satterthwaite theorem (§12.4) shows that in bilateral trade with private information, no mechanism simultaneously achieves efficiency, incentive compatibility, individual rationality, and budget balance. Apply this logic to redistribution: the government faces a version of the same impossibility. It cannot design a tax system that fully redistributes, respects incentives, and avoids deadweight loss. Furthermore, the mechanism design framework assumes a benevolent, well-informed planner who knows the distribution of types even if not individual types. In practice, redistributive policy is shaped by political economy — median voters, interest groups, lobbying. The design problem is well-understood; the implementation problem is not.
The mechanism design framework connects directly to optimal income tax theory. Mirrlees (1971) showed that the optimal tax schedule depends on the distribution of abilities and the elasticity of labor supply — both empirical quantities. The mechanism design approach gives the conceptual architecture; the quantitative answers require data. Myerson's optimal auction is structurally identical to optimal taxation: both maximize an objective subject to incentive compatibility and individual rationality. The same math that designs revenue-maximizing auctions designs welfare-maximizing tax schedules.
The efficiency-equity tradeoff is real, but mechanism design makes it precise rather than vague. The tradeoff isn't "redistribution is costly" — it's "redistribution is costly by exactly the amount that incentive compatibility constraints bind." The magnitude depends on specific parameters: how elastic is labor supply? How fat is the tail of the ability distribution? These are empirical questions with empirical answers, not ideological ones. Mechanism design transforms the inequality debate from philosophy into engineering — but the engineering is constrained by informational limits that no cleverness can circumvent.
Mechanism design gives you the framework; optimal tax theory gives the numbers. Come back in Chapter 16 (§16.7) for the Ramsey optimal tax result — tax inelastic goods more — and the quantitative estimates: optimal top marginal rates are probably 50–70% (Diamond & Saez 2011), higher than most countries implement but lower than "tax everything" implies. Then in Chapter 20 (§20.5, §20.8), the problem goes global: within-country inequality is dwarfed by between-country inequality, and the tools for addressing it — institutions, human capital, development interventions — are entirely different from domestic tax design.
Elizabeth Warren's proposal meets mechanism design: the binding constraint on redistribution is incentive compatibility — agents can hide their type. Wealth is harder to hide than income. Does that make wealth taxes better mechanisms?
AdvancedThis is the mechanism design analog of Arrow's impossibility theorem. It says that in general social choice settings, no non-dictatorial mechanism can elicit truthful preferences in dominant strategies.
The escape: restrict the domain. With quasi-linear preferences ($U_i = v_i(a) + t_i$, where $t_i$ is a monetary transfer), the Gibbard-Satterthwaite barrier falls. The VCG mechanism achieves efficiency and DSIC with transfers.
The Vickrey-Clarke-Groves (VCG) mechanism achieves efficient allocation with truth-telling as a dominant strategy, using monetary transfers.
Efficient allocation: $a^*(\theta) = \arg\max_a \sum_i v_i(a, \theta_i)$ — maximize total value.
Agent $i$ pays the externality she imposes on others — the difference between others' welfare with and without $i$.
Why is truth-telling dominant? Under truthful reporting, agent $i$'s payoff is:
$$v_i(a^*(\theta)) + t_i = v_i(a^*(\theta)) + \sum_{j \neq i} v_j(a^*(\theta_{-i})) - \sum_{j \neq i} v_j(a^*(\theta))$$
This simplifies to $\sum_j v_j(a^*(\theta)) - \sum_{j \neq i} v_j(a^*(\theta_{-i}))$. The second term doesn't depend on $i$'s report. So $i$ maximizes her payoff by choosing her report to maximize $\sum_j v_j(a^*(\theta))$ — which happens when she reports truthfully, since $a^*$ already maximizes total value.
Enter agent values for a single indivisible object. The calculator computes VCG payments (equivalent to a second-price auction for a single item).
Figure 12.2. Agent values and VCG payments. Each agent pays the externality they impose on others. The winner pays the second-highest value (in a single-item auction, VCG reduces to the Vickrey auction).
Three citizens value a bridge at $v_1 = 30$, $v_2 = 25$, $v_3 = 15$. The cost is $C = 60$.
Build if $\sum v_i > C$: \$10 > 60$ → yes.
Clarke tax payments:
Total collected: \$10 + 15 + 5 = 40 < 60$. There's a budget deficit of 20 — VCG does not generally achieve budget balance. Each agent pays their "pivotal" contribution.
| Format | Rules | Winner pays |
|---|---|---|
| English (ascending) | Bidders raise bids; last bidder wins | Second-highest value (approx.) |
| Dutch (descending) | Price drops until someone claims | Their bid |
| First-price sealed-bid | Highest bid wins | Their bid |
| Second-price sealed-bid (Vickrey) | Highest bid wins | Second-highest bid |
The Vickrey auction (second-price sealed-bid) is DSIC: each bidder's dominant strategy is to bid their true value $v_i$. Bidding above $v_i$ risks winning at a price above value; bidding below risks losing when the second-highest bid is below $v_i$.
This is a stunning result. It says that the seemingly vast differences between auction formats — open vs sealed bid, ascending vs descending, first-price vs second-price — are irrelevant for expected revenue under these conditions.
When revenue equivalence breaks down:
Set the number of bidders and their value distribution. Run single auctions to see individual outcomes, or run 100 rounds to observe revenue equivalence (average revenues converge across formats). Adjust the risk aversion slider to break equivalence.
Figure 12.3. Auction outcomes. In single runs, revenues differ across formats due to randomness. Over 100 runs, average revenues converge — demonstrating revenue equivalence. Increase risk aversion ($\rho > 0$) to break equivalence: first-price revenue rises above second-price.
When the seller wants to maximize revenue (not efficiency), Myerson showed the optimal mechanism uses the virtual value:
where $F$ is the CDF and $f$ is the PDF of the bidder's value distribution.
The optimal auction allocates to the bidder with the highest virtual value, provided it is positive. If all virtual values are negative, the seller retains the object. This implies a reserve price — the seller sets a minimum bid equal to $\psi^{-1}(0)$.
Values uniformly distributed on $[0, 1]$: $F(\theta) = \theta$, $f(\theta) = 1$.
$\psi(\theta) = \theta - (1-\theta)/1 = 2\theta - 1$
$\psi(\theta) = 0 \implies \theta = 1/2$. Optimal reserve price = \$1/2$.
A second-price auction with reserve \$1/2$ is optimal: the item is sold only if at least one bidder values it above \$1/2$.
For values drawn from Uniform$[0, V_{\max}]$, the virtual value is $\psi(\theta) = 2\theta - V_{\max}$. Drag the reserve price slider. The revenue curve shows expected revenue as a function of the reserve. The optimal reserve (maximizing expected revenue) is highlighted.
Figure 12.4a. Virtual value function $\psi(\theta) = 2\theta - 1$ (for $U[0,1]$). The reserve price is set where $\psi(r) = 0$. Bidders with $\theta < r$ are excluded (shaded red).
Figure 12.4b. Expected revenue as a function of reserve price. The green dot marks the optimal reserve that maximizes expected revenue. Your chosen reserve is shown as a blue dot.
A government allocates a license to one of two firms. Firm $i$ has private value $\theta_i \in \{L, H\} = \{10, 50\}$, each equally likely.
Proposed mechanism: Allocate to the firm reporting higher value; in case of a tie, allocate to firm 1. Payment: the winner pays 30.
Check IC for a high-value firm ($\theta = 50$):
Truthful is better. IC holds for type $H$.
Check IC for a low-value firm ($\theta = 10$):
Truthful is better. IC holds for type $L$. The mechanism is incentive compatible.
Two bidders with values drawn independently from $U[0, 100]$.
Second-price auction: Expected revenue = $E[\text{2nd highest value}] = 100/3 \approx 33.33$.
First-price auction: Optimal bid with 2 bidders: $b(\theta) = \theta/2$. Expected revenue = $E[\max(b_1, b_2)] = E[\max(\theta_1/2, \theta_2/2)] = E[\max(\theta_1, \theta_2)]/2 = (200/3)/2 = 100/3 \approx 33.33$.
Both formats yield expected revenue of \$100/3$, confirming revenue equivalence. The first-price auction generates less variable revenue (each winner pays exactly half their value) while the second-price auction has higher variance (the payment depends on the second-highest value, which can vary widely).
Intuition: The seller wants to overstate her cost (to extract a higher price). The buyer wants to understate his value (to pay less). Incentive compatibility requires leaving "information rents" to both parties. These rents are costly, and with budget balance, there isn't enough surplus to pay both rents and ensure all efficient trades occur.
Real-world bargaining under private information — salary negotiations, used car purchases, M&A deals — always involves some inefficiency. Institutions like posted prices, reputation systems, and standardized contracts mitigate the problem but cannot fully eliminate it.
Some goods cannot be allocated by prices — we don't (or shouldn't) sell school admissions, organ transplants, or residency positions. Matching markets use algorithms instead.
Theorem (Gale & Shapley, 1962). The algorithm terminates in at most $n^2$ rounds and produces a stable matching — no unmatched pair both prefer each other to their current match.
Properties:
Enter preference lists for students and schools. The algorithm animates each round: proposals, tentative holds, and rejections. Enter preferences as comma-separated names (e.g., "W,X,Y,Z").
Four students (A, B, C, D) and four schools (W, X, Y, Z). Students propose.
| Student | Preferences | School | Preferences |
|---|---|---|---|
| A | W > X > Y > Z | W | B > A > D > C |
| B | X > W > Y > Z | X | A > B > C > D |
| C | W > Y > X > Z | Y | C > D > A > B |
| D | Y > W > X > Z | Z | D > C > B > A |
Final matching: A-W, B-X, C-Y, D-Z. This is stable: no pair wants to deviate. Use the interactive above to verify step by step.
Run Gale-Shapley with students proposing vs. schools proposing. Compare the two stable matchings. The proposing side always gets their best stable matching; the responding side gets their worst.
Alvin Roth (Nobel 2012, shared with Lloyd Shapley) describes this as "the economist as engineer" — using economic theory not just to explain the world but to design real institutions that improve people's lives.
The broader lesson: Markets are not natural objects that arise spontaneously. They are designed institutions — rules, algorithms, and enforcement mechanisms that determine who gets what, at what price, and through what process. The design matters enormously.
The city decides to auction the exclusive right to operate a lemonade stand at the prime downtown corner. Three potential vendors: Maya ($v_M = 50$/day), Nate ($v_N = 35$/day), Olivia ($v_O = 20$/day). Values drawn from $U[0, 60]$.
Second-price auction (Vickrey): Dominant strategy is to bid truthfully. Maya bids 50, Nate bids 35, Olivia bids 20. Maya wins, pays 35.
Optimal auction (Myerson): Virtual values with $F(\theta) = \theta/60$, $f(\theta) = 1/60$:
$\psi(\theta) = \theta - (60 - \theta) = 2\theta - 60$
Reserve price: $\psi(\theta) = 0 \implies \theta = 30$.
Maya's virtual value: \$1(50) - 60 = 40$. Nate's: \$10$. Olivia's: $-20$ (excluded by optimal auction).
In a second-price auction with reserve 30: Maya wins, pays $\max(35, 30) = 35$.
Roth as "Economist as Engineer." Alvin Roth (Nobel Prize 2012) transformed mechanism design from pure theory into a practical discipline that redesigns real markets. His work demonstrates that markets are designed institutions, not natural phenomena.
The National Residency Matching Program (NRMP): Roth diagnosed why the original medical residency match was failing (instability, strategic manipulation) and redesigned it using deferred acceptance. The new system matches ~40,000 medical residents annually.
Kidney exchange: Roth, Sonmez, and Unver designed exchange protocols allowing incompatible donor-patient pairs to swap donors through chains of transplants, saving thousands of lives. This was pure market design — creating a market where none existed, without using prices.
School choice: Roth and colleagues replaced Boston's manipulable school assignment mechanism with a strategy-proof system. Under the old system, parents who reported their true preferences were punished; under the new system, honesty is always optimal.
Spectrum auctions: Milgrom and Wilson (Nobel Prize 2020) designed combinatorial auctions for the FCC, raising billions of dollars while efficiently allocating spectrum licenses. The 2017 incentive auction alone raised \$19.8 billion.
The common thread: economic theory provides the blueprint, but implementation requires understanding the specific institutional context — the "details" that pure theory abstracts away.
You now have the complete toolkit: the welfare theorems told you when markets work (Chapter 11); mechanism design and market design show you what to do when they don't. This is the final stop.
When traditional markets fail — when the welfare theorem conditions don't hold — you can engineer better institutions. The revelation principle says the design space is tractable: focus on direct truthful mechanisms. VCG implements efficient outcomes with dominant-strategy incentives when preferences are quasi-linear. And where prices can't work at all — kidneys can't be bought, school seats can't be auctioned — Gale-Shapley's deferred acceptance produces stable matchings without any monetary transfers. These aren't hypotheticals. Kidney exchange has saved thousands of lives by creating a market where none could exist. School choice redesigns replaced manipulable systems with strategy-proof ones, making honesty the optimal strategy for every parent. Spectrum auctions (Milgrom, Wilson — Nobel 2020) raised billions while allocating licenses efficiently. Roth's "economist as engineer" program demonstrates that economic theory can design real institutions that outperform both unregulated markets and blunt government intervention.
The Myerson-Satterthwaite impossibility is a cold shower for mechanism design optimism: in bilateral trade with private information, no mechanism can simultaneously achieve efficiency, incentive compatibility, individual rationality, and budget balance. This isn't a technical limitation — it's a fundamental impossibility. The success stories of market design (matching, auctions, kidney exchange) share a crucial feature: they operate in structured, well-defined environments where the "rules of the game" are clear and the designer has substantial control. In messier environments — healthcare systems, financial markets, labor markets, macroeconomic policy — the institutional design problem is orders of magnitude harder. The mechanism designer needs to know the distribution of types, the set of feasible allocations, and agents' utility functions. In complex real-world settings, this knowledge is precisely what the designer lacks. The mechanism design revolution may have succeeded in the easy cases while leaving the hard ones untouched.
Market design matured into a pragmatic discipline that takes the limitations seriously. Roth's methodology is explicitly "design, implement, observe, redesign" — not "prove optimality and deploy." The NRMP matching algorithm has been revised multiple times as new problems emerged (couples matching, rural hospital shortages). Spectrum auction formats evolved from simple simultaneous ascending auctions to complex combinatorial designs as the FCC learned from earlier rounds. The profession moved from proving impossibility results to asking: given the impossibilities, what's the best achievable mechanism? Computational mechanism design — integrating algorithmic constraints with incentive constraints — is the active frontier, particularly relevant as digital platforms become the dominant market institutions.
Markets allocate resources efficiently when the welfare theorem conditions hold — and they hold approximately enough to make markets the default for most goods. When they fail, mechanism design offers a genuine alternative: not "let the government decide" but "design an institution whose incentives produce the outcome you want." The success stories are real and important. But mechanism design is not a universal solvent. It works best in structured, well-defined settings. The frontier — digital markets, algorithmic pricing, AI-mediated transactions, platform monopolies — raises questions that existing theory doesn't fully address. The answer to "do markets allocate resources efficiently?" is: yes, when conditions hold; and when they don't, we can sometimes engineer something better — but "sometimes" is doing heavy lifting in that sentence, and the engineering is harder than the theory suggests.
This is the final stop on BQ #7. The arc ran from surplus as benchmark (Ch 3) through market failures (Ch 4), the formal welfare theorems (Ch 11), and now mechanism design. The question "do markets allocate resources efficiently?" turns out to be the wrong question — the right one is "under what conditions, and what can we build when conditions fail?" The answer involves welfare theorems and mechanism design and the practical wisdom that design is constrained by politics, information, and computation. The next frontier is where mechanism design meets behavioral economics (Chapter 19) — agents who aren't fully rational may not respond to incentive-compatible mechanisms the way theory predicts. Bounded rationality may be the binding constraint that mechanism design hasn't yet solved.
Bernie Sanders' rallying cry meets mechanism design: healthcare fails every welfare theorem condition. Can mechanism design do better? Kidney exchange says yes for organs. For the rest of healthcare, the design problem remains unsolved.
IntermediateKhan's antitrust paradox: platform markets are designed institutions — but designed by the platforms, for the platforms. The consumer welfare standard is blind to it.
Advanced| Label | Equation | Description |
|---|---|---|
| Eq. 12.1 | $U_i(\theta_i, \theta_i) \geq U_i(\hat{\theta}_i, \theta_i)$ for all $\hat{\theta}_i, \theta_{-i}$ | DSIC |
| Eq. 12.2 | $E[U_i(\theta_i, \theta_i)] \geq E[U_i(\hat{\theta}_i, \theta_i)]$ | BIC |
| Eq. 12.3 | $t_i = \sum_{j \neq i} v_j(a^*(\theta_{-i})) - \sum_{j \neq i} v_j(a^*(\theta))$ | VCG payment |
| Eq. 12.4 | $\psi(\theta) = \theta - (1-F(\theta))/f(\theta)$ | Myerson virtual value |
Coming in Part V: graduate macro. The models get serious — and so do the policy debates.