Revenue equivalence theorem

Overview

The revenue equivalence theorem (RET) is a benchmark result in auction theory. It says that, under a set of standard conditions, all “reasonable” auction formats that award the object to the highest-valuation bidder yield the same expected revenue to the seller (and the same expected payment from each bidder type). In practice, this means first-price and second-price sealed-bid auctions, ascending (English), and descending (Dutch) auctions are revenue-equivalent—provided the allocation rule is the same and the mechanisms give the lowest-valuation bidder zero expected surplus. RET matters for executives because it clarifies when format choice does not drive outcomes; it shifts attention to what does: bidder competition, reserves, participation, and information.

Origins and Credit

William Vickrey (1961) showed the equivalence between second-price and other formats in a symmetric private-values setting. Roger Myerson (1981) and John Riley and Richard Samuelson (1981) provided the general statement: for any mechanism that produces the same allocation rule, expected payments are pinned down by incentive compatibility and boundary conditions. Paul Milgrom and Robert Weber (1982) extended analysis to correlated/common values, linking format and information to revenue.

Core Idea and Mechanics

Consider n risk-neutral bidders with independent private values drawn from the same continuous distribution. An “allocation rule” q(v) gives the probability that a bidder with value v wins. In standard auctions, the highest value wins, so q(v) = F(v)ⁿ⁻¹, where F is the value distribution.

Revenue equivalence says: for any two incentive-compatible mechanisms that (a) implement the same q(v) for all v, and (b) give the same expected payoff to a particular type (often the lowest type gets zero), the expected payment of each type is identical. Intuitively, truth-telling and the same win probabilities fix the bidder’s surplus schedule up to a constant; matching the boundary condition pins down expected payments. A compact formula behind RET is that a type v’s expected payment equals v·q(v) minus the area under q up to v (plus a constant chosen to match the boundary condition). Because q is the same across formats, so are expected payments and total revenue.

Procurement (“reverse auctions”) is the mirror image: with independent private costs and the lowest-cost supplier winning, expected procurement cost is the same across analogous formats (e.g., first-price sealed-bid vs. descending clock), under the same assumptions.

Key Assumptions and Conditions

• Independent private values (IPV): each bidder knows their own value; values are independently drawn from a common continuous distribution.
• Risk-neutral bidders; linear utility for money.
• Symmetry: bidders are ex ante identical (same distribution of types).
• Efficient allocation rule: the highest value (or lowest cost) always wins (ties aside).
• Incentive compatibility and individual rationality: truthful bidding/reporting is optimal given beliefs; the lowest type earns zero expected surplus (common normalization).
• No binding budget constraints, entry frictions, or collusion; no reserve-price or participation-rule differences across compared formats unless explicitly held fixed.

Implications

• Format neutrality under the benchmark: with IPV, risk neutrality, and symmetry, first-price, second-price, English, and Dutch auctions deliver the same expected revenue/cost if they implement the same allocation and boundary condition.
• Design levers that matter:
  • Reserve prices and entry policy: a higher optimal reserve boosts expected revenue; RET compares formats holding the reserve rule fixed.
  • Number of bidders: more competition raises expected revenue; recruiting bidders often dominates fine-tuning formats.
  • Information disclosure: outside the strict IPV benchmark, disclosure (signals, reports) can change revenue.
• Robust prescriptions: in many private-values settings, format choice can be driven by administrative simplicity, transparency, or legal risk rather than revenue concerns.

Example in Practice

Procurement of a regional logistics contract. A retailer invites five qualified carriers to compete for a one-year lane bundle. Each carrier’s private unit cost is independently drawn from a common distribution; carriers are risk-neutral. The retailer can either run:

• A sealed-bid first-price reverse auction (lowest bid wins, pays their bid), or
• A descending-price “clock” auction (price ticks down; first to accept wins at the current price), or
• A second-price–style procurement (award to the lowest bidder, pay the second-lowest bid).

Under the RET conditions and with the same reserve or qualification rule, the expected procurement cost is the same across these formats. Management can choose based on ease of participation, auditability, and speed. If they add a reserve (a maximum acceptable price), expected cost falls further relative to no reserve—equally across formats—because the allocation/participation rule has changed in the same way.

However, if carriers are risk-averse or values are correlated (e.g., common fuel shocks), equivalence breaks: a first-price reverse auction typically yields lower expected cost (more aggressive bidding) with risk-averse suppliers, while an open, descending auction can mitigate winner’s-curse effects under common values.

Limitations and Common Misunderstandings

• Assumptions are strong: Risk aversion, budget constraints, entry costs, asymmetries (incumbent advantage), or collusion break equivalence. With risk-averse bidders, first-price tends to raise revenue relative to second-price.
• Common or affiliated values: When signals are correlated and the winner’s curse looms, ascending formats that reveal information often yield higher revenue (Milgrom–Weber linkage principle).
• Equivalence holds conditional on the allocation rule: Changing reserves, eligibility, lot structure, or activity rules changes the allocation and thus revenue—even within the same “format.”
• Not claim about realized prices: RET is about expected revenue ex ante, not the price in any single auction.
• Strategic environment matters: Shill bidding risk, dynamic entry, and platform features (soft closes, bid increments) can differentially affect formats in practice.
• Procurement mirror applies with care: In reverse auctions with supplier capacity constraints or multi-attribute scoring, the simple equivalence need not hold.

Related Concepts

Vickrey (second-price) auction; Myerson’s optimal auction; Independent private values; Linkage principle; Reserve prices; Winner’s curse.