📈 economics

Uniform price auction with quantity constraints

This paper analyzes uniform price auctions with asymmetric bidders and quantity constraints, presenting an iterative method to find equilibria and an ascending auction mechanism that achieves them as dominant strategies, while demonstrating that low-price equilibria are inevitable when no single bidder can cover the total supply.

Original authors: Kiho Yoon

Published 2026-04-09

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kiho Yoon

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are at a massive, chaotic bakery sale. The baker has a huge tray of 100 fresh croissants (the supply) and wants to sell them all at once.

There are many customers (bidders) in the room, but they are all different:

Some are hungry giants who can eat 20 croissants.
Some are small children who can only eat 2.
Some are very hungry (they value the croissant at $10), while others are just a little peckish (they value it at $2).

The rule of this bakery is the "Uniform Price": Everyone who gets a croissant pays the same price. That price is determined by the "last person" who successfully bought a croissant.

The Big Problem: The "Greed Trap"

In a normal auction, you might think, "I'll just bid my true value, and if I win, I win." But in this specific bakery, there's a twist: You can only submit one single bid for your entire appetite. You can't say, "I'll buy 10 at $5 and 10 at $2." You have to pick one price for all the croissants you want.

This creates a weird game of chicken.

The Analogy: The "Residual Crumbs" Strategy

Imagine you are a giant who wants 20 croissants. You see that the other customers combined only want 80 croissants.

Option A: You bid high ($9) to try to get all 20. If you win, you pay $9 each. Total cost: $180.
Option B: You realize that even if you bid $0, there are still 20 croissants left over because the others only bought 80. So, you decide to drop out of the bidding war. You let the others fight. When they stop, the baker has 20 croissants left. You step in and say, "I'll take the leftovers for free (or very cheap)."

The Result: You get your 20 croissants for almost nothing, while the others paid a high price.

This is called Demand Reduction. It's like a giant at a buffet deciding to eat only a tiny plate so they can get the rest of the food for free later, rather than fighting for the whole table.

The Paper's Solution: The "Smart Clock"

The author, Kiho Yoon, figured out a way to predict exactly what will happen in this chaotic bakery and even designed a better way to run the auction.

1. The "Iterative Procedure" (The Crystal Ball)

The author created a step-by-step math recipe (an algorithm) to figure out the outcome before the auction even starts.

Step 1: Look at the person with the lowest appetite/value. Would they rather fight for their full share at a high price, or give up and take the leftovers for free?
Step 2: If they give up, remove them from the list. Now look at the next lowest person. Do the same calculation.
Repeat: Keep peeling away the "weakest" bidders until the math says, "Okay, the remaining bidders are fighting, and here is the price they will settle on."

The Surprise: Sometimes, the person who values the croissants the most ends up with fewer croissants than someone who values them less, simply because the high-value person decided to "give up" to get a better deal on the leftovers.

2. The "Ascending Clock" (The Fair Play Auction)

The author also designed a new way to run the auction so that everyone plays fair, even if they are hiding how much they really want the croissants.

Imagine a giant clock in the room.

The price starts at $0 and slowly ticks up.
Everyone stands there. As the price goes up, people start to leave (drop out).
The Magic Rule: Every time someone leaves, the auction pauses. The baker checks: "If we stop right now, do the people still in the room have enough appetite to eat all 100 croissants?"
- If YES: The price ticks up again.
- If NO: The auction stops! The person who just left gets the "leftovers" (the croissants no one else wanted) at the current price (which is low). The people still in the room get their full share at that same low price.

Why this is brilliant: In this clock auction, your best strategy is always to stay in until the price hits exactly what the croissants are worth to you. You don't need to be tricky or try to "game" the system. It forces the "Demand Reduction" strategy to happen naturally and efficiently.

The Big Takeaway

The paper teaches us that in markets where people have limits on how much they can buy (like electricity grids or government bond sales), being greedy can actually make you lose money.

The Trap: If everyone tries to get their full share, they bid the price up so high that everyone loses.
The Trick: Smart bidders sometimes pretend to want less so they can get the "leftovers" cheap.
The Result: This often leads to a "Low Price Equilibrium" where the goods are sold very cheaply, sometimes even to people who didn't value them the most, just because they played the "leftover" game.

The author shows us that this isn't a bug; it's a feature of how these specific auctions work. And by using the "Clock Auction," we can make sure this happens in a way that is fair and predictable, rather than chaotic.

1. Problem Statement

The paper investigates the equilibrium outcomes of uniform price auctions (also known as pay-as-clear auctions) involving multiple asymmetric bidders who face quantity constraints (capacity limits).

Market Context: The study addresses real-world markets such as wholesale electricity, spectrum auctions, and treasury securities, where bidders often have flat demand curves up to a specific capacity limit ( $q_i$ ) and zero value beyond that.
The Core Conflict: In these auctions, bidders must decide whether to bid aggressively to secure their full demand at a potentially high clearing price or to strategically reduce their demand (or bid low) to secure only the "residual" supply (supply minus other bidders' demands) at a significantly lower price (potentially zero).
Research Gap: While demand reduction is known in multi-unit auctions, existing literature often assumes symmetric bidders, common values, or allows for multiple bids per bidder. This paper specifically analyzes the case where bidders submit a single lump-sum bid for their entire constrained demand, with private values and asymmetric capacity constraints.

2. Methodology

The author employs a two-pronged methodological approach:

A. Iterative Procedure (Complete Information)

First, the author assumes complete information (all valuations $v_i$ and constraints $q_i$ are common knowledge) to derive a Nash Equilibrium.

The Algorithm: An iterative procedure is constructed to systematically find the equilibrium outcome.
- Step Logic: At any step $k$ , the remaining bidders compare the payoff of winning their full capacity at the current price versus winning the residual supply at the price of the next bidder to drop out.
- Bidder Elimination: The procedure identifies the bidder with the lowest "maximum bid" (the threshold where they prefer the residual quantity over the full quantity). If this bidder's maximum bid equals their valuation, they are effectively "dropped out" (their constraint is removed from the active set), and the process repeats with the remaining bidders.
- Termination: The process stops when a specific condition is met regarding the aggregate demand of the remaining bidders relative to the supply $m$ .

B. Ascending Auction Design (Private Information)

Second, the author designs a dynamic ascending auction (a variant of the English auction) that implements the equilibrium found in the iterative procedure as a Dominant Strategy Equilibrium (DSE) under private information.

Mechanism: A clock price rises continuously. Bidders decide when to drop out.
Provisional Price: The auction tracks a "provisional price" ( $\hat{p}$ $\overset{p}{^}$ ). When a bidder drops out, the auctioneer checks if the remaining aggregate demand covers the supply.
- If remaining demand < supply: The auction ends. The price is set to the provisional price, and the dropping bidder receives the residual supply.
- If remaining demand $\ge$ supply: The provisional price updates to the current price, and the clock continues.
Strategy: The optimal strategy for a bidder is to remain active as long as the current price is below a calculated threshold that balances the value of full demand against the value of the residual demand at the provisional price.

3. Key Contributions

Systematic Equilibrium Characterization: The paper provides a constructive algorithm (iterative procedure) to find the equilibrium outcome for arbitrary numbers of asymmetric bidders with arbitrary quantity constraints, moving beyond the two-bidder or symmetric cases found in prior literature.
Dominant Strategy Implementation: It introduces a specific ascending auction format that transforms the Nash Equilibrium of the static game into a Dominant Strategy Equilibrium, solving the information revelation problem for private values.
Clarification of Demand Reduction Mechanism: The paper distinguishes the mechanism of demand reduction in this setting from previous literature. In standard multi-unit auctions, demand reduction often arises from shading bids on marginal units. Here, because bidders submit a single bid for all units, demand reduction occurs because a bidder strategically chooses to accept a smaller quantity (the residual) at a low price rather than fighting for the full quantity at a high price.

4. Key Results

A. Existence of Low-Price Equilibrium

The most significant finding is the inevitability of low-price equilibria under specific conditions.

Theorem: If no single bidder's quantity constraint is large enough to cover the total supply (i.e., $q_i < m$ for all $i$ ), then a low-price equilibrium is the only possible equilibrium.
Mechanism: When no bidder can monopolize the supply, bidders have a strong incentive to "give up" their full demand to let others take the bulk of the supply, thereby allowing the bidder to secure the residual amount at a price near zero (or the reserve price).

B. Inefficiency in Allocation

The equilibrium allocation is not necessarily efficient (i.e., goods do not always go to the bidders with the highest valuations).

Example: A bidder with a high valuation ( $v_1$ ) may end up with fewer units than a bidder with a lower valuation ( $v_2$ ) if the high-valuation bidder strategically reduces demand to secure a low price, while the lower-valuation bidder bids aggressively to fill the supply.

C. Equilibrium Price Determination

The equilibrium price is always set at one of the bidders' marginal values (or zero). Specifically, the price corresponds to the valuation of the bidder who is "dropped out" in the final step of the iterative procedure.

D. Algorithmic Summary

The iterative procedure can be summarized as follows:

Calculate the maximum bid each bidder is willing to make to secure their full demand versus the residual demand.
Identify the bidder with the lowest such maximum bid.
If this bidder's maximum bid is determined by the "residual" logic (i.e., they prefer the residual), they drop out, and the price is set to the valuation of the next lowest bidder.
Repeat until the supply is fully covered by the remaining bidders' constraints.

5. Significance and Implications

Policy and Regulation: The findings are crucial for regulators in electricity and spectrum markets. The paper suggests that in markets with fragmented capacity (where no single seller/buyer dominates), uniform price auctions are prone to severe price suppression and inefficient allocation due to strategic demand reduction.
Auction Design: The proposed ascending auction offers a practical mechanism to mitigate strategic manipulation. By making the "drop out" decision a dominant strategy, it encourages truthful revelation of valuations regarding the trade-off between quantity and price.
Theoretical Advancement: The paper bridges the gap between static Nash equilibrium analysis and dynamic mechanism design in the context of capacity-constrained multi-unit auctions. It highlights that the "single-bid" rule (common in practice) creates unique strategic incentives distinct from models allowing continuous demand schedules.

In conclusion, Yoon demonstrates that in uniform price auctions with quantity constraints, the fear of high prices drives bidders to strategically under-demand, leading to low clearing prices and potential inefficiencies. However, a properly designed ascending auction can align incentives to achieve the equilibrium outcome as a dominant strategy.