Optimal commitments in auctions with incomplete information

Zihe Wang, Pingzhong TangarXiv 2015 · OpenReview

Abstract

We are interested in the problem of optimal commitments in rank-and-bid based auctions, a general class of auctions that include first price and all-pay auctions as special cases. Our main contribution is a novel approach to solve for optimal commitment in this class of auctions, for any continuous type distributions. Applying our approach, we are able to solve optimal commitments for first-price and all-pay auctions in closed-form for fairly general distribution settings. The optimal commitments functions in these auctions reveal two surprisingly opposite insights: in the optimal commitment, the leader bids passively when he has a low type. We interpret this as a credible way to alleviate competition and to collude. In sharp contrast, when his type is high enough, the leader sometimes would go so far as to bid above his own value. We interpret this as a credible way to threat. Combing both insights, we show via concrete examples that the leader is indeed willing to do so to secure more utility when his type is in the middle. Our main approach consists of a series of nontrivial innovations. In particular we put forward a concept called equal-bid function that connects both players' strategies, as well as a concept called equal-utility curve that smooths any leader strategy into a continuous and differentiable strategy. We believe these techniques and insights are general and can be applied to similar problems.

Introduction

In standard auction theory, the search for a Bayes Nash Equilibrium (BNE) is the traditional approach to understanding bidder behavior. However, in asymmetric auctions where bidders have different valuation ranges, finding this equilibrium is a notoriously difficult open problem. This paper by Zihe Wang and Pingzhong Tang shifts the perspective to a Stackelberg (Commitment) setting. In this model, one bidder (the "leader") publicly announces their bidding strategy before the auction begins, and the other bidder (the "follower") best-responds.

While commitment is known to increase a leader's utility, calculating the optimal commitment strategy for continuous probability distributions was previously unsolved for general auction types like first-price and all-pay auctions. The authors develop a novel mathematical framework to solve for these optimal strategies, revealing counterintuitive behaviors such as passive bidding at low valuations and overbidding at high valuations.

Methodology

The authors introduce a general approach for "rank-and-bid" auctions, where payments depend on the bid and the outcome. Their framework relies on three primary innovations:

  1. The Equal-Bid Function ($g$): This function maps a leader’s type to the follower’s type that would submit the same bid, simplifying the calculation of winning probabilities.
  2. Equal-Utility Curves ($eu$): The authors prove that any arbitrary strategy can be "smoothed" into a continuous, differentiable strategy using these curves without reducing the leader's utility.
  3. Sorting and Optimization: By showing that the leader always prefers a monotone strategy, the problem is transformed into an optimization over the $g$ function, solvable using Lagrangian multipliers.
The relationship between a leader's strategy $s_A(x)$ and the equal-utility curve $eu_B(0.5, x)$, illustrating the smoothing process.

Key Insights and Results

The theoretical analysis, supported by synthetic probability distributions, yields several surprising insights into optimal strategic commitment:

  • Passive Bidding as "Credible Collusion": The optimal strategy often dictates that the leader bid zero when their valuation is below a certain threshold. This "throws the game" at low values, signaling to the follower that high bids are unnecessary, which drives down the follower's bids overall.
  • Overbidding as a "Credible Threat": Conversely, the leader may optimally commit to bidding above their own valuation for high types. While seemingly irrational, this acts as a threat that forces the follower to bid zero to avoid losses, allowing the leader to capture significant surplus when their valuation is in the middle range.
  • Closed-Form Solutions and Utility Gains: The paper provides the first closed-form solutions for optimal commitment in first-price and all-pay auctions. For example, with uniform distributions on $[0, 1]$, a leader might bid zero 56.7% of the time yet still achieve a 32% increase in expected utility compared to the standard Nash Equilibrium.

Conclusion and Future Work

This work provides a powerful new toolkit for analyzing commitment in auctions. However, several avenues for future research remain. The current framework is primarily limited to two-bidder scenarios. Additionally, the model assumes absolute credibility of the leader's commitment, which may be challenging to maintain in practice, especially when it involves overbidding. Future studies could explore "noisy" commitments where the follower has imperfect information about the leader's strategy.

Summary updated

Abstract

We are interested in the problem of optimal commitments in rank-and-bid based auctions, a general class of auctions that include first price and all-pay auctions as special cases. Our main contribution is a novel approach to solve for optimal commitment in this class of auctions, for any continuous type distributions. Applying our approach, we are able to solve optimal commitments for first-price and all-pay auctions in closed-form for fairly general distribution settings. The optimal commitments functions in these auctions reveal two surprisingly opposite insights: in the optimal commitment, the leader bids passively when he has a low type. We interpret this as a credible way to alleviate competition and to collude. In sharp contrast, when his type is high enough, the leader sometimes would go so far as to bid above his own value. We interpret this as a credible way to threat. Combing both insights, we show via concrete examples that the leader is indeed willing to do so to secure more utility when his type is in the middle. Our main approach consists of a series of nontrivial innovations. In particular we put forward a concept called equal-bid function that connects both players' strategies, as well as a concept called equal-utility curve that smooths any leader strategy into a continuous and differentiable strategy. We believe these techniques and insights are general and can be applied to similar problems.
BibTeX
@misc{wang2015optimal,
  author = {Zihe Wang and Pingzhong Tang},
  title = {{Optimal commitments in auctions with incomplete information}},
  year = {2015},
  eprint = {1502.07431},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1502.07431},
}
Figure