In traditional economic theory, markets are often treated as static rooms where all participants are known, and more information is universally better. In reality, modern digital economies—ranging from social network commerce to decentralized finance and real-time ad auctions—operate under a fundamental tension: the information-competition tradeoff.
For a buyer, sharing information about a market opportunity expands the market but invites competitors who might outbid them. For a seller or platform, revealing precise information about an asset can eliminate uncertainty, but it also risks destroying competitive tension by exposing a lack of rival interest.
To maximize efficiency and revenue, mechanism designers must strategically manipulate how information flows. This research direction focuses on two sides of the same coin: designing incentives that encourage participants to propagate information, and strategically withholding information to force healthy competition.
Overcoming the Forwarding Dilemma in Social Networks
Imagine you discover an online auction for a rare collectible. The seller promises that if you invite your friends, you might receive a discount. However, you know your friends share your tastes and have deeper pockets. Do you invite them?
As noted by Bin Li et al. in Mechanism Design in Social Networks, this creates a "forwarding dilemma." Bidders are naturally disincentivized from expanding a market because new entrants drive up transaction prices or reduce the original participant's probability of winning. As a survey by Yuhang Guo et al. in Emerging Methods of Auction Design in Social Networks highlights, this information suppression prevents markets from reaching high-valuation buyers, leaving social welfare and seller revenue suboptimal.
To solve this, researchers have designed diffusion auctions that reward information sharing. The core intuition is simple: the mechanism must ensure that the financial reward for inviting a competitor outweighs the risk of being outbid by them.
[Seller] ---> [Buyer A] ---> [Buyer B (Invited by A)]
|
v (Without diffusion incentives: A hides the auction)
v (With diffusion incentives: A shares because rewards > risk)
Several structural innovations make this possible:
- Conditional Activation: In Multi-Unit Auction in Social Networks with Budgets, Mingyu Xiao 0001 et al. propose delaying the entry of a buyer's social network neighbors until that buyer is no longer competing for the resource. Because descendants only activate when the referrer is out of the running, the referrer faces zero increased competition during their active bidding phase.
- Distance-Based Prioritization: In Strategy-Proof and Non-Wasteful Multi-Unit Auction via Social Network, Takehiro Kawasaki et al. introduce the DNA-MU mechanism. It uses a buyer's topological distance from the seller to prioritize bids, guaranteeing that inviting deeper network nodes cannot retroactively penalize the structural advantage of closer nodes.
- Decoupled Competition: Similarly, Dengji Zhao et al. developed the Generalized Information Diffusion Mechanism (GIDM) in Selling Multiple Items via Social Networks, which structures payments so that participants are explicitly compensated out of the surplus generated by the peers they successfully recruited.
Strategic Concealment: Why Less Information Generates More Revenue
While buyers strategically withhold information to suppress competition, sellers and platforms frequently do the exact opposite: they withhold information to manufacture competition.
In a second-price auction with perfect information, a dominant bidder who knows they face no serious rivals can bid low and capture massive "information rent." To prevent this, the auctioneer can strategically blur the details of the item being sold.
Preventing Thin Markets via Attribute Hiding
In Revenue Maximization via Hiding Item Attributes, Mingyu Guo et al. demonstrate that concealing specific item characteristics forces bidders with distinct preferences to compete for the same generalized item. While full transparency maximizes social welfare (ensuring the perfect match wins), hiding information transfers value from the bidders to the seller by preventing "thin markets" where a lack of localized competition would otherwise drive prices to zero.
Information Pooling and Asymmetric Signaling
In digital advertising, platforms must decide how much user data to reveal to advertisers. In Targeting and Signaling in Ad Auctions, Ashwinkumar Badanidiyuru et al. analyze information pooling.
Instead of revealing precise user states, the platform groups high-competition and low-competition states together. This forces dominant bidders to bid defensively against the statistical probability of a rival's interest, maintaining competitive pressure even when no natural rival exists. By keeping strong bidders slightly less informed than their competitors, the auctioneer compels them to hedge against uncertainty by bidding more aggressively.
Precise States: [User A: High Value] [User B: Low Value] -> Bidder drops bid on B
Pooled State: [User A or B: Mixed Value Bundle] -> Bidder must bid high to hedge
This concept is formalized by Hu Fu et al. in The Value of Information Concealment, proving that withholding private signals regarding a buyer's valuation mathematically caps the information rent a buyer can extract, bounding the seller's revenue advantage by a constant factor.
Information Asymmetry as a Market Distorter
When information withholding is not controlled by a central mechanism designer but occurs organically, it can lead to severe market distortions and centralization.
A prime example is found in blockchain block-building auctions. In Private Order Flows and Builder Bidding Dynamics: The Road to Monopoly in Ethereum's Block Building Market, Shuzheng Wang et al. apply asymmetric auction theory to show how dominant block builders leverage exclusive, private transaction data ("private order flow").
Because these builders possess information hidden from their competitors, they can optimize their bids to maintain high win rates while bidding less aggressively relative to the true underlying value of the block. This informational monopoly creates a feedback loop that drives the market toward centralization, pricing out participants who rely solely on public data.
Manipulating Competitors Through Partial Data Sales
The strategic manipulation of the information-competition tradeoff is not limited to auctioneers. Even direct competitors can benefit from sharing data if done selectively.
In Selling Data to a Competitor (Extended Abstract), Ronen Gradwohl et al. challenge the intuition that an incumbent firm should always guard its consumer data. They model an environment where an incumbent sells data about a specific subset of consumers to an uninformed rival.
By strategically choosing which segments to disclose, the incumbent alters the rival's optimal pricing strategy from uniform to personalized pricing for those specific individuals. Crucially, this manipulation shifts the rival's pricing floor for the remaining undisclosed market, allowing the incumbent to soften overall market competition and extract higher profits.
Ultimately, whether by expanding a social network auction or pooling ad impressions, modern market design relies on a singular realization: information is not merely a commodity to be traded, but a dial that directly controls the intensity of competition.
Key Foundational Work
Designing Auctions for Deliberative Agents (2004) — contributed 2 concept elements to this direction.
Collective rent seeking when sharing rules are private information (2006) — contributed 2 concept elements to this direction.