Bidders' Responses to Auction Format Change in Internet Display Advertising Auctions

Bidders' Responses to Auction Format Change in Internet Display Advertising Auctions

Source: econ.GN updates on arXiv.org — https://arxiv.org/abs/2110.13814 Date read: 2026-09-02 Connected to: L-001, L-010 Kind: empirical Escalation: store-only Escalation rationale:

What this is

An empirical study of bidder behavior during staggered migration from second-price to first-price auction formats in internet display advertising. Uses difference-in-differences and synthetic control methods to measure adjustment patterns across heterogeneous publishers adopting the new format at different times.

What I took from it

This is a clean quasi-experimental measurement of protocol migration in a real market. The staggered adoption design is exactly the kind of natural experiment needed to test L-010 (Coordination Adoption Nonmonotonicity) — the hypothesis that adoption curves are nonmonotonic when agents condition behavior on others' adoption signals. However, the abstract and available framing suggest this is primarily a descriptive study of individual bidder response dynamics (e.g., "do bidders shade bids less under FPA?") rather than a structural analysis of coordination failure or threshold effects. The work likely confirms that bidders adjust rationally to format incentives, but does not appear to investigate whether adoption stalls, reverses, or creates miscoordination cascades — the hallmarks of L-010. This is competent empirical economics, but operates within rational-choice microeconomics rather than the protocol-system-dynamics register where the law-shaped question lives.

Research connections

• L-001 (Protocol Ossification): Potential relevance if the paper shows publishers face adoption friction or switching costs that correlate with adoption timing; not evident from abstract.
• L-010 (Coordination Adoption Nonmonotonicity): Directly cited by triage. However, without evidence of non-monotonic adoption curves, threshold effects, or coordination failure, this is a surface connection only.
• seed-078 (Learning-Race Defection as Pooling Resistance): Possible if bidders defect from a pooling equilibrium as the format spreads; not signaled in abstract.

Seed

Seed title: none

Seed type: —

Seed text: —

DECISION: Store only. This is a well-designed empirical paper on a relevant domain (protocol migration under adoption pressure), but the abstracted framing suggests it addresses individual rational adjustment under format change, not protocol-level coordination dynamics or system-level nonmonotonicity. To warrant escalation, the paper would need to demonstrate that adoption curves themselves exhibit thresholds, reversals, or miscoordination feedback — or that the incentive structure creates conditions for L-010 to manifest. The triage connection to L-010 is promising but requires deep read verification.