Shumpei Goke
Job Market Candidate

Stanford University
Department of Economics
579 Jane Stanford Way
Stanford, CA 94305

Curriculum Vitae

Industrial Organization
Empirical Market Design
Empirical Operations
Data Science

Expected Graduation Date:
June, 2022

Thesis Committee:
Bradley J. Larsen (Primary):

Gabriel Y. Weintraub:

Liran Einav:

Working Papers

Non-Binding Secret Reserve Prices: The Case of Wholesale Used-Car Auctions
I study secret reserve prices in auctions that are non-binding in the sense that the sellers can accept bids below them. Such a reserve price has a bite only when the winning bid exceeds it, in which case the winning bid is accepted without seller's action. This work investigates the motivation for this puzzling practice that many real-world auctions take, such as wholesale used-car auctions. I estimate a structural model of ascending auctions using the auction data in the wholesale used-car market. To microfound seller's decision of the secret reserve price, I posit that the seller has uncertainty as to the value of the item when she sets the reserve price and that this uncertainty is resolved after she observes the auction price. I compare the status quo with two counterfactual auction formats: (i) no reserve prices and the seller gets to accept or reject every winning bid, and (ii) the seller commits to the secret reserve price. I observe very little difference among them in terms of probability of trade, seller's payoff and revenue. I discuss how the current format may be rationalized as reducing transaction costs for asking sellers' confirmation of all winning bids and avoiding sellers' cognitive cost of committing to a reserve price.

Learning New Auction Format by Bidders in Internet Display Ad Auctions (joint with Gabriel Y. Weintraub, Ralph Mastromonaco, and Sam Seljan) [Coming Soon]
We study actual bidding behavior when a new auction format gets introduced into the marketplace. More specifically, we investigate this question using a novel data set on internet display ad auctions that exploits a staggered adoption by different publishers (sellers) of first-price auctions (FPAs), in place for the traditional second-price auctions (SPAs). Event study regression estimates indicate a significant jump, immediately after the auction format change, in revenue per sold impression (price) of the treated publishers relative to that of control publishers, ranging from 35% to 75% of pre-treatment price levels of the treated group. Further, we observe that in later auction format changes the lift in price relative to SPAs dissipates over time, reminiscent of the celebrated revenue equivalence theorem. We take this as evidence of initially insufficient bid shading after the format change rather than an immediate shift to a new Bayesian Nash equilibrium. Prices then went down as bidders learned to shade their bids. We also show that bidders sophistication impacted their response to the auction format change. Our work constitutes one of the first field studies on bidders' responses to auction format changes, providing an important complement to theoretical model predictions. As such, it provides valuable information to auction designers when considering the implementation of different formats.

Designing Efficient Mortgage Foreclosure Sale
Lenders with delinquent mortgages recover their lending by foreclosure, which is a legal process to sell the mortgage property via public auction. In the U.S., mortgage lenders are allowed to bid in such foreclosure auctions, and they win in such auctions very frequently. I study the question of why mortgage lenders win in most of those auctions. I develop a theoretical model of ascending auctions with private values. I find that the lender's optimal bidding strategy is the same as the optimal reserve price of an auction seller, if it is below the debt balance. In other words, the lender exercises monopoly power as would an auction seller, up to the remaining debt. I next derive a mechanism that achieves efficient allocation of the foreclosed property. Finally, I match the model with data for 15,860 foreclosure auctions in Miami-Date County, Florida. The lender's monopoly power partially explains the reason why the lender wins in 88% of the auctions. I also find evidence that mortgage lenders may have a systematically higher valuation of the mortgage property compared to other bidders.

Work in Progress

Will the "Link Tax" Succeed in the US? Collective Bargaining of Newspaper Publishers with Content Distributors
News publishers are concerned that online content distributors such as Google and Facebook are "stealing" page views and ad revenues by linking online news articles without fees from internet users. Accordingly, they advocate for a temporary exemption from antitrust law so they can "collude" and collectively bargain with online content distributors to earn fees for linking news articles. However, even if the legislation is passed, the result of such a collective bargaining is not obvious. This is because the news websites may benefit from page views via online content distributors. Indeed, Europe has seen some failed legislative attempts to force payments ("link tax") from online content distributors to news publishers. Using Nash bargaining framework, I show that the news publishers may potentially succeed in negotiating payments from online content distributors through such a collective bargaining, even if separate negotiation attempts by each news publisher fails to secure payments.