Pavel Zryumov

Lending Competition and Funding Collaboration with Yunzhi Hu
2nd round R&R at the RFS August 2026

We examine competition and collaboration between banks and fintech firms in a market with adverse selection. Banks have cheaper funding, while fintechs have better screening technology. Our innovation is to allow the bank to lend to the fintech, i.e., to finance its competitor. This partnership affects competition through two channels: the funding-cost channel, which lowers fintech funding costs and operates through the winner's curse effect when adverse selection is severe and cost pass-through when it is mild, and the toehold channel, which reduces bank competition incentives. Lenders collaborate when adverse selection is severe but compete when it is mild. Partnership funding always benefits fintechs, while its effects on banks and borrowers depend on adverse selection and the timing and scope of partnership terms.

Divide and Inform: Carving Up a Common Value Among Rival Bidders with Dmitry Orlov
September 2026

A monopolist data seller designs and sells signals about borrower quality to competing lenders. Optimal signals raise lenders' profits, and hence seller revenue, by softening lending competition through two channels. Segmentation negatively correlates good grades across lenders, limiting competition for high-quality borrowers. Deterrence makes a bad grade precise enough that its recipient does not lend at all. The channels conflict: higher precision strengthens deterrence but weakens segmentation. Among all information structures, the optimal one relaxes this trade-off by carving out for each lender a distinct set of core borrowers, priced at the monopoly rate. It thus tells a lender not only whether a borrower is creditworthy but also how hard to compete. Adding lenders weakens both channels, but revenue never falls below that from informing only two of the lenders. A rival data seller cuts the incumbent's fees to a correlation premium without changing its design or the lending equilibrium.

Data versus Information Sales under Financial Constraints with Dmitry Orlov and Andy Skrzypacz
August 2025

We study a dynamic problem of selling data without commitment to a budget-constrained receiver. The sender has access to a data-generating process, informative about a fundamental state, and can sell it either as granular observations (raw data) or summary statistics (information). Properly designed, such statistics ensure the residual uncertainty declines predictably along with the receiver's budget, supporting efficiency under gradual information sales. In contrast, selling raw data poses a risk that future observations increase residual uncertainty, exceeding the receiver's remaining budget. Consequently, selling data is inefficient if the fundamental is discrete and requires excess budget if the fundamental is non-Gaussian.

Exchanging Information with Dmitry Orlov and Andy Skrzypacz
September 2025

We analyze a class of dynamic games of information exchange between two players. Each agent possesses information about a binary state that is of interest to the other player and cares about the other player's actions. Preferences are additively separable over own and the other player's actions. We fully characterize the set of equilibrium payoffs that can be sustained in such games and construct equilibria that achieve those payoffs. We show that gradual information exchange dominates static (one-shot) communication. Moreover, the whole set of outcomes that Pareto-dominate static communication can be supported in equilibrium.

Cream Rises First: Endogenous Entry and Adverse Selection over the Cycle
June 2026
Subsumes working paper "Dynamic Adverse Selection: Time Varying Market Conditions and Endogenous Entry"

I study how endogenous asset creation and a time-varying cost of capital jointly shape liquidity, the quality of financed assets, and efficiency in a dynamic market with adverse selection. When gains from trade are low, owners of good assets delay sale to signal quality, and the market is illiquid. An improvement in conditions triggers a wave of deals that, contrary to standard theories, opens with the highest-quality assets, as withheld inventory clears at once; quality then deteriorates as high prices attract low-grade entrants. Illiquidity is both a cause of and a cure for inefficiency: delay screens entrants and restores first-best origination without intervention.