Zhu Mingxi | Searcharxiv

arXiv2025

Learning by Doing: The Case of Online Lending

Online lending, a phenomenon which is becoming mainstream due to the migration of consumer finance to the Internet and the adoption of AI based lending models, is an example of learning by doing. This paper studies optimal policies for a direct online lender. This is an instance of a more general problem: how should a decision-maker experiment sequentially in the face of unknown customer (or other) information? Conventional wisdom suggests the decision-maker should take advantage of sequential learning opportunities by conducting multiple small, lean experiments, each building incrementally on the results of earlier ones. Can a single grand experiment, uninformed by earlier experiments, do as well? We find that lean incremental experiments are optimal when the interest rate is exogenous. However, when we extend the lender's action space to setting both the interest rate and the loan amount, we find conditions under which a single grand experiment is optimal. In both cases, income variability can benefit the lender by enabling more effective experimentation. We also study the consumer segmentation associated with each strategy and show that the lender cannot achieve more than half the profit obtained under perfect information.

econ.TH↗

arXiv2024

Design Information Disclosure under Bidder Heterogeneity in Online Advertising Auctions: Implications of Bid-Adherence Behavior

Bidding is a key element of search advertising, but the variation in bidders' valuations and strategies is often overlooked. Disclosing bid information helps uncover this heterogeneity and enables platforms to tailor their disclosure policies to meet objectives like increasing consumer surplus or platform revenue. We analyzed data from a platform that provided bid recommendations based on historical bids. Our findings reveal that advertisers vary significantly in their strategies: some follow the platform's recommendations, while others create their own bids, deviating from the provided information. This highlights the need for customized information disclosure policies in online ad marketplaces. We developed an equilibrium model for Generalized Second Price (GSP) auctions, showing that adhering to bid recommendations with positive probability is suboptimal. We categorized advertisers as bid-adhering or bid-constructing and developed a structural model for self-bidding to identify private valuations. This model allowed for a counterfactual analysis of the impact of different levels of information disclosure. Both theoretical and empirical results suggest that moderate increases in disclosure improve platform revenue and market efficiency. Understanding bidder diversity is crucial for platforms, which can design more effective disclosure policies to address varying bidder needs and achieve their goals through costless information sharing.

econ.GN↗