arXiv subjects

Walter W. Zhang

Publications and source records attributed to Walter W. Zhang.

4 recordsLinked to original sources

arXiv2026

Buy Now, Pay Later: Academic Insights and Open Policy Questions

Buy Now, Pay Later (BNPL) has moved from a novelty product to mainstream consumer finance in the space of a few years. Before 2020, BNPL was a niche offering at the checkouts of online fashion merchants. BNPL then experienced substantial growth (Consumer Financial Protection Bureau, 2022; Salem & Udis, 2025), coinciding with increased online shopping since the onset of the COVID-19 pandemic in 2020. BNPL is now a pervasive payment option. In 2026, you can use BNPL to delay and split up payments for anything ranging from a pizza to your rent, and many items in between. As the market has grown, so has the academic literature. There is now sufficient research into BNPL to take stock and review what we have learned. In this article, we summarize insights from the economics, finance, and marketing academic literature. While we have learned a great deal about BNPL in the last few years, we also discuss many remaining open questions and current concerns impacting policy decisions. We start by explaining what BNPL products are and the business models of BNPL lenders. Then, we summarize the economics and psychology behind why some consumers use these products. Next, we describe the customer base that uses BNPL. We show the effects of BNPL on consumers, an area of research with substantial policy interest. Another area of policy interest is whether BNPL loans should be reported in a consumer's credit report. We explain why BNPL is typically not on credit reports and its implications. Finally, we offer some concluding thoughts.

econ.GN↗

arXiv2026

Valuing Winners: When and How to Correct for Selection Bias in Randomized Experiments

Decision-makers often deploy the best-performing treatment from a randomized experiment, creating a winner's curse: selection favors treatments whose observed outcomes are high partly because of statistical noise, so the naïve estimate of the winner is upward biased. We distinguish two forms of winner's curse, bias relative to the true best treatment (global) and bias relative to the selected treatment's true mean (selective), and link them to regret from deploying a suboptimal treatment. This framework defines seven decision-relevant evaluation targets: mean bias, mean squared error, and confidence interval coverage for the global and selective winner's curse, and mean regret. We then show that methods that perform well on one target can perform poorly on others, so corrections should be matched to the manager's objective. Across simulations with varying effect sizes, multiple-arm settings, and data calibrated to an online A/B testing platform, no method dominates uniformly: the plug-in estimator performs best when treatment differences are large, cross-fitting performs best when treatments are similar, and resampling methods often achieve low mean squared error for moderate differences. We also introduce an adaptive empirical likelihood procedure that delivers asymptotically valid confidence intervals across settings without the tuning sensitivity of resampling-based methods.

econ.EM↗

arXiv2025

Optimal Comprehensible Targeting

Developments in machine learning and big data allow firms to fully personalize and target their marketing mix. However, data and privacy regulations, such as those in the European Union (GDPR), incorporate a "right to explanation," which is fulfilled when targeting policies are comprehensible to customers. This paper provides a framework for firms to navigate right-to-explanation legislation. First, I construct a class of comprehensible targeting policies that is represented by a sentence. Second, I show how to optimize over this class of policies to find the profit-maximizing comprehensible policy. I further demonstrate that it is optimal to estimate the comprehensible policy directly from the data, rather than projecting down the black box policy into a comprehensible policy. Third, I find the optimal black box targeting policy and compare it to the optimal comprehensible policy. I then empirically apply my framework using data from a price promotion field experiment from a durable goods retailer. I quantify the cost of explanation, which I define as the difference in expected profits between the optimal black box and comprehensible targeting policies. Compared to the black box benchmark, the comprehensible targeting policy reduces profits by 7.5% or 23 cents per customer.

econ.GN↗

arXiv2025

Coarse Personalization

With advances in estimating heterogeneous treatment effects, firms can personalize and target individuals at a granular level. However, feasibility constraints limit full personalization. In practice, firms choose segments of individuals and assign a treatment to each segment to maximize profits: We call this the coarse personalization problem. We propose a two-step solution that simultaneously makes segmentation and targeting decisions. First, the firm personalizes by estimating conditional average treatment effects. Second, the firm discretizes using treatment effects to choose which treatments to offer and their segments. We show that a combination of available machine learning tools for estimating heterogeneous treatment effects and a novel application of optimal transport methods provides a viable and efficient solution. With data from a large-scale field experiment in promotions management, we find our methodology outperforms extant approaches that segment on consumer characteristics, consumer preferences, or those that only search over a prespecified grid. Using our procedure, the firm recoups over $99.5\%$ of its expected incremental profits under full personalization while offering only five segments. We conclude by discussing how coarse personalization arises in other domains.

econ.EM↗