The following is a list of current working papers.
Personalized Discounts and Consumer Search
(with Zikun Liu and Jidong Zhou), revise and resubmit at Management Science
The growing availability of big data enables firms to predict consumer search outcomes and outside options more accurately than consumers themselves. This paper examines how a firm can utilize such superior information to offer personalized buy-now discounts intended to deter consumer search. However, discounts can also serve as signals of attractive outside options, potentially encouraging rather than discouraging consumer search. We show that, despite the firm’s ability to tailor discounts across a continuum of consumer valuations, the firm-optimal equilibrium features a simple two-tier discount scheme, comprising a uniform positive discount when the consumer outside option is intermediate and no discount when the outside option is low or high. Furthermore, compared to a scenario where the firm lacks superior information, we find that the firm earns lower profits, consumers search more while their welfare remains unchanged, and total welfare declines. More…
Designing Detection Algorithms for AI-Generated Content: Consumer Inference, Creator Incentives, and Platform Strategy
(with Jieteng Chen and Tony Ke), revise and resubmit at Management Science
Generative AI has transformed content creation, enhancing efficiency and scalability across media platforms. However, it also introduces substantial risks, particularly the spread of misinformation that can undermine consumer trust and platform credibility. In response, platforms deploy detection algorithms to distinguish AI-generated from human-created content, but these systems face inherent trade-offs: aggressive detection lowers false negatives (failing to detect AI-generated content) but raises false positives (misclassifying human content), discouraging good creators. Conversely, conservative detection protects creators but weakens the informational value of labels, eroding consumer trust. We develop a model in which a platform sets the detection threshold, consumers form beliefs from content labels and decide whether to engage, and creators choose whether to adopt AI and how much effort to exert to create content. A central insight is that detection does not affect outcomes continuously: instead, equilibrium structure shifts discontinuously as the threshold changes. At low thresholds, consumers trust human labels and partially engage with AI-labeled content, disciplining AI misuse and boosting engagement. But when detection threshold becomes higher, this inference breaks down, AI adoption rises, and both trust and engagement collapse. Thus, the platform’s optimal detection strategy balances these risks, influencing content creation incentives, consumer beliefs, and overall welfare. More…
Demand Externalities from Co-Location
(with B. Sen, K. Sudhir and N. Yang), revise and resubmit at QME
We illustrate an approach to measure demand externalities from co-location by estimating household-level changes in grocery spending at a supermarket among households that also buy gas at a co-located gas station, relative to those who do not. Controlling for observable and unobserved selection in the use of the gas stations, we find significant demand externalities; on average a household that buys gas has a 7.7% to 9.3% increase in spending on groceries. Accounting for differences in gross margins, the profit from the grocery spillovers is 130% to 150% the profit from gasoline sales. The spillovers are moderated by store loyalty, with the gas station serving to cement the loyalty of store-loyal households. The grocery spillover effects are significant for traditional grocery products, but 23% larger for convenience stores. Thus co-location of a new category impacts both inter-format competitions with respect to convenience stores (selling the new category) and intra-format competition with respect to other supermarkets (selling the existing categories). More…