Government policy, strategic consumer behavior, and spillovers to retailers: The case of demonetization in India, with Pradeep Chintagunta and Bhuvanesh Pareek. Marketing Science 41(6), 2022.

This paper studies strategic consumer shopping behavior in response to a macroeconomic policy in the case of currency reform, and quantifies its unintended consequences for retailers vis-a-vis the policy goal. Using transaction-level data from a large retail chain in India, we document consumer strategies that leverage the presence of retailers to avoid costs associated with the country's currency reform policy. We observe both an increase in returns of cash-purchased items that were bought before demonetization (strategic returns) and a spike in final (unreturned) cash purchases with soon-to-be-illegal notes (strategic purchases). Both practices serve consumer incentives either to receive legal notes from the retailer or to avoid depositing cash in formal bank accounts. Our analysis suggests that strategic consumers benefited the retail chain overall while partly hindering the intended effect of the policy, leaving more than 20 million INR (0.3 million USD) of demonetized notes outside the formal tax network through this retail chain only; when we scale up the estimates to the country's market size, the estimated total impact is at least 100 billion INR (1.5 billion USD). Our findings (i) offer implications for policy makers by quantifying a wide spillover effect of government interventions that goes beyond the target group, and (ii) document a new role of the retail industry - of absorbing, and facilitating a response to, macro shocks.

Sustainability and Strategic Differentiation in Unregulated Consumer Goods Market, with Kristina Brecko, 2026. R&R, JMR.  

- Presented at 2022 Stanford-Berkeley IO Fest, 2023 Marketing for Sustainability Conference, 2023 VQMS, 2023 Choice Symposium, 2024 INSEAD Marketing  seminar, 2024 Cornell Marketing seminar, 2024 Bass FORMS Conference, 2024 Stanford Human Behavior and Sustainability Conference, 2024 Stanford BGS FlashTalks, 2024 Yale InsightsOn Conference, 2025 International Conference at CUHK, 2025 CityU (Hong Kong) Marketing seminar
- Previously presented under the title of "Voluntary provision of sustainability claims: Evidence from Consumer Packaged Goods", and "Supply and Demand for Sustainability Features: Drivers and Deterrents under Voluntary Labeling"
- Funded by Stanford BGS grant ($50,000)

Using large-scale field data from the U.S.  personal care categories from 2012 to 2019, we find that sustainability remains a relatively low-priority feature for the average consumer, leading to firms' strategic differentiation in the absence of regulation. Descriptive and structural analysis reveal that sustainability features are primarily offered by smaller brands through premium-priced products with high marginal cost, while dominant brands tend to offer fewer and lower-cost sustainable alternatives. This pattern reflects underlying consumer demand. Mass-market brands -- serving a broad, price-sensitive consumer base that places low value on sustainability -- have limited incentives to invest in high-cost sustainable products. In contrast, fringe brands -- including those owned by large multi-brand manufacturers -- use sustainability to differentiate and target a niche but growing segment of consumers who value it more highly. Implications for brands and policymakers are discussed. 

Marketing a Highly Durable Technology in a Developing Country, with Amrita Kundu and Erica Plambeck, 2026. (Preliminary draft available upon request)

- First and corresponding author. Presented at 2024 China-India Insights Conference (plenary session), 2024 INFORMS Annual Meeting, Stanford Celebration of Scholarship, University of Hawaiʻi at Mānoa, KAIST, Business and Marketing Analytics Seminar, 2026 Bass  FORMS conference, POMS 2026
- Funded by Asian Development Bank ($77,000 grant) and Stanford BGS Research Grant ($20,000) 

We propose and evaluate market-based solutions to promote the adoption of a highly durable product in a developing country, focusing on batteries for electric three-wheeled vehicles (EVs) in Bangladesh. The market is dominated by short-lived lead-acid batteries - lasting less than a year - whose manufacturing and recycling generate severe lead pollution. Highly durable alternatives such as high-performance lithium batteries are available but hold negligible market shares. Using an incentivized choice experiment, we marketed batteries with different durability claims (3-year or 7-year) and randomly offered combinations of a warranty, a buyback program, and microfinance loans with varying maturities. We find that a warranty alone does not increase willingness to pay (WTP) and a buyback offer produces only modest gains. In contrast, a microfinance loan substantially increases WTP and renders warranties effective -- but only when loan duration aligns with the promised battery life. Combining elicited WTPs with survey data suggests that a loan whose duration matches a product’s durability claim creates demand not only by relaxing financial constraints but also by reducing distrust in the claimed durability.

Financing & Policy Innovations to Reduce Lead Emissions Through a Transition to Lead-Free Batteries, with Erica Plambeck, Qiong Wang, and Amrita Kundu, 2026. Submitted. 

Problem Definition: Bangladesh has widespread lead poisoning associated with its circular lead-acid battery (LAB) industry, and negligible use of lead-free batteries. How could microfinance institutions, development banks and the government of Bangladesh induce adoption of lead-free batteries and reduce lead emissions from the LAB industry?
Method and Results: In a model of a circular LAB industry with formal and informal processes, we introduce lead-free batteries. This is based on a battery choice field experiment, in which introducing microfinance loans creates the demand for lead-free batteries. We derive the transient dynamic equilibrium salvage value of Used LAB (ULAB), the fractions of EVs with lead-free batteries vs. formally-manufactured LAB vs. informally- manufactured LAB, and the resulting lead emissions over time. The introduction of lead-free batteries can temporarily increase lead emissions through dumping of excess ULAB or a shift to informal lead emission-intensive processes for recycling ULAB.
Policy Implications: We recommend a smelting fee scheme: Provide a subsidy per ULAB processed by a qualified low-emission formal smelter, and impose a fee on each new LAB sold domestically or exported. This reduces lead emissions during a transition to lead-free batteries. In numerical scenario analysis for Bangladesh, the smelting fee scheme also generates net positive government revenue. As an alternative, concessionary financing for ULAB inventory incentivizes low-emission smelting and reduces emissions. Microfinance loans for durable LAB slightly reduce lead emissions. Greatly reducing the tariff on lead-free batteries greatly reduces government revenue yet, during the transition period, fails at lead emission reduction.

Are Display Ad Content and Landing Page Ad Content Complements or Substitutes? A Field Experiment, with Kirthi Kalyanam, 2025. Submitted.

- Presented at 2025 Workshop in Management Science, 2025 Marketing Dynamics Conference, 2025 Bay Area Marketing Symposium
- MSI Working Paper Series 2025, Report No. 25-151

We study whether sequential ad content - display ads and landing page ads - act as substitutes or complements using a field experiment with a semiconductor firm. Ad headlines emphasizing product quality or transactional ease were randomly assigned across two ad placements in eight markets. Using pre-experiment organic traffic to infer prior knowledge about the advertised products, we find that the relationship between the two contents varies with prior information. In low-information markets, two quality-focused contents across placements showed substitutability, consistent with ad content serving as a source of information. In high-information markets, they exhibited complementarity, suggesting their preference-matching role. Decomposing quality-focused content reveals that content emphasizing horizontal match drives information provision, whereas content highlighting vertical quality supports preference matching. Our findings suggest that even coarse, market-level coordination of sequential ad content substantially improves online customer acquisition - lowering cost per action by 37%.

Customer retention under imperfect information, 2025.  R&R, QME. (Draft available upon request)

- Presented at 2019 Stanford Marketing seminar, 2019 NYU Marketing seminar, 2019 Columbia Marketing seminar, 2019 Minnesota Marketing seminar, 2019 UT Dallas Marketing seminar, 2021 Yale Marketing seminar, 2021 KAIST Marketing seminar, 2021 QME Conference, 2022 Emory Marketing seminar, 2022 Berkeley Marketing seminar, 2022 Northwestern Marketing Seminar

Firms often see customers churning quickly after limited product experiences. This paper examines whether such early churn is solely driven by customers' low preferences for a given firm or influenced by incomplete information about available products. Using ticket purchase data from a major U.S. symphony center, I find that concert choices exhibit within-customer changes over time that align with predictions under consumer learning. These patterns suggest that 1) first-time customers have incomplete information about concerts and 2) the choices of experienced customers reflect a concert's true average consumption utility  ("concert value"). Reduced-form analyses show that the estimated concert values inferred from experienced customers' choices are not correlated with first-timers' initial concert choices but are strongly correlated with their subsequent churn, indicating that an initial visit made under imperfect information shapes a customer's expectations about future concerts.  Counterfactual analyses using a structural model highlight the challenge of regaining customers after an initial low-value experience, emphasizing the importance of pre-visit interventions (e.g., introductory marketing, assortment design for first-time customers) in reducing early churn.

Valuing brand collaboration: Evidence from a natural experiment, with Sanjog Misra and Bradley Shapiro, 2020. 

We study complementarities between brands in the context of collaborations across museums. Over the course of our sample, one major museum with a highly recognized brand closed temporarily and sequentially collaborated with two established local museums. With individual panel data on museum memberships around these events, we measure how collaborations affect demand using an empirical framework of complementarities that are newly applied to the branding context. We observe two counter-acting demand patterns. First, customers with no history of buying membership from either museum enter the market, suggesting brand complementarities. Second, a sub-group of customers who previously purchased from either or both of the museums display decreased demand, consistent with brand dilution. Any structural approach that models the demand for collaboration with existing preferences for separate brands fails to create accurate demand predictions. The magnitude of the offsetting forces varies between collaboration events, which makes demand prediction even more challenging. These results call for a theory of brand being beyond a fixed utility primitive and have implications for counterfactuals that involve combining or altering of brands.

Learning to Reduce Food Waste, with Samina Lutfeali and Szu-chi Huang, 2025. Manuscript in progress. 

- On-campus field experiment completed in June 2025
- In collaboration with Stanford Food Institute and Residential & Dining Enterprises 

In collaboration with four Stanford University dining halls, we test whether information interventions reduce food waste in the focal dining halls and generate spillover effects to non-treated dining halls. Two treatments were implemented in the field experiment: (i) displaying the weight of discarded food (\textit{waste} condition), and (ii) adding prompts encouraging mindful food selection \textit{(waste + selection} condition). Food waste fell significantly in treated dining halls, with the largest reduction under the combined treatment. Spillover effects also reduced waste in non-treated dining halls, though short-lived. The findings suggest that students learn to reduce food waste in response to information interventions, but the learning fades quickly once the intervention is removed. We discuss implications for marketers and policymakers aiming to reduce institutional food waste.

Reducing Food Waste in Hospitality, with Samina Lutfeali and Szu-chi Huang, 2025. Field experiment in progress.

- Field experiment in progress (Oct 2025  - Feb 2026)
- In collaboration with six hotels in Dubai

Local Sustainability Mandates, with Kristina Brecko, 2025. Data collection and analysis in progress. 

Range of Information, with Yu Ding, 2025. Pilot studies in progress. 

Syncing Preferences: Dynamic Reference Formation in Digital Service Platforms, with Abhishek Deshmane, Noah Askin, and Khwan Kim, 2025. Analysis in progress.