This paper integrates a randomized controlled trial and structural analysis to investigate how misconceptions about drug efficacy distort consumer decisions in the over-the-counter market and how clearer information can improve consumer welfare. A key challenge is that beliefs about efficacy are unobservable, which complicates the assessment of consumers’ initial perceptions and responses to new information. To address this, I gathered pairwise product comparisons from a control group and three treatment arms to develop product-level measures of efficacy beliefs. The most effective treatment — emphasizing equivalent efficacy —increases substitution elasticities by 22% on average, saves the average consumer $1.09, and generates a welfare gain of $1.20 per consumer. It results in $6.5 million annual savings across the 50 largest markets. However, this approach also leads to second-degree price discrimination based on consumers’ preferences for symptom labels, highlighting the need for social planner to make informed policy decisions.

No one can remain neutral regarding health information about their conditions. While some focus on the worst-case scenario, others seek justifications for not visiting the doctor. The complexity and diversity of current online health information (OHI) cater to both groups. This paper explains this behavior through the modification of the information acquisition model and further examines its relationship with healthcare utilization. It introduces a rational inattention model with a modified cost function that accounts for individual heterogeneity in perceiving the absence of illness. The model demonstrates that information acquisition costs are influenced by individuals’ concerns about overlooking signs of illness, causing information sources to act as thought accelerators rather than purely educational material. Empirical analysis supports the theoretical framework. First, the data shows that OHI usage generally leads to higher healthcare utilization. Second, OHI users who are concerned about missing illness signs use healthcare services more than those less worried. Third, the statistically insignificant difference in how OHI affects healthcare usage between groups that might benefit from increased concern and those that do not suggests that patients’ overconfidence, rather than knowledgeable worry, drives higher healthcare use. These findings raise critical policy questions about managing OHI-induced overconfidence and offer recommendations for enhancing physician–patient interactions in the context of OHI.

This paper discuss theoretically and empirically second degree price discrimination under the condition when companies use pricing strategies to prevent consumers form learning about equivalence in quality.

This study examines whether Medicare reimbursement schemes play a critical role in addressing market failures, particularly the slow adoption of innovative and consumer-beneficial technologies. Additionally, it proposes policy interventions to accelerate adoption rates. Using the case of Transcatheter Aortic Valve Replacement (TAVR), an initially costly procedure where Medicare adjusted hospital reimbursements later than physician payments, we analyze the rate and pattern of adoption. Empirically, we find that 70% of adopted by 2020 hospitals adopted TAVR before the reimbursement change, suggesting that factors beyond direct financial incentives influenced early adoption. The diffusion of TAVR does not follow the standard bell-shaped curve but instead exhibits distinct spikes in adoption — first year after the DRG code creation. To explain this pattern, we develop a structural model that accounts for cost reductions due to learning effects, anticipation of future Medicare reimbursement changes, and contractual relationships between hospitals and physicians. All this patterns are observed in the reduce form evidence. By separately identifying these key drivers of innovation adoption, our model allows us to simulate diffusion patterns under alternative Medicare reimbursement policies.