Research and Market Structure: Evidence from an Antibiotic-Resistant Pathogenic Outbreak

Working Paper: NBER ID: w28840

Authors: Mayank Aggarwal; Anindya Chakrabarti; Chirantan Chatterjee; Matthew J. Higgins

Abstract: Do upstream research shocks impact unconnected downstream product markets? We explore this question using a natural experiment involving a publication that identified a pathogenic outbreak in India involving a carbapenem antibiotic resistant superbug. Consistent with theory, we find that this upstream research shock caused multinational firms selling carbapenem antibiotics in India to reduce their downstream market exposure. Rational antibiotic stewardship implies that we should observe a similar response by domestic Indian firms. Surprisingly, we observe the opposite; domestic Indian firms filled the void in the market left by multinational firms. We confirm this aggregate finding with prescription level data, Indian physicians prescribed fewer focal multinational products relative to domestic firm products. Results are robust to alternate control groups and placebo testing. Implications for antibiotic resistance, global health policy and innovation policy are discussed.

Keywords: antibiotic resistance; pharmaceutical industry; market structure; natural experiment

JEL Codes: I18; L1; L65


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Upstream research shock from the discovery of NDM1 superbug (O36)Multinational firms withdraw from the antibiotic market in India (F23)
Multinational firms withdraw from the antibiotic market in India (F23)Prescriptions for multinational firm drugs decline relative to domestic firm products (F23)
Domestic firms fill the market void left by multinational firms (F23)Domestic firms increase their production (D21)
Withdrawal of multinational firms (F23)Domestic firms increase their output (D21)
Response of domestic firms to market shift (L19)Driven by quantity considerations rather than R&D activities (O39)

Back to index