Department of Economics and Business Economics

Piling Pills? Forward-Looking Behavior and Stockpiling of Prescription Drugs

Research output: Working paperResearch


  • wp17_08

    Final published version, 1 MB, PDF document

This paper provides evidence of forward-looking behavior in the demand for prescription drugs, while relying on registry-based, individual-level information about the universe of Danish prescription drug purchases from 1995–2014. We exploit a universal shift in policy in early 2000 from a flat-rate to a non-linear insurance plan for prescription drugs that incentivizes stockpiling at the end of the coverage year. We extend the original framework of Keeler et al. (1977) and discuss how the institutional features of most health insurance contracts, at least theoretically, incentivize intertemporal substitution in purchases across coverage years. We describe how consumers react to the introduction of the non-linear plan by increasing spending by 80% immediately before the implementation of the new regime. Next, our main analysis takes advantage of the policy experiment to formally analyze behavior immediately prior to the end-of-year reset in the non-linear plan using a difference-in-difference strategy. We provide evidence that consumers react to this reset by stockpiling toward the end of the coverage year: consumers buy what amounts to an additional 20%. We detect heterogeneity in the size of the response by individual-level characteristics, proxies for health status, and drug type. We find no evidence of any immediate adverse health utilization effects associated with the stockpiling. We round off the paper with an analysis of the importance of stockpiling for estimates of price sensitivity. We find that ignoring intertemporal substitution across coverage years inflates price sensitivity estimates by a non-negligible amount.
Original languageEnglish
Place of publicationAarhus
PublisherInstitut for Økonomi, Aarhus Universitet
Number of pages66
Publication statusPublished - 18 Sep 2017
SeriesEconomics Working Papers

See relations at Aarhus University Citationformats

Download statistics

No data available

ID: 117207481