Switching Costs, Brand Premia and Behavioral Pricing in the Pharmaceutical Market (Download) (Online Appendix)
[Unicredit & Universities Best Job Market Paper Award, November 2018]
This article examines the market power of branded prescription drugs faced with generic competition. Using prescription-level and matched socioeconomic panel data of the entire Swedish population between 2010 and 2016, I provide evidence for the key role of switching costs. A discontinuity surrounding patent expirations establishes that the effect is causal. Further, by comparing patients with and without medical education, I show that those without medical education experience higher brand premia. A unique feature of the Swedish market allows me to rule out patients’ inattention due to information costs as a source of market power. Therefore, switching costs and perceived quality differences are the key determinants of market power. I then estimate a dynamic oligopoly model with forward-looking firms which is used in counterfactual studies of the effect of switching costs and perceived quality differences on prices. First, an increase in the length of procurement mimics a reduction of switching costs and increases prices. While the effect of switching costs on prices in theory is ambiguous, moderate switching costs and sufficient competition for new patients increase competitive pressure. Second, if everyone acts as a medical expert and experiences fewer brand premia, prices decrease.
This paper investigates price patterns of off-patent pharmaceuticals in Sweden. I show that price dynamics are dependent on the number of competitors in the market. The price patterns follow predictions from a model of dynamic price competition in which the demand for pharmaceuticals incorporates the known biases of consumers: habit persistence and brand preferences. Using the regulated market of Swedish pharmaceuticals, I show that price may help in identifying possible tacit collusion by manufacturers in markets where consumers experience behavioral frictions.
In most democracies members of parliament are either elected over a party list or by a district. We use a discontinuity in the German parliamentary system to investigate the causal effect of a district-election on an MP’s conformity with her party-line. A district-election does not affect roll call voting behavior causally, possibly due to overall high adherence to party voting. Analyzing the parliamentary speeches of each MP allows us to overcome the high party discipline with regard to parliamentary voting. Using textual analysis and machine learning techniques, we create two measures of closeness of an MP’s speeches to her party. We find that district-elected members of parliament do not differ, in terms of speeches, from those of their party-peers who have been elected through closed party lists. However, both speeches and voting correlate with district characteristics suggesting that district-elections allow districts to select more similar politicians.
Pregnancy and Alcohol Purchases: Evidence from Scanner Data (SSRN)
(Together with Elle Parslow)
We analyze household-level changes in alcohol consumption in response to pregnancy. Using scanner data, we identify households with a pregnant household member. Within an event study and a dynamic difference-in-differences estimation, we find that during a first pregnancy, households reduce their alcohol purchases by 35%. After pregnancy, purchases of alcohol are 31% lower than before pregnancy. We do not find any effect during the second pregnancy. We argue that lower consumption during pregnancy changes habits and reduces consumption in the long term. We exclude other explanations and comment on policy implications.
Does Precise Case Information Limit Precautionary Behavior? Evidence from COVID-19 in Singapore (SSRN)
(Together with Matthew Shapiro)
Limiting the spread of contagious diseases can involve both government-managed and voluntary efforts. Governments have a number of policy options beyond direct intervention that can shape individuals’ responses to a pandemic and its associated costs. During its first wave of COVID-19 cases, Singapore was among a few countries that attempted to adjust behavior through the public provision of detailed case information. Singapore’s Ministry of Health maintained and shared precise, daily information detailing local travel behavior and residences of COVID-19 cases. We use this transparency policy along with device-level cellphone data to quantify how local and national COVID-19 case announcements trigger differential behavioral changes. We find evidence that individuals are three times more responsive to outbreaks in granularly defined locales. Conditional on keeping infection rates at a manageable level, the results suggest economic value in this type of transparency by mitigating precautionary activity reductions.
Work in Progress:
Obfuscation and Rational Inattention
(Together with Johannes Kasinger)
Retail Pharmacies and Drug Diversion during the Opioid Epidemic
(Together with Xuan Zhang)