I am an assistant professor of economics at the University of Vienna. I work in the fields of applied microeconomics, information economics, political economy and industrial organization.
This paper proposes two novel ways of testing for moral hazard in settings where adverse selection is potentially present using panel data on insurance. The first way uses variation in risk between different items owned by the same person with different levels of coverage to "factor out" the person's risk type and separate variation that can be only due to moral hazard. The second way uses some contractual features of many real-world contracts to establish theoretical intertemporal patterns in risk that can be attributed to moral hazard but not to adverse selection and then derives an econometric test of these patterns. I apply my methods to an insurance data from a large virtual world EVE Online and find that daily hazard rates of insured items are, on average, six-fold compared to uninsured items due to moral hazard.
This paper shows how heterogeneous beliefs in the population of voters about the appropriate policy can alone explain the following three stylized facts: First, in a number of recent elections parties or candidates that are often labeled populist have been able to attract a winning coalition of voters with ambiguous, unrealistic or self-contradicting platforms. Second, the support for these candidates has come from both ends of the traditional voter spectrum. Third, many of these candidates have lost half or more of their support after getting into office. The model also allows discussing the role of political marginalization in the rise of populist candidates.
Work in Progress
Political Ambiguity and Heterogeneous Beliefs in the Lab (with James Tremewan and Alexander Wagner)
We use a laboratory experiment to study how appeal of ambiguous platforms depends on the uncertainty that voters have about candidates' preferences. The experiment builds on the predictions from Tolvanen (2017). Our initial results show that even though most extreme voters behave as predicted by the theory and are willing to vote for ambiguous extremists, most extreme candidates do not realize this. We study how much of this puzzling behavior can be attributed to correlation neglect and how much to mistakes in higher order beliefs.
Bargaining with Private Deadlines (with Nemanja Antic, early draft available)
This paper studies an alternating offers bargaining model where the players have private information about their own deadline after which they have to accept any non-negative offer. Such situations arise, for example, when a firm is bargaining with its employees and both sides are facing fixed costs and finite budgets. We construct a highly intuitive screening equilibrium that has delay. The equilibrium is wasteful in the strict sense that all types of both players would prefer to commit to verifiably reveal their private information if such commitment device was available. The equilibrium highlights the potential benefits from third-party arbitration assuming that the arbitrator can serve the role of this commitment device. We also show that outcome becomes essentially unique when the parties become increasingly patient.