My research interests are anything to do with applied and theoretical microeconomics, especially industrial organization and the economics of information.
Here is a brief research statement (pdf).
My current work focuses on two themes. The first is optimal stopping problems, which include a range of economic problems such as investment under uncertainty, job search with variable employment opportunities, and selling with unknown demand. In all of these situations, an economic agent faces a trade-off between exploration---waiting for better information in the future---and exploitation---taking what is available now. Crucially, in many of these situations, the economic agent is able to influence the amount of information it receives: it can experiment as it explores.
The second theme is understanding situations of economic relevance that have indeterminate outcomes. More precisely, when do economic models have a definite (unique) prediction about how economic agents will behave, and when are they able only to say that a number of outcomes are possible?
On this page, I describe some of the academic research projects in which I am currently involved. Some of the work described has been supported financially by the ESRC under the Research Fellowship Award R000271265 and Research Grant RES-062-23-0925.
I also work with a number of regulators (OFCOM and Ofgem in the UK, Comreg in Ireland) on regulation and competition issues.
Joint work with Akos Valentinyi provides a sufficient condition for existence and uniqueness of equilibrium, which is in monotone pure strategies, in games of incomplete information. First, we show that if each player's incremental ex post payoff is uniformly increasing in its own action and type, and its type is sufficiently uninformative of the types of its opponents (independence), then its expected payoff satisfies a strict single crossing property in its own action and type, for any strategy profile played by its opponents. This ensures that a player's best response to any strategy profile is a monotone pure strategy. Secondly, we show that if, in addition, there is sufficient heterogeneity of the conditional density of types, then the best response correspondence is a contraction mapping. This ensures equilibrium existence and uniqueness. In contrast to existing results, our uniqueness result does not rely on strategic complementarities; this allows for a wider range of applications.
In joint work with Juuso Valimaki (Helsinki School of Economics and Southampton), we analyse smooth stopping problems. Optimal stopping problems are used as the workhorse model in diverse fields of economics. Employees' search for new jobs, optimal selling procedures over time, models of irreversible real investments are a few examples where this family of models features prominently. The success of these models stems from the fact that they incorporate in an easily understood manner the trade-off between the risky wait for potentially superior alternatives in the future and the riskless acceptance of the current alternative.
Analytically, such problems are often assumed to take place in a stable environment, and in particular, it is assumed that current decisions do not affect the statistical structure of future opportunities. While this may be a useful first approximation of many actual situations, it is also clear that in many other contexts, such independence is crucial. In this project, we enlarge the domain of problems to include the case where there is learning about future opportunities. The framework is also flexible enough to allow for instantaneous investment decisions for the decision makers that change the future environment. The key analytical insight that allows us to make progress in characterizing optimal stopping problems of this type is that we require both the stopping probability and the future evolution of the environment to be continuous in the instantaneous choices.
Again with Juuso Valimaki (Helsinki School of Economics and Southampton), we look at the provision of incentives in settings where a principal hires an agent to complete a project. The objective is to analyse the dynamics of contracts: for example, does the completion bonus decrease over time?; does the principal use deadlines? More generally, we are looking to develop a framework that is sufficiently simple that (i) the dynamic incentives can be identified clearly; (ii) more complicated settings, e.g., with both adverse selection and moral hazard over time, can be analysed. This allows us to look at situations where agents differ in the quality of the completed project. The principal then aims to provide effort incentives while at the same time learning about the quality of the agent that is employed.