Andrei DubovikA. Dubovik

Discussion and Working Papers


  • Andrei Dubovik, Alexei Parakhonyak

    We argue that cooperation can become more fragile if (i) there are sufficiently many intermediate levels of cooperation and (ii) players cannot respond with large punishments to small deviations. A failure to credibly commit to disproportionate punishments can stem from legal or political feasibility, or from historical precedent. Specifically, we show that regardless of how patient the players are, any prisoner’s dilemma game can be extended with intermediate levels of cooperation in such a way that full conflict is the only equilibrium outcome of the extended game in a class of strategies with limited punishment.

  • CPB Discussion Paper, 2022
    Andrei Dubovik, Clemens Fiedler, Alexei Parkhonyak

    We study the duration of topics in economics research by looking at how much time passes between publication of textually similar papers. Using the corpus of abstracts of economics papers, as available from the RePEc dataset, we find that most papers match to papers from the same year, indicating strong common trends in the economics literature. Nevertheless, matches as long as 14 years apart are statistically significant, suggesting there are topics that last as long. Finally, the average duration of a match has dropped from around 4 years during 1990–2005 to about 1 year starting in 2010.

  • MPRA Discussion Paper, 2018
    Andrei Dubovik

    I study mergers where each firm owns multiple shops across a country. Presently, the European Commission views every shop, together with the shops from its catchment area, as an isolated market. Such an approach is internally inconsistent. I show how to extend the European Commission’s approach to consistently take overlaps in catchment areas into account. My model is a specialization of the existing network theory that is aimed to be feasible in real merger cases. As a demonstration, I study a past merger case and I find that neglecting overlaps in catchment areas can result in substantial biases.

Run of the Mill

Working Papers
Working Papers