Atahan Afsar
Postdoctoral Researcher,
The Institute for Futures Studies, Stockholm/Sweden
email: atahan.afsar@hhs.se
Bio
I am currently a Postdoctoral researcher at the Institute for Futures Studies in Stockholm, Sweden. I completed my Ph.D. at Stockholm School of Economics in 2024.
My research areas include Behavioral Economics, Computational Economics, Game Theory, Microeconomic Theory, and Political Economics. My current research focuses on (i) theoretical and empirical analysis of models of learning in games, (ii) information acquisition and information design, with an emphasis on political and normative behavior, and (iii) Political economics, with an emphasis on parliamentary democracies.
Publications
A Behavioral Hybrid New Keynesian Model: Quantifying the Importance of Belief Formation
Economic Modelling, Volume 132, Article 106626
Abstract: Recent evidence points towards significant belief formation frictions and forecast sluggishness. In this paper, we build a bounded rationality New Keynesian model, estimated to match the degree of forecast sluggishness present in the data. We find that bounded rationality induces enough myopia and intrinsic persistence, diminishing the influence of consumption habits and price indexation. Additionally, the bounded rationality model generates impulse response dynamics to monetary policy shocks that resemble those observed in empirical estimations. This study highlights the significance of bounded rationality in capturing real-world dynamics and provides valuable insights into the role of belief formation frictions in macroeconomic modeling.
Working Papers
Judgement Aggregation and Rational Inattention
Latest version: June 2024. Draft available upon request.
Abstract: From central banks to company boards, important decisions are made in committees where members are expected to be informed about the alternatives. We analyze the effects of strategic flexible information acquisition for collective decision-making under uncertainty in finite committees with two alternatives. Extending previous literature that assumes equal attention across states, which is arguably a strong assumption when members are not ex ante indifferent between type-I and type-II errors, we allow committee members to buy costly Blackwell experiments prior to voting. The cost function is based on Shannon entropy, aligning with the standard Rational Inattention framework. We recast the strategic considerations of each player as a single-agent rational inattention problem, utilizing the results of Matějka and McKay (2015) and Caplin, Dean, and Leahy (2019) to establish the existence of a unique symmetric informative equilibrium. We show that although the acquired information decreases as the committee size increases in a sequence of equilibria, the probability of choosing the correct alternative remains higher than if members vote without acquiring any information.
A Framework for Spatial Political Analysis
[PDF] Latest version: September 2023.
Abstract: In order to understand current political events, it appears that the multidimensionality of political ideology is essential. Political candidates and parties differ not only in the traditional left-right dimension, but also concerning environmental issues, gender roles, law and order, migration, defense, etc. In addition, an arguably important phenomenon in democracies is abstention from voting because of alienation. We here outline a mathematical framework for the analysis of\ direct and representative democracy, political competition, political power, and coalition formation in a multidimensional setting with endogenous voter abstention. Our approach combines elements from non-cooperative game theory, cooperative game theory, geometry, and random-utility theory. We illustrate the framework by numerical examples and by applying it to data for the Swedish parliament.
Work in Progress
Compete to Persuade Inattentive Voters
Abstract: What is the effect of one or more information providers on the quality of collective decision reached by a rationally inattentive committee? Afsar (2024) shows that the full aggregation of distributed information across voters, which Condorcet saw as a justification of universal suffrage under majority rule, fails to arise as we increase the size of the committee if the information each voter acquires is endogenous and costly, since in the limit the probability of being pivotal is zero, and thus no voter acquires any information. However, in contrast to Afsar (2024), the standard committee setting assumes that voters receive some free information. This role has traditionally been assumed by public broadcasters, but their role has been weakened significantly in recent times. Here, we extend Afsar (2024) to a setting with one or more information designers together with a committee of rationally inattentive voters. This work also extends the setting of Matyskova and Montes (2023) with a single sender, a single rationally inattentive receiver, as well as the setting of Alonso and Camara (2016) with a single sender and a committee of receivers.
Comparing Models of Learning in Games
Dynamics of Political Discourse: Agenda Setting in Multidimensional Electoral Competition
Experimental Cost of Information and Stochastic Choice Probabilities
Iranian Protests and Gender Norms
Protests, Image Concerns, and Preference Aggregation
Voting with Dependent Information Acquisition
Teaching Experience
Lecturer, Ph.D. Preparatory Mathematics
Teaching Assistant, Ph.D. Mathematics I, Professor Jörgen Weibull
Teaching Assistant, Ph.D. Microeconomics II, Professor Karl Wärneryd
Teaching Assistant, M.Sc. Advanced Microeconomics, Professor Karl Wärneryd