Stochastic Stability in Networks and Markets

• Applicant: Professor Dr. Carlos Alós-Ferrer
• Subject Area: Economic Theory
• Term: from 2016 to 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 273553573

• Final Report Year: 2024

Final Report Abstract

The theory of learning in games has become one of the major modeling tools for the analysis of bounded rationality and economic behaviour. Most models rely on a technique known as stochastic stability, where agents are endowed with boundedly rational behavioural rules (for example imitation, myopic best reply, or reinforcement) and the possibility of decision mistakes is explicitly incorporated. Formally, this results in a family of perturbed, discrete-time Markov chains which enable the analysis of long-run behaviour and the stability of economic outcomes. The project has examined two main research lines within this research field. The first has considered the selection, evolution, and design of market institutions (as e.g. B2B market platforms) when traders are boundedly rational. Here we have obtained results for general market institutions which might even violate the “law of one price” and new theoretical insights for the case of production with constant returns to scale, which point out that coordination on efficient trading institutions is not guaranteed. Using simulations, we have also found out that the boundary conditions on previous theoretical results in dynamic oligopolies with boundedly rational firms were not binding. Laboratory experiments have shown the relevance of the approach for actual human behavior, and especially the impact of techologies: under constant returns to scale, multiple institutions remain active and coordination is slower than under decreasing returns to scale. The main insight is that sellers trade off higher efficiency in a market with dwindling profits for biased-up profits in a market with vanishing customers. As a consequence, also empirically efficiency alone does not suffice to guarantee coordination on a single market institution if the surplus distribution is asymmetric. The second line concentrated on games on networks, that is, has taken explicitly into account that economic interactions are local in nature. The aim to obtain full characterizations of the network characteristics leading to each class of economic outcomes has proven elusive at a theoretical level, and the focus has shifted to extensive simulations (in excess of one million) carried out with supercomputers, which have helped identify a number of general principles underlying the selection of efficient conventions.

Publications

• On the convergence of logit-response to (strict) Nash equilibria. Economic Theory Bulletin, 5(1), 1-8.
  Alós-Ferrer, Carlos & Netzer, Nick
  
• Cournot vs. Walras: A reappraisal through simulations. Journal of Economic Dynamics and Control, 82, 257-272.
  Alós-Ferrer, Carlos & Buckenmaier, Johannes
  
• Market selection by boundedly-rational traders under constant returns to scale. Economics Letters, 153, 51-53.
  Alós-Ferrer, Carlos & Kirchsteiger, Georg
  
• Trader matching and the selection of market institutions. Journal of Mathematical Economics, 69, 118-127.
  Alós-Ferrer, Carlos & Buckenmaier, Johannes
  
• Imitation, network size, and efficiency. Network Science, 9(1), 123-133.
  Alós-Ferrer, Carlos; Buckenmaier, Johannes & Farolfi, Federica
  
• When are efficient conventions selected in networks?. Journal of Economic Dynamics and Control, 124, 104074.
  Alós-Ferrer, Carlos; Buckenmaier, Johannes & Farolfi, Federica
  
• Do traders learn to select efficient market institutions?. Experimental Economics, 25(1), 203-228.
  Alós-Ferrer, Carlos; Buckenmaier, Johannes & Kirchsteiger, Georg