Divergence and convergence in Scarf cycle environments: experiments and predictability in the dynamics of general equilibrium systems

  • Benjamin J. Gillen
  • , Masayoshi Hirota
  • , Ming Hsu
  • , Charles R. Plott
  • , Brian W. Rogers

    Research output: Contribution to journalArticlepeer-review

    3 Scopus citations

    Abstract

    Previous experimental work demonstrates the power of classical theories of economic dynamics to accurately characterize equilibration in multiple market systems. Building on the literature, this study examines the behavior of experimental continuous double auction markets in convergence-challenging environments identified by Scarf (Int Econ Rev 1:157–171, 1960) and Hirota (Int Econ Rev 22:461–467, 1981). The experiments provide insight into two important economic questions: (a) Do markets necessarily converge to a unique interior equilibrium? and (b) which model, among a set of classical specifications, most accurately characterizes observed price dynamics? We observe excess demand-driven prices spiraling outwardly away from the interior equilibrium prices as predicted by the theory of disequilibrium price dynamics. We estimate a structural model establishing that partial equilibrium dynamics characterize price changes even in an unstable general equilibrium environment. We observe linkages between excess demand in one market and price changes in another market, but the sign of expected price change in a market does not depend on the magnitude of excess demand in other markets unless disequilibrium is severe.

    Original languageEnglish
    Pages (from-to)1033-1084
    Number of pages52
    JournalEconomic Theory
    Volume71
    Issue number3
    DOIs
    StatePublished - Apr 2021

    Keywords

    • Experiments
    • General equilibrium
    • Stability

    Fingerprint

    Dive into the research topics of 'Divergence and convergence in Scarf cycle environments: experiments and predictability in the dynamics of general equilibrium systems'. Together they form a unique fingerprint.

    Cite this