Path dependence in economic processes: implications for policy analysis in dynamical system contexts

Abstract

Introduction: credo and context I believe that the future of economics as an intellectually exciting discipline lies in its becoming an historical social science. Much of my work as an economic historian has sought to convey a strong sense of how ‘history matters’ in economic affairs by undertaking applied studies, focused on the behaviour of stochastic processes at either the micro- or the macroeconomic level, in which the proximate and eventual outcomes could be said to be path-dependent . By the term ‘path-dependent’ I mean that the process is non-ergodic : a dynamical system possessing this property cannot shake off the effects of past events, and, consequently, its asymptotic distribution (describing the limiting outcomes towards which it tends) evolves as a function of its own history. Although path dependence will be found where a dynamic resource allocation process is not described by a first-order Markov chain, it can be encountered also in some stochastic systems where the transition probabilities are strictly state-dependent (first-order Markovian) – namely, in those systems where a multiplicity of absorbing states exists. Under such conditions, small events of a random character – especially those occurring early on the path – are likely to figure significantly in ‘selecting’ one or other among the set of stable equilibria, or ‘attractors’, for the system. Although the nature of the specific selecting events themselves may elude prediction, it does not follow that the particular equilibrium configuration – among the multiplicity that are available ex ante – remain unpredictable.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 100 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.