An experimental test of trade hysteresis: market exit and entry decisions in the presence of sunk costs and exchange rate uncertainty

Authors
Citation
D. Ansic et G. Pugh, An experimental test of trade hysteresis: market exit and entry decisions in the presence of sunk costs and exchange rate uncertainty, APPL ECON, 31(4), 1999, pp. 427-436
Citations number
17
Categorie Soggetti
Economics
Journal title
APPLIED ECONOMICS
ISSN journal
00036846 → ACNP
Volume
31
Issue
4
Year of publication
1999
Pages
427 - 436
Database
ISI
SICI code
0003-6846(199904)31:4<427:AETOTH>2.0.ZU;2-5
Abstract
In the 1980s, the unresponsiveness of trade flows to exchange rate swings i nspired models in which sunk costs in combination with exchange rate instab ility generate trade hysteresis, meaning that temporary exchange rate misal ignments have a persistent effect on trade. This paper furnishes an empiric al complement to the theoretical literature. First, it describes a computer ized experiment in which 100 subjects generated over 1000 decisions on mark et entry and exit under conditions congruent with a model of trade hysteres is developed by Paul Krugman. Secondly, these data are used to test the mai n predictions arising from the model. Our experiment bears out the main qua litative predictions of Krugman's model; in particular, that firms' trading policy is unresponsive to exchange rate movements over a wide range of val ues. Moreover, in the repeated-decision setting of the experiment, we find evidence that the stochastic behaviour of subjects' entry- and exit-price d ecisions tend towards consensus as they gain experience, even though they d o not interact with one another during the experiment. This effect, which i s not predicted in Krugman's model, supports the supposition that behaviour at the firm level in the presence of sunk costs and exchange rate uncertai nty is a plausible microeconomic foundation for otherwise puzzling macroeco nomic phenomena.