The impact of changes in government policies on economic performance (the ARCH model)

Abstract

The impact of changes in government policy from direct control to free market on spatial market integration is investigated with a bivariate exponential autoregressive conditional heteroskedasticity (ARCH) model. Causal relationships and the magnitude of impact of new information on price discovery between geographic markets during different policy regimes are determined by the Granger causality test and dynamic multipliers, respectively. Our findings show that intervention policies impair the efficiency of arbitrage in integrating spatial markets. Price adjustments between regional markets are slower during periods of government regulation than during periods of decontrol. Highly volatile prices and a longer ARCH process also tend to characterize the former more than the latter. These findings for the Philippine copra markets support the importance of taking heteroskedasticity into account in testing spatial integration.