Bayesian estimation of exchange rate regime choice with spatial effect

Exchange rate arrangements and trade are inherently connected. Exchange rate volatility has a significant impact on trade volumes, and trading partners thus could affect each other’s exchange rate regime choice. This spatial effect among trading partners has been overlooked in empirical studies of countries’ exchange rate regime choices. This paper addresses this deficit by presenting a specifically designed ordered probit model with spatial effect. The model is estimated using a Bayesian method that utilizes data augmentation with Gibbs sampling. The results obtained here differ markedly from those presented in the existing literature.