EXCHANGE RATE MODEL AND INTEREST RATE EVIDENCE FROM THAILAND

Abstract

This research is to explore the relationship between the interest rate difference of Thailand and the United States and exchange rate in Baht and US dollar. The finding shows that such comovement is not consistent with uncovered interest rate parity. Additionally, the exchange rate model demonstrates that money supply of home country at time t , foreign price level at time t, foreign interest rate in the next period, and real output of home country at time t determine the current exchange rate. Moreover, the result indicates that the uncovered interest parity condition is violated. That is, it cannot capture the exchange rate movement. The multiple linear regression model reveals that the Thai broad money supply has a positive impact on exchange rate. On the contrary, the US consumer price index is negatively related to its exchange rate. In particular, such model clearly outperforms the other models. In addition, there are two cointegrating relationships. VECM explains that if there is a deviation from the long-run relationship, the exchange rate will suddenly move toward the equilibrium again with speed of adjustment of 2.219%. Such model is not appropriate for exchange rate forecasting. In contrast, ARIMA (1, 1, 1) model is more effective than the other one. Consequently, the Bank of Thailand should not focus solely on the interest rate difference between Thailand and the United States. Instead, they should not only concentrate on broad money supply of Thailand, but also pay attention to the movement of consumer price index in the United States. This is because both of them significantly affect on the Thai exchange rate.