سیاست های پولی و تغییر نرخ ارز لحظه ای اول و دوم در طول بحران مالی جهانی: شواهد از تایلند

Using a sample of monetary policy announcements in Thailand over the period 2003–2011, I show that a monetary policy surprise tends to affect the return and volatility of the Thai baht. In the full sample, a 1% unexpected increase in the policy rate leads to an about 1.8% depreciation of the baht against the Japanese yen. During periods of high interest rate differentials, an unexpected increase in the policy rate leads to a substantial depreciation of the baht against the US dollar (about 1%) and the British pound (about 2.6%). While Thai monetary policy surprises have no effect on the baht against the dollar in the spot market, they have a significant effect on the baht against the dollar in the forwards market. During the non-financial crisis period, an unexpected increase in the policy rate on average results in a large depreciation of the baht/dollar forward rates: 6.6% and 13.7% for two-month and three-month forward rates, respectively.

Do monetary policy decisions affect the first- and second-moment exchange rate change differently during financial crises in the context of emerging market countries? While this question is of importance to both investors and policy makers as monetary policy can be implemented together with foreign exchange interventions to manipulate exchange rate movements, it has rarely been examined together in detail. In one of the earlier studies on the relation between monetary policy and exchange rate volatility, Jansen and De Haan (2005) show that the European Central Bank's comments have a substantial effect on the volatility of the euro-dollar exchange rate and exert only a relatively small effect on the mean of the euro-dollar exchange rate. As this paper is perhaps one of the first studies in this line of research in the context of developing countries in Asia, I choose Thailand, a medium-sized developing country, as a sample setting for at least five reasons. First, Coudert and Mignon (2013) show that the Thai baht is one of the most rewarding currencies for carry-trade strategies, and as such, a better understanding of how Bank of Thailand's actions affect exchange rates is beneficial to both investors and policy makers. Second, the Bank of Thailand has been conducting its monetary policy under an inflation-targeting, relatively high capital mobility, and a de facto managed-floating exchange rate regime since 2000, and so the findings of this study will shed light on the impact of monetary policy actions on the return and volatility of exchange rates in Thailand during tranquil and 2007–2009 global financial crisis periods. Third, to some extent, the findings may be generalizable to other developing countries in Asia (such as India and Indonesia) that have recently begun to conduct their monetary policy under similar frameworks.1 Fourth, in terms of size and economic significance, Thailand is not only a gateway to markets in Southeast Asian countries but also is a major production base of small and large multinational corporations (e.g., all major Japanese automobile manufacturers have had their assembly/production facilities in Thailand). Last but not least, the initiative for ASEAN integration (e.g., the establishment of an ASEAN Economic Community (AEC) by 2015) that gradually results in a greater level of economic integration among the ASEAN members provides both investment opportunities and challenges to the private sector. In this paper I therefore examine whether the Bank of Thailand's monetary policy (i.e. repurchase rate) announcement has an impact on the return and volatility of the Thai baht (THB) against four major currencies: the US dollar (USD), the euro (EUR), the Japanese yen (JPY), and the British pound (GBP). Using a sample of the Bank of Thailand's 66 repurchase rate (hereafter, policy rate) announcements during 2003–2011, I show that, to some extent, it does. Over the full sample period, an unexpected change in the policy rate has a negative effect on the THB/JPY exchange rate return after controlling for term spread, interest rate differential, foreign exchange reserves, stock market attractiveness, and past exchange rate volatility. But the response of the THB/JPY exchange rate return to an unexpected change in the policy rate is asymmetric with respect to the direction of a policy rate change. An unexpected tightening monetary policy action has a positive effect on the THB/JPY exchange rate. That is, the Thai baht depreciates against the yen as a response to a tightening monetary policy surprise: depending on specifications, an unexpected 1% hike in the Thai policy rate was associated with about an almost 1.81% depreciation in the Thai baht against the yen. A monetary policy surprise has no effect on the return of the Thai baht against the euro, the US dollar, and the British pound over the full sample period. My results indicate that the THB/JPY exchange rate return reacts asymmetrically to monetary policy surprises during the 2007–2009 global financial crisis period in relation to the non-financial crisis period. Furthermore, the interaction between an interest rate differential and an unexpected change in the policy rate has a negative effect on the return of the Thai baht against the US dollar and the British pound. In a high interest rate subsample, an unexpected hike in the policy rate resulted in a large appreciation of the Thai baht against the US dollar (about 1.08%) and the British pound (about 2.63%). Prior studies, such as that of Fatum and Scholnick (2008) and of Rosa (2011), do not examine the asymmetric effect of Federal funds target rate changes on exchange rates. While the THB/USD exchange rate in the spot market does not seem to react to a Thai monetary policy action in the full sample, I find that an unexpected change in the policy rate has a negative effect on the THB/USD exchange rate in the forwards market. That is, an unexpected increase in the policy rate has a negative effect on the return of the THB/USD forwards at three horizons: one-month, two-month, and three-month. My results further show evidence of an asymmetric response of the return on the THB/USD forwards to the direction of a monetary policy change during the non-financial crisis period. For instance, during the tranquil period, a hypothetical unexpected 1% cut in the Thai policy rate leads to an approximate 6.67% depreciation of the two-month THB/USD forward rate and an almost 13.7% depreciation of the three-month THB/USD forward rate. In a standard rational expectations model, a credible central bank action (e.g., foreign exchange interventions) should either reduce exchange rate volatility or have no effect on exchange rate volatility. Frenkel et al. (2005) suggest that when actions are not credible, such actions are likely to increase exchange rate volatility. In the context of monetary policy actions in emerging market countries, the effect of monetary policy actions on exchange rate volatility is conditional on, among other things, whether central banks’ monetary policy is credible and/or effective. Jumps in exchange rate volatility following a monetary policy announcement may signal that a monetary policy action is not credible. I find that an expected change in the policy rate has a negative effect on the volatility of the Thai baht against the euro and the British pound, and an unexpected change in the policy rate has a positive effect on the volatility of the Thai baht against the euro and the British pound. There is evidence of an asymmetric effect of a monetary policy surprise on exchange rate volatility. My results show that the interaction between an unexpected change in repurchase rate and a monetary tightening dummy has a positive effect on the volatility of the Thai baht against the euro, the Japanese yen, and the British pound. The pattern of intraday exchange rate volatility results is largely similar. For instance, the interaction between an unexpected change in the policy rate and a monetary policy easing action has a positive effect on the intraday volatility of the Thai baht against the euro and the British pound. Studies such as Bonser-Neal et al. (1998), Fatum and Scholnick (2008) and Rosa (2011) examine the impact of monetary policy actions on exchange rate return by using the standard event study. To be consistent with prior studies, I adopt the event study approach to test to the relationship between monetary policy and exchange rate performance in Thailand. As Fatum and Scholnick (2008) and Rosa (2011) do not examine the effect of monetary policy actions on exchange rate volatility, I therefore test the response of exchange rate volatility to monetary policy actions by adopting an approach from prior studies, as in Fatum (2008), that uses the event study analysis to examine the effect of the central bank's foreign exchange interventions on exchange rate return and volatility. For comparison purposes, I use a realized volatility measure to proxy for the volatility of the Thai baht because previous studies on exchange rate volatility, as in Cai et al. (2008), use the standard deviation of the daily nominal exchange rates to measure the unconditional volatility of the exchange rate. More specifically, I use the 10-day standard deviation of daily returns on the Thai baht against each of the four major currencies. To test the robustness of my results, I use the 20-day standard deviation of daily returns and two range-based volatility measures: the range-based volatility measure proposed by Parkinson (1980), and the range-based volatility measure suggested by Garman and Klass (1980). In addition, I use the intraday exchange rate data to test the robustness of my results.2 After having established the above main findings, I explore a list of potential explanations for these seemingly conflicting results. My interpretation of these findings is that sufficiently large interest rate differentials induce carry-trade activities in the spot market of the Thai baht,3 which in turn cause the spot exchange rate of the Thai baht to be more sensitive to a policy rate change, given that a depreciation of the Thai baht following expansionary policy actions reduces the carry-trade strategies’ profits. In low interest rate differential environments, carry-trade activities are limited or non-existent, reducing the effect of a monetary policy on the spot exchange rate. In high interest rate differential environments, speculators are more likely to take speculative positions in the currency, and thus, become active in the spot market. Accordingly, carry-trade strategies lead to large capital inflows into high-yield currencies, amplifying the effect of a monetary policy action on the spot exchange rate. In this sense, the effect of a monetary policy surprise on exchange rate performance is conditional on the size of interest rate differentials. This explanation is built upon the findings of Menkhoff et al. (2012) that high-yield currencies are negatively related to exchange rate volatility shocks in the sense that the performance of carry-trade strategies is poor during times of market turmoil (i.e. unexpectedly high volatility in exchange rate markets). As carry-trade activities are primarily conducted in the spot market, the finding of the negative effect of a monetary policy surprise on the return of the Thai baht against the US dollar and the British pound in the high interest rate differential subsample is consistent with this explanation. In summary, I present several new results that complement and extend the existing literature in several dimensions. As with Bjørnland (2009), who reports that monetary policy shocks affect exchange rates in Australia, Canada, New Zealand, and Sweden, I extend the carry-trade literature by showing that the level of interest rate differentials can explain the relationship between monetary policy and exchange rate performance. When interest rate differentials are small, the inverse relationship between interest rates and exchange rate performance appears to have weakened. In other words, when interest rate differentials are small and carry-trade activities are very limited, unexpected policy rate cuts are no longer seen as bad news by participants in currency markets. My findings also contribute to the exchange rate exposure literature. That is, as an unexpected policy rate change affects exchange rate volatility, it may have an indirect effect on stock prices through changes in exchange rate volatility (via a firm-level currency exposure).4

I close with some remarks on my findings. Returning to the main question that I asked at the beginning of the paper about whether the return and volatility of the exchange rate respond differently to the monetary policy announcements during the 2007–2009 global financial crisis period, I find that to a large extent, in the context of Thailand, they do. In comparison with the non-financial crisis period, the pattern of the Thai baht reaction to a monetary policy surprise during the financial crisis period is different. My results also suggest that the impact of a monetary policy surprise on the Thai baht is conditional on the level of interest rate differentials. While the THB/USD exchange rate in the spot market does not seem to react to a Thai monetary policy action over the entire period sample, I find that an unexpected change in the policy rate has a negative effect on the THB/USD exchange rate in the forwards market. That is, an unexpected increase in the policy rate has a negative effect on the return of the THB/USD forwards at three horizons: one-month, two-month, and three-month. The findings reveal an asymmetric response of the return on the THB/USD forwards to direction of a monetary policy change during the non-financial crisis period. Overall, my work contributes to the literature by examining the role of a monetary policy action on the return and volatility of exchange rates during the 2007–2009 global financial crisis in the context of Thailand. By focusing on not only exchange rate return but also exchange rate volatility, my work complements previous studies, such as that of Jansen and De Haan (2005), that show that European Central Bank's statements affect the volatility of the euro–dollar exchange rate.