تاثیر تصمیمات سیاست پولی بر بازده سهام: شواهد از تایلند

Although numerous studies have examined the effect of monetary policy on stock prices, empirical research in the international setting remains relatively scant. Therefore, this topic is reexamined in the context of Thailand. In a sample of 50 repurchase rate announcements of the Bank of Thailand during 2003–2009, our regression results suggest that the raw change in the repurchase rate has a negative effect on stock returns at the market level, which is inconsistent with the literature. Contrary to the results of numerous studies, we find that at the market level the expected change in the repurchase rate has a negative effect on stock returns and the unexpected change in the repurchase rate exhibits no effect on stock returns. However, the effect of the unexpected change in the repurchase rate on stock returns is evident at the firm level. Our findings also suggest that the stock market's response to the repurchase rate change is asymmetric. The unexpected change in the repurchase rate that is considered as good news has a negative effect on stock returns. Overall, the evidence lends support to the notion that the monetary policy announcements have a significant effect on stock prices and further adds to the debate on whether the creditability of the monetary authority may contribute to the stock market's response to the monetary policy actions.

As the levels of foreign investments in developing countries have been increasing in recent years, understanding how monetary policy in developing countries affects the economy is crucial to making investment decisions in these markets. Since the nature of financial markets and economic frameworks in developing countries differs from that of developed countries, our knowledge of how monetary policy works in developed economies may not be accurately applied in the context of developing economies. Consider, for instance, in many developing countries, there exists no organized futures markets or there are a limited number of futures contracts being available in the futures markets; therefore, domestic investors are likely to use different approaches to the estimation of assets’ future prices. Furthermore, publicly available information available to investors in developing countries tends to be in relatively short supply. Given these contextual differences, a key question is then whether we would observe a similar pattern of stock price reactions to a change in monetary policy in small and developing countries as that of developed countries. There is a large body of research on the implications of monetary policy for asset prices in the context of developed countries (see e.g., Bredin et al., 2007, Farka, 2009, Konrad, 2009, Lobo, 2000, Rigobon and Sack, 2004 and Thorbecke, 1997), suggesting that monetary policy actions are associated with changes in stock prices. Research on the effect of monetary policy on equity prices at both the aggregate and disaggregate levels has shown that the unexpected changes in monetary policy affect equity prices (see e.g., Bernanke and Kuttner, 2005, Chuliá et al., 2010, Lobo, 2000, Reinhart and Simin, 1997 and Thorbecke and Alami, 1994). Several studies have suggested that equity prices react negatively to a hike in interest rates (e.g., Bernanke and Kuttner, 2005, Bredin et al., 2007, Chuliá et al., 2010, Farka, 2009 and Rigobon and Sack, 2004). For instance, Farka (2009) reports a negative relationship between an unexpected hike in the Federal funds target rate and stock returns in the US. Several studies (e.g., Basistha and Kurov, 2008, Bernanke and Kuttner, 2005, Chuliá et al., 2010 and Farka, 2009) also show that the effect of monetary policy on equity prices is asymmetric. For example, Farka (2009) finds that an unexpected tightening monetary policy tends to have a smaller effect on equity prices than does an unexpected easing monetary policy. In the same vein, Chuliá et al. (2010) find that stock prices generally react more strongly to positive surprises in the Federal funds target rate in the US than to negative surprises in the Federal funds target rate. In the context of developing countries, the evidence of the impact of monetary policy on equity prices is limited to a few empirical studies (see e.g., Goodhart et al., 2003). Therefore, we attempt to make a contribution to the literature by examining the effect of monetary policy on stock prices in a small and open economy. In particular, we investigate the effect of monetary policy on equity prices in Thailand, which is one of emerging market countries in Asia. As a result of the demand for the further development of the capital market in Thailand in the early 1970s, the Stock Exchange of Thailand (SET) was established in 1975. There were 21 listed companies on the SET by the end of 1975. As the Thai economy had expanded rapidly during 1980s–1990s, the SET had been growing steadily during this time period. By the end of 1996 (prior to the 1997-financial crisis in Asia), there were 454 listed firms on the SET with the market capitalization of 2,559,578 million Baht (approximately 98,445 million USD at the exchange rate of 26 Baht/USD at the time). At the end of 2003 the SET had 407 listed firms with the total market capitalization of 4,789,857 million Baht (approximately 121,262 million USD at the exchange rate of 39.5 Baht/USD), and at the end of 2009 the SET had 501 listed firms with the total market capitalization of 5,873,100 Million Baht (approximately 176,901 million USD at the exchange rate of 33.2 Baht/USD). As the Thai economy has continued to grow in recent years, the capital market in Thailand has also further developed. By the end of 2010 the market capitalization of the SET was 8,247,635 million Baht (approximately 274,921 million USD at the exchange rate of 30 Baht/USD). For a comparison purpose, Malaysia, which is also one of the emerging market countries in Asia, had the market capitalization of 1,275,280 million RM (approximately 411,380 million USD at the exchange rate of 3.1 RM/USD) at the end of 2010, which was about 50% larger than the SET in terms of total market capitalization. Given that financial markets in Thailand are still relatively small but rapidly developed, the results reported in this paper can provide us a better understanding of the relationship between monetary policy and equity prices in emerging market countries. More specifically, we test whether in the context of Thailand there is evidence of an inversed relationship between repurchase rate changes and stock prices. Furthermore, we test whether the effect of repurchase rate changes on stock prices is asymmetric. To test the effect of monetary policy on equity prices, we primarily adopt two approaches used by Bernanke and Kuttner (2005) and He (2006). We measure the stock market's response to the Bank of Thailand's repurchase rate announcement, used as a proxy for a monetary policy decision, during 2003–2009, at both aggregate (market) and disaggregate (firm) levels. We make several discoveries in this study. First, our regression results suggest that a raw increase in the repurchase rate has a negative effect on stock returns at the market level and the firm level. Second, at the market level the expected change in the repurchase rates has a negative effect on stock returns. This finding is in contrast with prior studies indicating that the expected change in the Federal funds rate target has no effect on stock returns (e.g., Bernanke and Kuttner, 2005). Third, the unexpected change in the repurchase rate has no effect on stock returns at the market level. However, at the firm level the expected change in the repurchase rate has a negative effect on stock returns while the effect of the unexpected change on stock returns is evident but inconclusive. Fourth, we also find evidence from our regression results indicating that the market's reaction to the repurchase rate changes is asymmetric. The unexpected change in the repurchase rate has a negative (positive) effect on stock returns when it is considered as good (bad) news. Fifth, using the time-series analysis framework of He (2006), we find that at the market level the change in the repurchase rate has no effect on stock returns. The results also reveal that the unexpected change in the repurchase rate cannot explain the stock return variation. In addition, the time-series results do not lend support to the notion that the stock market reacts asymmetrically to the direction of the repurchase rate change. Overall, the results of the time-series regressions, which suggest that the monetary policy change exerts no effect on stock returns, are in contrast with our regression analyses that indicate the negative relation between the change in the repurchase rate and stock returns at both the market and firm levels. The paper is organized as follows. Section 2 describes data and methodology used in this study. Section 3 reports empirical results. Section 4 concludes the paper.

The existing literature on the effect of monetary policy on equity prices in the context of the developed countries suggests that stock prices respond only to the unexpected component of a change in monetary policy and that stock markets, on average, react asymmetrically to monetary policy actions (e.g., Bredin et al., 2007, Farka, 2009, Konrad, 2009, Lobo, 2000, Rigobon and Sack, 2004 and Thorbecke, 1997). In this paper we extend the monetary policy research to an international setting by examining the effect of monetary policy decisions on stock prices in Thailand, at the market and firm levels. The Bank of Thailand database is our primary source of information on the 1- and 14-day repurchase rates, which are used as a proxy for the change in the monetary policy, over the period January 2003 to December 2009. We use the Bloomberg database and the SETSMART database as the key sources of the sample firms’ stock prices, financial statements, and the SET Index over the January 1998 to December 2009 period. We investigate whether there is a negative relationship between monetary policy and stock prices at the market and firm levels. We also examine whether equity prices react asymmetrically to the direction of monetary policy actions as well as to good/bad news. We use two methods to examine the effect of monetary policy announcement on stock prices. First, we adopt the regression-based event study, which is used to avoid the endogeneity and omitted variables problems (see e.g., Bernanke and Kuttner, 2005, Lobo, 2000 and Reinhart and Simin, 1997). Second, we use the time-series analysis approach (see e.g., Bernanke and Kuttner, 2005 and He, 2006) to test the robustness of our results. Due to the fact that futures repurchase rates do not exist in Thailand, we are unable to follow the approach used by Bernanke and Kuttner (2005) to measure the unexpected component of the repurchase rate. Therefore, we infer the market's expectations by using the survey data on the repurchase rate expectations from the Bloomberg database, which is consistent with the approach used by Reinhart and Simin (1997). Using the approaches of Roberts (1997), Jonsson and Österholm (2011) to test the biasness of the survey data, we find that the Bloomberg's survey of repurchase rate appears to be an unbiased predictor of the Bank of Thailand's repurchase rate. The regression-based event study results show that stock prices respond negatively to the expected change in the repurchase rate. Surprisingly, however, the unexpected change appears to have no effect on stock returns at the aggregate level, which is inconsistent with the results reported by numerous studies reporting that the surprise component has a negative effect on stock prices (see e.g., Bernanke and Kuttner, 2005, Chuliá et al., 2010 and Smirlock and Yawitz, 1985). Interestingly, we find that at the firm level the expected change in the repurchase rates has a negative effect on stock returns while the unexpected change has a positive effect on stock returns. Furthermore, the effect of the unexpected repurchase rate change on stock returns is statistically insignificant when the direction interactions are present. The evidence suggests that stock prices react asymmetrically to the direction of the repurchase rate movements. The interaction between the expected increase direction and the expected repurchase rate change has the detrimental effect on stock prices. Likewise, the interaction involving the expected decrease direction with the expected change in the repurchase rate has a positive effect on stock prices. Surprisingly, the interaction between the unexpected increase direction and the unexpected repurchase rate change has a positive effect on stock returns. More strikingly, the unexpected repurchase rate change has a negative effect on stock returns when it is considered good news but has a positive effect on stock returns when it is considered bad news. The relationship between the surprise component of a monetary policy and stock returns is inconclusive; the evidence lends some support to the notion that the market's reaction to the unexpected change in the monetary policy is asymmetric. We adopt the monthly time-series methods of Bernanke and Kuttner (2005) and He (2006) to test the robustness of the results. We find some contradictory results. It should be noted that due to the limitation of data availability (i.e., there was no futures contract on the Bank of Thailand's repurchase rate), our definition of the month-t repurchase rate surprise cannot be defined analogously as that of Bernanke and Kuttner (2005). To measure the surprise component of the repurchase rate change, we use the residuals from the autoregressive model with lag up to 12 months, similar to the model of He (2006). The time-series results indicate that neither the repurchase rate change nor the unexpected component of the repurchase rate change has a significant effect on stock prices. Furthermore, there is no evidence to support the asymmetric movement of stock prices to the policy directions and to good news and bad news. Having documented the insignificant effect of the repurchase rate change on stock prices with the time-series methodology, we then adopt the VAR methodology (see e.g., Craine and Martin, 2008, Rigobon and Sack, 2003, Rigobon and Sack, 2004 and Thorbecke, 1997) to examine the stock market response to monetary policy shocks. The VAR estimation results also indicate that there is no significant effect of monetary policy shocks on stock returns. Skeptics about the finding of the significant effect of the expected change in the repurchase rate on stock returns may argue that the use of the surveys data to imply the market's expectations on the repurchase rate may be impropriate in the context of developing countries. As discussed earlier (see the robustness test section for more detail), we find that the Bloomberg's survey of repurchase rate is an unbiased predictor of the Bank of Thailand's repurchase rate, which could thus be used as a proxy for the market's expectations. There are three plausible explanations to the significant effect of the expected component of the repurchase rate change on stock returns. First, it might be possible that the average investors in Thailand, as well as in other developing countries, are less informed (of the monetary policy expectations) and thus react to the monetary policy announcement in such a way that the monetary policy decisions appear to be simply unexpected, implying that the stock market in Thailand is inefficient in a sense that there exists a substantial discrepancy in the set of information held by a diverse group of investors. This reasoning may partially explain the significant effect of the expected component of monetary policy on stock returns. Second, the investors’ expectation of the monetary authority's actions does not necessarily infer that they believe that such actions are appropriate. It is plausible that a discrepancy between what the monetary authority should do (from the investors’ perspective) and what the monetary authority actually does may be relatively wider in developing countries than in developed countries, which could, to some extent, explain the negative response to the expected change in the repurchase rate. Of course, given the data limitation, our study cannot address this issue. We hope that future research can move beyond what we have been able to do here (i.e., in the case of Thailand). Third, is it possible that a lack of credibility of the monetary authority (see e.g., Blinder, 2000, Faust and Svensson, 2001 and Moscarini, 2007) in developing countries plays a key role here? This argument leads us to ask what would be the outcome for the credibility of monetary policy actions if we incorporate heterogeneous expectations into consideration. To the extent that a large proportion of the investors questions the monetary authority's decisions, and further, if this fraction grows over time as the credibility of the monetary authority deteriorates, regardless of whether such decisions are justified, may have a profound effect on how stock prices react to the monetary policy announcement. In this study we cannot address this possibility. Future research examining the effect of monetary policy actions on stock prices in other emerging market countries may be able to shed some light on this puzzle. There are some important implications of our findings for monetary policy makers as well as investors in Thailand, which could, to some extent, be generalized to other emerging market countries. First, the negative effect of the expected component of monetary policy on stock returns could be viewed as a warning to policy makers as to better understand how (and what) average investors make of the monetary policy announcements and find a way to further develop the country's capital markets. In more developed financial markets, the expected component of monetary policy is likely to have no effect on stock returns (see e.g., Bernanke and Kuttner, 2005 and Farka, 2009), thereby indicating that current stock prices incorporate the expectations of future monetary policy. In the case of Thailand, the findings suggest that it is reasonable to draw a conclusion that current stock prices do not fully incorporate the expectation of future monetary policy actions and therefore react to the expected component of the “actual” monetary policy. Second, the negative stock market reaction to the monetary policy announcement that is considered to be good news could indicate the market's disappointments at the actions undertaken by the Bank of Thailand, which could be viewed as the questioning of the monetary policy's actions and credibility. As Blinder (2000) and Moscarini (2007) suggest that a central bank's competence implies its credibility, which is crucial to conducting monetary policy, the investors seem to ask why the Bank of Thailand did what it did, suggesting that monetary authorities in emerging markets are being gradually held more accountable of their actions as the financial market development in these countries further progresses. Last but not least, because a large portion of investors in Thailand appears to not incorporate the expected component of monetary policy into current stock prices, there exist trading strategies that could benefit “better informed” investors. Our findings offer new insights to both domestic and foreign investors that in Thailand (probably as in other emerging market countries) the expected component of the monetary policy does still matter. Overall, this paper contributes to the monetary policy literature in several ways. First, as noted by Konrad (2009) that the empirical research on the effect of monetary policy decisions on stock returns in the international context is still limited, we attempt to fill the gap in the literature by analyzing the impact of monetary policy actions on stock returns using data from Thailand, allowing us to better understand the stock market's reaction to the monetary policy decisions in one of emerging market countries in Asia. Second, with respect to this line of research in the context of Thailand, our sample of the repurchase rate announcements is perhaps among the largest examined to date. Finally, and most importantly, we are able to show that contrary to the literature examining the stock market's response to the monetary policy announcements in the developed countries (e.g., Bernanke and Kuttner, 2005, Farka, 2009, Rigobon and Sack, 2004 and Smirlock and Yawitz, 1985), both the expected and unexpected changes in the monetary policy in a developing country with less developed financial markets such as Thailand appear to have a significant effect on stock returns.