An Analysis of The Impact of Political News on Thai Stock Market.
[Dissertation (University of Nottingham only)]
Political risk is one of the crucial factors influencing the operation of a country’s financial market as widely claimed by numerous studies. Many examples of the events which are considered as a political risk are the switch in regimen, a revolution, a coup incident, or a civil war. In Thailand, one of the most recent incidents contribute to the instability of the country’s political situation which received much of public attention worldwide is the 2006-military coup. Not only the coup itself, the demonstrations by all groups of people both for and against the government also play an important role in causing political risk in the country and affect investors’ decision to invest in Thai stock market. The most common way I think for this risk to transmit its effect on the country’s stock market is through the announcement of political news. That is, investors reevaluate their investment plans after the recognition of political situation in the country from those news.
Regarding to this assumption, it is thus interesting to investigate the impact of political news on Thai stock market. Though there are abundant of existing academic literatures in this area, Thailand provides a unique research context due to the complication in its political system and its position as a developing country. The main model I used here for the analysis is the Generalized Autoregressive Conditional Heteroscedasticity in mean or GARCH-in-mean model. The distinctiveness of this model is that it takes account of risk as measured by the time-varying variance (or volatility term) in the estimation of stock returns. In particular, this implies the relationship between risk and return of the assets. Based on the concept of the GARCH-M model, three model specifications were identified in order to examine the impact of favorable and unfavorable political news on stock returns and volatility which are GARCH (1,1)-M, EGARCH, and APARCH model where the APARCH is the most general form of the other two. Note that the EGARCH and APARCH model specifications are selected to be used because their ability in capturing the asymmetric effect of the impact between favorable and unfavorable news on stock return. For the stock data, I employed a historical daily data of the SET index returns. This is a composite index which composes the price of all common stock traded in Thai stock market. The study period of this dissertation starts from January 1, 2004 to October 1, 2008. Moreover, this is separated into two sub-periods which pre- and post-coup period.
Mainly, there are four hypotheses to be tested in this study including 1) there is a difference in stock market volatility between the event day and non-event day 2) the difference in stock market volatility between the event and non-event day is higher after the coup 3) major political news has a significant impact on both stock market returns and volatility and 4) the impact of major political news on stock market returns and volatility was amplified after the coup incident. For the first two hypotheses, the results from a nonparametric Mann-Whitney test show that stock volatilities in the event day are significantly difference from those in the non-event days in both pre- and post-coup period. Besides, it is apparently proved that SET index returns appear to be negative on the day in which unfavorable news was released and the opposite occur on the day in which favorable news was released.
For the third and fourth hypotheses, the results from estimating the GARCH (1,1)-in-mean, the EGARCH, and the APRACH model specifications indicate a significant impact of favorable(unfavorable) news on the increase(decrease) in SET index returns. Similarly, both types of news also have a significant effect on SET index volatility in a way that it increases the volatility of the stock market. An incredibly surprising finding is that the effect of positive and negative political news on stock returns is symmetric which is rare in the real world. However, we failed to find an evidence of the latest coup on stock return and its amplification effect as well as the risk premium, though a significant impact of coup on stock volatility is evidenced. Lastly, the leverage effect of political news on stock market has been found also the persistence of shocks from the news on stock volatility. In conclusion, these results imply the importance of political instability on the health of Thai stock market.
Actions (Archive Staff Only)