Abstract

The study explores the dynamics of volatility spillover from U.S. political uncertainties to the emerging economies of South Asia by using the Markov-Switching and a DCC-GARCH model. By using the daily data from 2000 to 2024, the study has found that the U.S. political risk has significantly affected the majority of the South Asian stock markets during the first regime. On the other hand, it has affected the stock markets of Bangladesh and Maldives during the second regime. Moreover, the transition probabilities also show a higher likelihood of remaining in the second regime of high volatility along with longer expected durations. Then, the DCC-GARCH model confirms the dependence of these markets' returns on their lagged squared residuals. Overall, the volatility spillover from the U.S. political uncertainties exists in the short run to Indian, Bangladeshi, and Maldivian stock markets, while this spillover exists in the long run for all the stock markets.

Key Words

Volatility Spillover, Markov-Switching, DCC-GARCH, Political Uncertainties, South Asia

Introduction

Economic integration between the world's financial markets has increased due to the impact of economic globalization. This led to the expansion of international trade in the world. The world has witnessed a huge change in the international finance system, like the adoption of a floating exchange rate system and the removal of blockades for foreign exchange and capital flow within the nations (Aloui, 2007). Because of this improved trade, fast processing of information, and advancement in information technology, the transmission of returns and volatilities has amplified. The financial and portfolio managers need to comprehend the nature and degree of the connection between these financial markets (Maghyereh & Abdoh, 2021). The relationship between these financial markets is examined by finding the mean and volatility spillover across them because it is crucial for designing the optimal portfolio and smart hedging strategies. 

The crisis brings a negative trend to the stock markets. The relationship between different classes of assets is different in different financial markets of the world. This relationship has also changed during political and global health crises like COVID-19. International investors' perceptions change during a crisis (Hoffmann et al., 2013). This happens due to their decreased expectation and level of risk tolerance during the crisis, and their perception improved at the end of the crisis (Yousaf et al., 2018). Literature has shown that individual investors started showing herding behavior compared to institutional investors during the period of crisis because they panicked due to the alarming situation of the economy (Kim & Wei, 2002). Therefore, the crisis brought about a change in the behavior of investors.

When the big financial markets of the world face a financial crisis, then the effect of this crisis also affects the other markets of the world, which have strong integration with it, and vice versa. The fractal market hypothesis argues that the stock market prices show fractal properties because their sharp uncertainties in the market can lead to market crises. Secondly, the meteor shower effect explains that the volatility in the market can spill over to the other market, and it continues in that market and keeps producing volatility in geographically distant markets, which open many hours later (Susmel & Engle, 1994). The global integration of the stock markets does create opportunities for international investors to diversify their portfolios, but at the same time, it does bring challenges for them. 

Political crises are shaping the financial and economic interactions of the economies (Dogan et al., 2021). These investors are giving more importance to geopolitical risk than political instabilities. High levels of geopolitical crises reduce the economic activities of a country and its stock returns. The political risks for this study have been measured with the help of Geopolitical risk, which is the risk coming from terrorist attacks and tensions between the countries that affect their correlations. This will capture the direct and indirect risks originating from the political incidents of a country. These geopolitical crises negatively impact stock market returns (Caldara & Iacoviello, 2022). The political crisis contributes to the volatility spillover in the stock market. These crises negatively affect the stock and positively impact the oil markets (Smales, 2021). Scholars have used the Markov-Switching GARCH model to study the impact of political risk on the stock markets because of the non-linear relationship between them (Hoque & Zaidi, 2021). 

The political uncertainties in the USA affect the South Asian stock markets. This will increase disturbance in other countries of South Asia because investors are not willing to invest in a country with risky returns (Shear et al., 2020). Therefore, it becomes mandatory for international portfolio managers to adjust the allocation of their assets between the different financial markets during a crisis. It also calls for the attention of policymakers to take sound actions for the smooth functioning of their financial and commodity markets during a period of crisis. Therefore, the aim of this investigation is to find the pattern of Spillover from the U.S. political uncertainties to the stock markets of South Asia. 

The study brings novelty by using the two-regime bivariate MS-EGARCH model. In contrast, most of the studies in the literature have used the basic tests of univariate and bivariate GARCH models for finding the patterns and trends in the data. Secondly, the model is suitable for detecting the dependence on the regime of the impact, asymmetric and persistent response to the shocks because the conditional variance is dependent on the past and the present shocks along with the past states of the economies. Thirdly, the study has also used the non-parametric approach for checking the spillover from the political uncertainties of the USA by applying the Markov regime-switching model, while the existing literature has made use of the parametric models (Kilian L., 2010). The DCC-GARCH model has also been used to study volatility spillover in the short and long run. Moreover, this study checks the impact of the political uncertainties of the USA on the stock markets of South Asia. The interconnections, average effects, and volatility transmission from advanced economies to South Asian developing nations are influenced by various political upheavals. These include the consequences of terror strikes, significant political occurrences, armed conflicts, and associated contentions (Balcilar et al., 2018; Bouras et al., 2019; Elsayed & Helmi, 2021). 

Literature Review

Political crises create uncertainties in the world economies because investor shies away from the stock markets during the political crises to avoid their losses. They tend to sell their stocks and that further increases the volatility in this market which further reduces the prices. The political crisis increases the volatility spillover from the world's developed nations to the developing and emerging nations. Political instability is part of the crisis and many nations of the world have faced. The international political crisis increased the level of volatility, which impact the South Asian countries. During political crises, there are greater chances of spillover (Onyeama, 2021). Irshad (2017) has checked the impact of inflation, industrial Production, political instability, and exports on the returns and volatility of the stock markets by using the ARDL and ARCH/GARCH families and found that political instability negatively affects the country's stock market. 

Literature has shown that the political crisis of the world has resulted in a significant decline in the stock market returns and increased volatilities. Sharma & Bangur (2024) systematically reviewed the literature that studies the impact of political events on the stock markets. Bhatia (2023) has checked the impact of political crises on the stock market efficiency within the Indian Banking Sector by using the Wild Bootstrap Automatic Variance Ratio. Their findings show that internal crises of the markets have a more negative and significant impact on the market as compared to the pandemic or global financial crises. Study finds that internal crises in the industry are shaping the efficiency of the markets. 

The Ukraine Crisis has led to increased fluctuations in crude oil prices, reaching new highs. This development is expected to affect China's economic landscape (CNBC, 2022). To identify enhanced diversification opportunities during times of crisis, researchers have examined the influence of geopolitical risk on oil, gold, and stock returns throughout the Global Financial Crisis, the COVID-19 pandemic, and the Russia-Ukraine War. This analysis employed Wavelet Power Spectrum and Wavelet Coherence Transformation techniques (Shaik et al., 2023). Their findings indicate that gold provides superior diversification prospects compared to stocks and oil markets during financial turmoil. This research contributes to a deeper comprehension of how geopolitical risk impacts financial markets in turbulent periods and aids in identifying potential safe-havens. Overall, existing literature demonstrates the adverse effects of wars on stock markets (Kamal et al., 2023). Studies utilizing event study methodology have observed negative abnormal returns on the event date, with this unfavorable market reaction dissipating after the event.

Murtaza & Ali (2015) also studies the impact of political uncertainties on the stock market of Pakistan. He has taken a total of nine events and grouped them into two categories. The first category involves those events that cause changes in government policies and the other are those events that do not bring any change in government policy. By using t-statistics, the findings show that those political uncertainties that affect the government policies significantly affect the stock returns of the KSE. Scholars find the impact of political events on the KSE and find that elections of the state result in increased stock returns, while the selection of prime minister results in decreased stock returns in Pakistan (Audi et al., 2022). Other studies have also been done to find the impact of political instability on stock returns and they have found a negative relationship between them. The increase in political instabilities results in the decline of stock prices (Irshad, 2017). 

Studies have found that political uncertainties are affecting stock prices. They have further confirmed that political uncertainties reduce the prices of the stocks in the year prior to the elections and during the elections as well because the level of uncertainty increased during this period and that decreased the stock returns (Fulgence et al., 2023). Literature also shows that presidential elections also increase the volatilities in the stock markets because the high level of pre-election uncertainty increases the volatilities in the stock markets (Musah et al., 2023). 

On the other hand, exports positively impact stock prices, and inflation harms the economy. Saeed Meo (2017) has studied the influence of political instability, corruption, and effectiveness of the government on the performance of the stock market by using the VECM model for the countries of South Asia. They found a positive relationship exists between political stability and corruption control in South Asian countries. 

Ghanem & Rosvall (2014) have studied the impact of the major events in the world on the international prices of the stock markets by using the event study methodology. They have found that the stock market responds differently to the different events of the world. The spillover is significantly negative for the economic events, and they have found neutral and positive reactions to the stock market. Moreover, they have found evidence that the stock markets of Europe follow a similar pattern after the effect of international events of the world.

Globalization has increased the dependencies between the financial markets of the globe (Ali, 2022). The domestic and international markets are closely integrated. The information flows and volatility from the currency markets in the international markets affect the domestic stock markets of the country. The political and natural crisis in the international markets transmits to the highly integrated domestic markets, and these crises reduce the benefits of portfolio diversification for international investors. International investors can benefit from their portfolios only when the markets have weak integration. The crisis in the immense international markets affects other countries' financial markets. The level of integration between the international markets also changes during a crisis.

Sulehri & Ali (2020) have studied the impact of political events on the stock markets of Pakistan by using the event study methodology. They have taken a total of 18 events, which included the positive and negative events, and they have found that major events in Pakistan have significant influence on the stock markets, like the nuclear test and reinstating of Nawaz Sharif by the Supreme Court. The Ukraine Crisis has increased the volatility in the crude oil price and reached heights. This aims to hit China's economy (CNBC, 2022). Overall, they have found that the major events bring instability to the stock market returns of Pakistan. The political events bring volatility to the stock markets, which is of major concern for investors and policymakers in making investment decisions. Based on this literature, the following hypothesis has been developed:

H_1: Their volatility spillover has increased from the political uncertainties of the USA to the stock markets of South Asian countries.

Methodology

The study has also used the Markov-Switching Regression model to address the non-linear nexus of the South Asian stock markets. This technique has the benefit over the conventional techniques because it addresses the non-linear nature of the time series data. The switching regression helps find non-linearity and asymmetry in the financial data of the markets. The model is also helpful whenever adjustments are driven mainly by exogenous actions (Baghchi, 2017). 

Moreover, for estimating the spillover from the political uncertainties of the U.S. to the stock markets of South Asia, the DCC-GARCH model has been used. The model enables estimating the correlations in the volatilities that are dynamic in nature and it also helps devise the optimal diversification alternatives for the investors and portfolio managers. The study has used the daily data of South Asia's stock markets. The stock markets include Maldives, Pakistan, Sri Lanka, Afghanistan, Nepal, Bhutan, and Bangladesh stock exchanges. Afghanistan was subsequently excluded from the research due to the absence of stock data, as the country lacks a stock exchange. This study employs geopolitical risk as a substitute for political crisis, which is characterized as the risk stemming from terrorist activities, armed conflicts, and interstate political tensions (Elsayed & Helmi, 2021). Such risks disrupt the smooth and typical progression of international relations (Caldara & Iacoviello, 2022).

Results & Discussion:

Descriptive Statistics>Table 1 recapitulates the findings and shows that the average returns of NEPSE stock markets have a maximum value of 0.000728, while crude oil markets have the lowest mean return of 0.000205. The table also shows that the Maldives Stock Exchange Index (MASIX) has a maximum value of returns of 0.614748. The Royal Securities Exchange of Bhutan Limited (BSI) has having least amount of standard deviation of 0.005200. Moreover, the values of Skewness help in understanding the nature of data distribution. Moreover, the Maldives Stock Exchange Index (MASIX) has a maximum value of skewness of 6.102631, which makes it positive or right-skewed. Then, the Kurtosis value is larger than three for all the market returns. The high value of Kurtosis tends to produce extreme returns in the market that can also be positive and negative. Moreover, Jarque-Bera's values are also significant for the returns of all the series and endorse that these returns distributions are not normal.




 >Table 1>Descriptive Statistics align="center">

> > > > > > > > > > >

width="12%" nowrap=""> RNIFTY50 width="11%" nowrap=""> RKSE100 width="10%" nowrap=""> RNEPSE width="9%" nowrap=""> RDSEX width="10%" nowrap=""> RMASIX width="7%" nowrap=""> RBSI width="8%" nowrap=""> RCSE width="13%"> RGPR USA

Mean width="12%" nowrap=""> 0.000 width="11%" nowrap=""> 0.000 width="10%" nowrap=""> 0.000 width="9%" nowrap=""> 0.000 width="10%" nowrap=""> 0.000 width="7%" nowrap=""> 0.000 width="8%" nowrap=""> 0.000 width="13%"> 0.000

Median width="12%" nowrap=""> 0.000 width="11%" nowrap=""> 0.000 width="10%" nowrap=""> 0.000 width="9%" nowrap=""> 0.000 width="10%" nowrap=""> 0.000 width="7%" nowrap=""> 0.000 width="8%" nowrap=""> 0.000 width="13%"> 0.000

Maximum width="12%" nowrap=""> 0.163 width="11%" nowrap=""> 0.085 width="10%" nowrap=""> 0.0829 width="9%" nowrap=""> 0.097 width="10%" nowrap=""> 0.614 width="7%" nowrap=""> 0.061 width="8%" nowrap=""> 0.182 width="13%"> 2.282

Minimum width="12%" nowrap=""> -0 152 width="11%" nowrap=""> -0.077 width="10%" nowrap=""> -0 162 width="9%" nowrap=""> -009 width="10%" nowrap=""> -0 244 width="7%" nowrap=""> -0 021 width="8%" nowrap=""> -0138 width="13%"> -0.871

Std. Dev. width="12%" nowrap=""> 0.014 width="11%" nowrap=""> 0.013 width="10%" nowrap=""> 0.016 width="9%" nowrap=""> 0.01 width="10%" nowrap=""> 0.026 width="7%" nowrap=""> 0.005 width="8%" nowrap=""> 0.011 width="13%"> 0.055

Skewness width="12%" nowrap=""> -0.594 width="11%" nowrap=""> -0.299 width="10%" nowrap=""> -0.632 width="9%" nowrap=""> -0.203 width="10%" nowrap=""> 6.102 width="7%" nowrap=""> 3.052 width="8%" nowrap=""> 0.015 width="13%"> 11.817

Kurtosis width="12%" nowrap=""> 14.529 width="11%" nowrap=""> 7.062 width="10%" nowrap=""> 13.19 width="9%" nowrap=""> 17.525 width="10%" nowrap=""> 147.019 width="7%" nowrap=""> 34.399 width="8%" nowrap=""> 28.463 width="13%"> 538.482

Jarque-Bera width="12%" nowrap=""> 31127.76 width="11%" nowrap=""> 3979088 width="10%" nowrap=""> 6031.87 width="9%" nowrap=""> 16909 width="10%" nowrap=""> 3337257 width="7%" nowrap=""> 32913 width="8%" nowrap=""> 148779 width="13%"> 68580

Probability width="12%" nowrap=""> 0.000 width="11%" nowrap=""> 0.000 width="10%" nowrap=""> 0.000 width="9%" nowrap=""> 0.000 width="10%" nowrap=""> 0.000 width="7%" nowrap=""> 0.000 width="8%" nowrap=""> 0.000 width="13%"> 0.000

Sum width="12%" nowrap=""> 2.442 width="11%" nowrap=""> 3.418 width="10%" nowrap=""> 0.999 width="9%" nowrap=""> 4.29 width="10%" nowrap=""> 1.385 width="7%" nowrap=""> 0.297 width="8%" nowrap=""> 2.877 width="13%"> 0.819

Sum Sq. Dev. width="12%" nowrap=""> 1.184 width="11%" nowrap=""> 0.99 width="10%" nowrap=""> 0.393 width="9%" nowrap=""> 0.196 width="10%" nowrap=""> 2611 width="7%" nowrap=""> 0.02 width="8%" nowrap=""> 0.766 width="13%"> 17.754

> 




Unit Root Test>The ADF and Phillip-Perron tests have been used and results are reported in Table 2. The findings show that all the series of prices are non-stationary at their first level, and consequently, these get stationary at their first difference after differencing them. Hence, the study can discard the null hypothesis of stationary series at their level. This suggests that all of these series have a unit root, and their first difference should be used for further analysis to avoid spurious regressions.




 >Table 2>Unit Root Test align="center">

> > > > > > > > >

width="37%" colspan="2"> Augmented dicky Fuller Test width="44%" colspan="2"> Phillips-Perron Test

width="16%"> At Level width="20%"> 1st difference width="21%"> At Level width="22%"> 1st difference

KSE-100 width="16%"> -2.290 align="center">(0.438) width="20%"> -66.852 align="center">(0.000) width="21%"> -2.379 align="center">(0.390) width="22%"> -67.233 align="center">(0.000)

NIFTY-50 width="16%"> -1.367 align="center">(0.870) width="20%"> -72.848 align="center">(0.000) width="21%"> -1.428 align="center">(0.852) width="22%"> -72.848 align="center">(0.000)

NEPSE width="16%"> -1.944 align="center">(0.630) width="20%"> -17.743 align="center">(0.000) width="21%"> -1.791 align="center">(0.708) width="22%"> -34.273 align="center">(0.000)

DSEX width="16%"> -1.608 align="center">(0.478) width="20%"> -18.364 align="center">(0.000) width="21%"> -1.558 align="center">(0.503) width="22%"> -41.324 align="center">(0.000)

MASIX width="16%"> -2.118 align="center">(0.534) width="20%"> -62.551 align="center">(0.000) width="21%"> -2.064 align="center">(0.565) width="22%"> -62.562 align="center">(0.000)

BSI width="16%"> -0.183 align="center">(0.993) width="20%"> -17.341 align="center">(0.000) width="21%"> -0.867967 align="center">(0.9575) width="22%"> -28.962 align="center">(0.000)

CSE-All Share width="16%"> -3.371 align="center">(0.055) width="20%"> -11.303 align="center">(0.000) width="21%"> -2.978903 align="center">(0.1382) width="22%"> -62.647 align="center">(0.000)

GPR-USA width="16%"> -6.185307 align="center">(0.000) width="20%"> -56.647 align="center">(0.000) width="21%"> -7.406319 align="center">(0.070) width="22%"> -34.876 align="center">(0.000)

> 




Heteroscedasticity & ARCH-LM Test>For the application of ARCH and GARCH models on the data, it is required to check the heteroscedasticity and autocorrelation of the returns series. This family is used for analysis and results have been reported in Table 3. For checking these two conditions, firstly, the heteroscedasticity has been checked, and it shows significant results at a 5% level for all the returns series except the returns of The Royal Securities Exchange of Bhutan Limited (BSI), which were significant at a 10% level. Secondly, the study used the ARCH-LM test to check the ARCH effects in the returns series. The results favor the alternate hypothesis and reject the null hypothesis of no autocorrelation in the data, and the ARCH-LM Test has been used. Consequently, it confirms the ARCH effect's existence in the series. After fulfilling both conditions, the volatility spillover has been checked with the help of Markov-& DCC-GARCH approaches.




 >Table 3>Heteroscedasticity & ARCH-LM Test align="center">

> > > > > > > >

width="27%"> Heteroscedasticity width="28%"> ARCH-LM

KSE-100 width="20%"> F-Statistics align="center">P-value width="27%"> 407.931 align="center">0.000 width="28%"> 2.712 align="center">0.099

NIFTY-50 width="20%"> F-Statistics align="center">P-value width="27%"> 250.740 align="center">0.000 width="28%"> 3.498 align="center">0.003

NEPSE width="20%"> F-Statistics align="center">P-value width="27%"> 159.908 align="center">0.000 width="28%"> 2.310 align="center">0.042

DSEX width="20%"> F-Statistics align="center">P-value width="27%"> 11.058 align="center">0.000 width="28%"> 5.454 align="center">0.000

MASIX width="20%"> F-Statistics align="center">P-value width="27%"> 1.566 align="center">0.210 width="28%"> 0.674 align="center">0.642

BSI width="20%"> F-Statistics align="center">P-value width="27%"> 3.309 align="center">0.069 width="28%"> 8.141 align="center">0.000

CSE-All Share width="20%"> F-Statistics align="center">P-value width="27%"> 192.744 align="center">0.000 width="28%"> 6.297 align="center">0.000

GPR-USA width="20%"> F-Statistics align="center">P-value width="27%"> 21.01111 align="center">0.000 width="28%"> 35.66187 align="center">0.000

> 




Volatility spillover from Political Uncertainties (Global Political Risk) of USA to South Asian Stock Markets>The study now checks the volatility spillover that comes from the political uncertainties of the United States of America. The Markov Switching Model has been employed for investigating the volatility spillover from the Global Political Risk of the United States of America to the South Asian Stock markets, and its results are presented in Table 4. The results show that in the first regime, the Global Political Risk of the USA has positively and significantly affected the KSE-100, BSI, and CSE-All Share. It has negatively affected the returns of the NEPSE, the DSEX, and MASIX. At the same time, the National Stock Exchange of India (NIFY-50) remains unaffected. On the contrary, during the second regime, the Global Political Risk of China had a positive and significant impact on the Dhaka Stock Exchange (DSEX) and a negative impact on the Maldives Stock Exchange (MASIX). This shows that the political uncertainties of the USA are affecting the stock markets of Bangladesh and Maldives only.>Moreover, the standard error of regression coefficients has lower values in the second regime for all the South Asian stock markets, which means the volatility has decreased for these stock markets during the political uncertainties in the USA. The lower values of standard errors bring more accuracy for the model estimation. Overall, the values of standard errors are low for the second regime, and both values are closer to zero, which indicates the absence of random error in the model. It also confirms the data adequacy and low difference between the real and predicted values by the model.




 >Table 4>Bi-Variate Markov Switching Model align="center">

> > > > > > > > > > > > > >

Stock Markets width="12%"> Regimes width="12%"> Variable width="14%"> Coefficient width="11%"> St. Error width="14%"> Z-Statistics width="14%"> Probability

KSE-100 width="12%"> Regime 1 width="12%"> RNYSE width="14%"> 0.001 width="11%"> 0.000 width="14%"> 4.132 width="14%"> 0.000

width="12%"> Regime 2 width="12%"> RNYSE width="14%"> 0.000 width="11%"> 0.000 width="14%"> 1.339 width="14%"> 0.181

NIFTY-50 width="12%"> Regime 1 width="12%"> RNYSE width="14%"> 0.000 width="11%"> 0.000 width="14%"> -0.669 width="14%"> 0.503

width="12%"> Regime 2 width="12%"> RNYSE width="14%"> 0.000 width="11%"> 0.001 width="14%"> -0.454 width="14%"> 0.650

NEPSE width="12%"> Regime 1 width="12%"> RNYSE width="14%"> -0.001 width="11%"> 0.000 width="14%"> -3.618 width="14%"> 0.000

width="12%"> Regime 2 width="12%"> RNYSE width="14%"> -0.002 width="11%"> 0.001 width="14%"> -1.345 width="14%"> 0.179

DSEX width="12%"> Regime 1 width="12%"> RNYSE width="14%"> -0.021 width="11%"> 0.012 width="14%"> -1.742 width="14%"> 0.082

width="12%"> Regime 2 width="12%"> RNYSE width="14%"> 0.000 width="11%"> 0.000 width="14%"> -2.099 width="14%"> 0.036

MASIX width="12%"> Regime 1 width="12%"> RNYSE width="14%"> -19.852 width="11%"> 0.854 width="14%"> -23.234 width="14%"> 0.000

width="12%"> Regime 2 width="12%"> RNYSE width="14%"> -10.139 width="11%"> 0.503 width="14%"> -20.161 width="14%"> 0.000

BSI width="12%"> Regime 1 width="12%"> RNYSE width="14%"> 0.002 width="11%"> 0.001 width="14%"> 2.003 width="14%"> 0.045

width="12%"> Regime 2 width="12%"> RNYSE width="14%"> 0.000 width="11%"> 0.000 width="14%"> -0.578 width="14%"> 0.564

CSE-All Share width="12%"> Regime 1 width="12%"> RNYSE width="14%"> 0.010 width="11%"> 0.001 width="14%"> 7.490 width="14%"> 0.000

width="12%"> Regime 2 width="12%"> RNYSE width="14%"> 0.000 width="11%"> 0.000 width="14%"> 0.769 width="14%"> 0.442

> 




Transition Probabilities and Expected Durations from Markov Switching Model>The study has reported the estimates of probabilities for transition along with the forecasted duration in each Regime during Political Uncertainty in the USA in Table 5. The transition probabilities from the first regime to the second regime are higher for most South Asian Stock markets than the transition probabilities from the second to first regime. This shows that the second regime is relatively permanent, and the transition process is comparatively slow from the second regime to the first regime. However, the KSE-100, the National Stock Exchange of India Limited (NIFTY-50), and the Nepal Stock Exchange (NEPSE) have high values of transition from the second to the first regime. The majority of stocks have a larger expected duration of remaining in the second regime, while the expected duration of being in the second regime also has higher values; this confirms that the second regime is more stable than the first one. The Maldives Stock Exchange (MASIX) has maximum values of expected duration in the second regime. Additionally, the one-step ahead regime switching probabilities for all the stock markets are shown in Figure 1.




 

Table 5

Transition Probabilities and Expected Durations

> > > > > > > >

Stock Markets width="45%" colspan="4"> Transition Probabilities width="30%" colspan="2"> Expected Duration

width="11%"> width="11%"> width="11%"> width="15%"> width="15%">

KSE-100 width="11%"> 0.976 width="11%"> 0.024 width="11%"> 0.593 width="11%"> 0.407 width="15%"> 42.144 width="15%"> 1.687

NIFTY-50 width="11%"> 0.985 width="11%"> 0.015 width="11%"> 0.708 width="11%"> 0.292 width="15%"> 68.341 width="15%"> 1.412

NEPSE width="11%"> 0.961 width="11%"> 0.039 width="11%"> 0.750 width="11%"> 0.250 width="15%"> 25.904 width="15%"> 1.334

DSEX width="11%"> 0.475 width="11%"> 0.525 width="11%"> 0.004 width="11%"> 0.996 width="15%"> 1.905 width="15%"> 268.227

MASIX width="11%"> 0.996 width="11%"> 0.004 width="11%"> 0.001 width="11%"> 0.999 width="15%"> 283.781 width="15%"> 960.297

BSI width="11%"> 0.228 width="11%"> 0.772 width="11%"> 0.025 width="11%"> 0.975 width="15%"> 1.295 width="15%"> 39.629

CSE-All Share width="11%"> 0.368 width="11%"> 0.632 width="11%"> 0.008 width="11%"> 0.992 width="15%"> 1.582 width="15%"> 121.542

Figure 1

Regime Switching Probabilities for all the South Asian Stock Markets



Short and Long Run Volatility Spillover from Political Risk of USA and South Asian Stock Markets>For checking the short and long-run volatility spillover in the South Asia stock markets, the study has employed the DCC-GARCH model, and its results have been reported in Table 6. The results show that ? is significant for all the stock markets except the Royal Securities Commission of Bhutan (BSI), confirming the dependence on the lagged squared residuals. Then, the ? values are also significant for all the stock markets, showing the persistence of volatility. It also confirms that volatility can be estimated by its lagged values. Moreover, the sum of ? and ? is less than for all the series, which signifies the deterioration in the persistence of volatility with time, and it further confirms that conditional correlation does not remain constant and keeps changing with time. Overall, it confirms the diffusion of information from the Political uncertainties of the USA to the stock markets of South Asia.>The value of DCC? shows significant values for the National Stock Exchange of NIFTY-50, DSEX, and MASIX, while the remaining stock markets remain unaffected. This confirms that transmission of volatility spillover from the political uncertainties does exist in the short run for the stock markets of India, Bangladesh, and Maldives only. On the contrary, the value of DCC? shows significant values for all the stock markets at a 5% level, which means volatility spillover does exist from the political uncertainties of the USA to these markets in the long run. This means investors need to look for alternate options for diversifying their portfolios because all the South Asian markets are being affected by the political uncertainties of the USA in the long run.




 >Table 6>DCC-GARCH Approach align="center">

> > > > > > >

Stock Markets width="12%"> Estimates width="10%"> µ width="10%"> ? width="9%"> ? width="10%"> ? width="10%"> DCC ? width="10%"> DCC ? width="12%"> Elapsed Time

KSE-100 width="12%"> Coefficients align="center">Sig-Value width="10%"> 0.000    0.000 width="10%"> 0.000    0.000 width="9%"> 0.153   0.000 width="10%"> 0.807   0.000 width="10%"> 0.441    0.440 width="10%"> 0.554    0.000 width="12%"> 16.610

NIFTY-50 width="12%"> Coefficients align="center">Sig-Value width="10%"> 0.000    0.000 width="10%"> 0.000    0.046 width="9%"> 0.110    0.000 width="10%"> 0.878    0.000 width="10%"> 0.162    0.000 width="10%"> 0.763    0.000 width="12%">10.905

NEPSE width="12%"> Coefficients align="center">Sig-Value width="10%"> 0.000    0.010 width="10%"> 0.000    0.000 width="9%"> 0.216    0.000 width="10%"> 0.735    0.000 width="10%"> 0.012    0.775 width="10%"> 0.826    0.007 width="12%">4.925

DSEX width="12%"> Coefficients align="center">Sig-Value width="10%"> 0.000    0.101 width="10%"> 0.000    0.068 width="9%"> 0.247    0.000 width="10%"> 0.740    0.000 width="10%"> 0.125    0.002 width="10%"> 0.841    0.000 width="12%"> 4.776

MASIX width="12%"> Coefficients align="center">Sig-Value width="10%"> 0.000    0.626 width="10%"> 0.000    0.462 width="9%"> 0.021    0.000 width="10%"> 0.973    0.000 width="10%"> 0.745    0.000 width="10%"> 0.250    0.000 width="12%"> 15.678

BSI width="12%"> Coefficients align="center">Sig-Value width="10%"> 0.000    0.653 width="10%"> 0.000    0.822 width="9%"> 0.032    0.153 width="10%"> 0.953    0.000 width="10%"> 0.001    0.504 width="10%"> 0.586    0.000 width="12%"> 2.218

CSE-All Share width="12%"> Coefficients align="center">Sig-Value width="10%"> 0.000    0.138 width="10%"> 0.000    0.339 width="9%"> 0.280    0.000 width="10%"> 0.718    0.000 width="10%"> 0.058    0.139 width="10%">0.8240.000 width="12%"> 10.083

> 




DCC GARCH Forecasts>The forecasts for the Conditional Correlation Matrices are shown in Table 7. The forecasted conditional correlation between all the series is almost close to zero, which suggests that no significant relationship exists among the forecasted returns of both series. The National Stock Exchange of India (NIFTY-50), the Maldives Stock Exchange (MASIX), and the Royal Securities Exchange of Bhutan (BSI) have negative correlation forecasts with the Global Political Risk of the USA. The conditional correlation remains constant over 5 periods, suggesting that the model expects a constant and stable relationship between both series in the near future.




 >Table 7>DCC GARCH Forecasts align="center">

> > > > > > >

Stock Markets width="15%"> 1 width="15%"> 2 width="15%"> 3 width="15%"> 4 width="15%"> 5

KSE-100 width="15%"> 0.107 width="15%"> 0.107 width="15%"> 0.106 width="15%"> 0.106 width="15%"> 0.106

NIFTY-50 width="15%"> -0.191 width="15%"> -0.177 width="15%"> -0.163 width="15%"> -0.151 width="15%"> -0.140

NEPSE width="15%"> 0.006 width="15%"> 0.006 width="15%"> 0.006 width="15%"> 0.005 width="15%"> 0.005

DSEX width="15%"> 0.067 width="15%"> 0.065 width="15%"> 0.065 width="15%"> 0.062 width="15%"> 0.061

MASIX width="15%"> -0.003 width="15%"> -0.003 width="15%"> -0.003 width="15%"> -0.003 width="15%"> -0.003

BSI width="15%"> -0.039 width="15%"> -0.039 width="15%"> -0.039 width="15%"> -0.039 width="15%"> -0.039

CSE-All Share width="15%"> 0.012 width="15%"> 0.008 width="15%"> 0.004 width="15%"> 0.001 width="15%"> -0.002

> 




Correlation between US & South Asian Political Risk>The association between the Global Political has been checked, and its results are shown in Table 8 and graphically in Figure 2. Overall, the findings show a positive and weak relationship between the majority of South Asians and the political uncertainties of the USA. This shows that these series show some dependence on each other. On the contrary, the Royal Securities Exchange of Bhutan (BSI) has a negative Correlation with the political uncertainties of the USA. Figure 4 also shows that patterns of their correlation are not constant. Instead, it keeps changing with time and returns to its mean position that exists in the long term.




 

Table 8

Correlation among US & South Asian Political Risk

> > > > > > >

Stock Markets width="50%"> Correlation

KSE-100 width="50%"> 0.009

NIFTY-50 width="50%"> 0.100

NEPSE width="50%"> 0.015

DSEX width="50%"> 0.136

MASIX width="50%"> 0.671

BSI width="50%"> -0.030

CSE width="50%"> 0.019

Figure 3

Correlation among US & South Asian Political Risk 



Covariance among US & South Asian Political Risk>The covariance among the Political uncertainties is shown in Table 9 and graphically in Figure 4. The results show that covariance between both the series is low and propose that fluctuations in the political conditions of the USA cannot predict the volatility in the South Asian' stock markets.




 

 

Table 9

Covariance among US & South Asian’s Political Risk

> > > > > > >

Stock Markets width="27%"> Variance of NYSE Returns width="36%"> Variance of Stock Market Returns width="17%"> Covariance

KSE-100 width="27%"> 1.97E-04 width="36%"> 1.77E-04 width="17%"> 1.68E-06

NIFTY-50 width="27%"> 3.07E-03 width="36%"> 2.67E-04 width="17%"> 9.02E-05

NEPSE width="27%"> 3.27E-03 width="36%"> 1.57E-04 width="17%"> 1.05E-05

DSEX width="27%"> 3.10E-03 width="36%"> 6.05E-05 width="17%"> 5.88E-05

MASIX width="27%"> 1.46E-03 width="36%"> 5.10E-04 width="17%"> 5.78E-04

BSI width="27%"> 1.15E-02 width="36%"> -1.47E-05 width="17%"> 2.14E-05

CSE-All Share width="27%"> 3.20E-03 width="36%"> 1.11E-04 width="17%"> 1.11E-05

Figure 4

Covariance among US & South Asian’s Political Risk



Results Discussion

The Markov Switching Model has shown that in the first regime, the Global Political Risk of the USA affected the majority of the stock markets in South Asia. This shows that the political uncertainties of the USA are affecting the stock markets of Bangladesh and Maldives only during periods of high volatility. Moreover, the standard error of regression coefficients has lower values in the second regime for all the South Asian stock markets, which means the volatility has decreased for these stock markets during the political uncertainties in the USA. The transition probabilities from the first to the second regime are higher for most South Asian Stock markets. This shows that the second regime is relatively permanent, and the transition process is comparatively slow from the second to the first regime. Then expected duration is large for second regime, while the expected duration of being in the second regime also has higher values; this confirms that the second regime is more stable than the first one. The Maldives Stock Exchange (MASIX) has maximum values of expected duration in the second regime, which shows that volatility does not spill over to this market from the political uncertainties in the USA.

The DCC-GARCH model confirms the dependence on the lagged squared residuals for all the stock markets and the persistence of volatility. It also confirms that volatility can be estimated by its lagged values. Moreover, the sum of ? and ? is less than for all the series, which signifies the deterioration in the persistence of volatility with time. Overall, it confirms the diffusion of information. It also confirms that transmission of volatility spillover from the political uncertainties does exist in the short run for the stock markets of India, Bangladesh, and Maldives only. In contrast, long-term volatility spillover is evident across all South Asian stock markets. This implies that investors should seek alternative diversification strategies for their portfolios, as the political uncertainties in the United States have a long-term impact on all South Asian markets. The research indicates a weak but positive correlation between most South Asian markets and US political uncertainties. On the contrary, the Royal Securities Exchange of Bhutan (BSI) has a negative Correlation with the political uncertainties of the USA. Results show that patterns of their correlation are not constant. Instead, it keeps changing with time and returns to its mean position that exists in the long term.

Conclusion

The study aimed to find the volatility spillover from the U.S. political uncertainties to the stock markets of South Asia. The findings have found a significant spillover from the USA geopolitical risk to the South Asian stock markets. The results of the Markov-Switching model have shown a significant spillover during the period of high volatility, especially for the stock markets of Maldives and Bangladesh. Moreover, the transitions from the second regime are slow due to reduced volatility. Then, DCC-GARCH's findings have shown the short-run spillover in India, Bangladesh, and Maldives, while spillover does exist in the long run for all the markets. Overall, there exists a positive but limited association between South Asian stock markets and the geopolitical risk of the United States. Therefore, it is crucial for investors to look for alternative options for investments during the time of heightened political uncertainties in the USA. 

References

• Ali, A. (2022). Determining Pakistan's Financial Dependency: The Role of Financial Globalization and Corruption. Journal of Business and Economic Options, 5(1), 24-38. https://mpra.ub.uni-muenchen.de/116097/

• Aloui, C. (2007). Price and volatility spillovers between exchange rates and stock indexes for the pre- and post-euro period. Quantitative Finance, 7(6), 669–685. https://doi.org/10.1080/14697680701302653
• Audi, M., Sulehri, F. A., Ali, A., & Al-Masri, R. (2022). An event based analysis of stock return and political uncertainty in Pakistan: revisited. International Journal of Economics and Financial Issues, 12(5), 39–56. https://doi.org/10.32479/ijefi.13239
• Balcilar, M., Bonato, M., Demirer, R., & Gupta, R. (2018). Geopolitical risks and stock market dynamics of the BRICS. Economic Systems, 42(2), 295–306. https://doi.org/10.1016/j.ecosys.2017.05.008
• Bhatia, S. (2023). Bank capital channel of monetary policy: panel data evidence for India. Indian Economic Review, 58(S2), 423–443. https://doi.org/10.1007/s41775-023-00173-0
• Bouras, C., Christou, C., Gupta, R., & Suleman, T. (2018). Geopolitical Risks, Returns, and Volatility in Emerging Stock Markets: Evidence from a Panel GARCH Model. Emerging Markets Finance and Trade, 55(8), 1841–1856. https://doi.org/10.1080/1540496x.2018.1507906
• Caldara, D., & Iacoviello, M. (2022). Measuring geopolitical risk. American Economic Review, 112(4), 1194–1225. https://doi.org/10.1257/aer.20191823
• Dogan, E., Majeed, M. T., & Luni, T. (2021). Analyzing the impacts of geopolitical risk and economic uncertainty on natural resource rents. Resources Policy, 72, 102056. https://doi.org/10.1016/j.resourpol.2021.102056
• Elsayed, A. H., & Helmi, M. H. (2021). Volatility transmission and spillover dynamics across financial markets: the role of geopolitical risk. Annals of Operations Research, 305(1–2), 1–22. https://doi.org/10.1007/s10479-021-04081-5
• Fulgence, S., Kwabi, F., Boateng, A., Hu, W., & Paudyal, K. (2023). Cross-country analysis of the effects of political uncertainty on stock price informativeness. Journal of International Financial Markets Institutions and Money, 88, 101829. https://doi.org/10.1016/j.intfin.2023.101829
• Ghanem, D., & Rosvall, D. (2014). Major world events impact on stock market prices: An event study.
• Hoffmann, A. O., Post, T., & Pennings, J. M. (2012). Individual investor perceptions and behavior during the financial crisis. Journal of Banking & Finance, 37(1), 60–74. https://doi.org/10.1016/j.jbankfin.2012.08.007
• Hoque, M. E., Zaidi, M. a. S., & Hassan, M. K. (2021). Geopolitical uncertainties and Malaysian stock market returns: Do market conditions matter? Mathematics, 9(19), 2393. https://doi.org/10.3390/math9192393
• Irshad, H. (2017). Relationship among political instability, stock market returns and stock market volatility. Studies in Business and Economics, 12(2), 70–99. https://doi.org/10.1515/sbe-2017-0023
• Kamal, M. R., Ahmed, S., & Hasan, M. M. (2023). The impact of the Russia-Ukraine crisis on the stock market: Evidence from Australia. Pacific-Basin Finance Journal, 79, 102036. https://doi.org/10.1016/j.pacfin.2023.102036
• Kilian, L. (2010). Oil price volatility. WTO Working Papers. https://doi.org/10.30875/9b9f6bc8-en
• Kim, W., & Wei, S.-J. (2002). Foreign portfolio investors before and during a crisis. Journal of International Economics, 56(1), 77-96.
• Maghyereh, A., & Abdoh, H. (2021). Time–frequency quantile dependence between Bitcoin and global equity markets. The North American Journal of Economics and Finance, 56, 101355. https://doi.org/10.1016/j.najef.2020.101355
• Murtaza, H., & Ali, R. (2015). Impact of major political events on stock market returns of Pakistan. Public Policy and Administration Research, 5(4), 68-84.
• Musah, G., Domeher, D., & Musah, A. (2023). Presidential elections and stock return volatility: evidence from selected sub-Saharan African stock markets. Journal of Financial Economic Policy, 15(3), 248–262. https://doi.org/10.1108/jfep-02-2023-0033
• Onyeama, A. E. (2021). Positive Spillover of Political and Technological Crises on Persons, Organizations, and Destinations across Covid-19: A Data Triangulation Study. Open Journal of Political Science, 11(03), 488–521. https://doi.org/10.4236/ojps.2021.113033
• Saeed Meo, M. (2017). Impact of political stability, government effectiveness and corruption on stock markets of South Asia. Journal of the Punjab University Historical Society, 30(1). 
• Shaik, M., Jamil, S. A., Hawaldar, I. T., Sahabuddin, M., Rabbani, M. R., & Atif, M. (2023). Impact of geo-political risk on stocks, oil, and gold returns during GFC, COVID-19, and Russian – Ukraine War. Cogent Economics & Finance, 11(1). https://doi.org/10.1080/23322039.2023.2190213
• Sharma, S., & Bangur, P. (2023). A literature review on political uncertainty and stock market behaviour. Asian Journal of Comparative Politics, 9(2), 291–307. https://doi.org/10.1177/20578911231170436
• Shear, F., Ashraf, B. N., & Sadaqat, M. (2020). Are Investors’ Attention and Uncertainty Aversion the Risk Factors for Stock Markets? International Evidence from the COVID-19 Crisis. Risks, 9(1), 2. https://doi.org/10.3390/risks9010002
• Smales, L. (2021). Geopolitical risk and volatility spillovers in oil and stock markets. The Quarterly Review of Economics and Finance, 80, 358–366. https://doi.org/10.1016/j.qref.2021.03.008
• Sulehri, F. A., & Ali, A. (2020). Impact of political uncertainty on Pakistan stock exchange: An event study approach. Journal of Advanced Studies in Finance, 11(2), 194-207. http://dx.doi.org/10.14505//jasf.v11.2(22).10
• Susmel, R., & Engle, R. F. (1994). Hourly volatility spillovers between international equity markets. Journal of International Money and Finance, 13(1), 3-25.
• Yousaf, I., Ali, S., & Shah, S. Z. A. (2018). Herding behavior in Ramadan and financial crises: the case of the Pakistani stock market. Financial Innovation, 4(1). https://doi.org/10.1186/s40854-018-0098-9