Beyond the Silicon Valley of the East: Exploring Portfolio Diversification with India and MINT Economies

Abstract

In the past few decades, India’s tech industry has boomed, making it a leader in the digital world. Today, India has many big tech companies, well-trained software developers, and cutting-edge technology like AI and cloud computing. This success shows India’s innovative spirit and makes the country a good example for other developing nations. However, global portfolio managers often overlook potential diversification opportunities beyond India’s dynamic stock market. This study investigates the viability of MINT (Mexico, Indonesia, Nigeria, and Turkey) as diversification targets, specifically analyzing spillover effects and volatility dynamics between their stock markets and that of India. Leveraging vector autoregressions (VARs) and dynamic conditional correlation (DCC)–GARCH models, we uncover intricate relationships. Further, DCC–GARCH analysis reveals varying degrees of volatility spillover, offering valuable insights for risk management. Our findings suggest that MINT economies, particularly Mexico and Turkey, hold promise for Indian portfolio diversification. By strategically incorporating these markets, investors can potentially mitigate India-specific risks and enhance portfolio returns. We urge global portfolio managers to consider Turkey as a viable diversification avenue, acknowledging the nuanced market growth dynamics highlighted in this study.

1. Introduction

Globalization has increased significantly after 1980s. This can be attributed to the liberalization of capital markets worldwide, which has resulted in greater integration of financial markets. This integration is evident through the correlation of financial markets (Singh et al. 2015). While financial market integration offers countries wider investment opportunities, especially in the case of insufficient saving, it also exposes the markets to shocks from other financial markets. Owing to market integration, the volatility of financial markets spreads to the other markets and induces portfolio shifts (Zhong and Liu 2021; Bala and Takimoto 2017).

In this respect, financial integration has the potential to turn a local financial crisis in one country into a regional or even global crisis. The Global Financial Crisis (GFC) is a case in point, as it had its origins in mortgage markets but quickly spread to other markets, as a result of market integration. In addition, the spillover effect between integrated markets is amplified by shocks of a financial or real nature. Several studies (Bekiros 2014; Lien et al. 2018; Hung 2019; Vo and Tran 2020) suggest that the Global Financial Crisis (GFC) led to a significant increase in volatility spillovers, especially from advanced to emerging markets. In the context of shock transmission, it is not imperative for the origin of the shock to exclusively emanate from developed economies. Indeed, shocks may also originate within emerging markets, with the resultant volatility being transmitted to more advanced financial markets. For instance, Li and Giles (2015) find evidence of contagion from Asian markets to advanced markets during the Asian financial crisis. Likewise, the COVID-19 pandemic, which initially caused a shock in Chinese markets, resulted in significant volatility in US and European markets (Zehri 2021; Vuong et al. 2022).

These volatility spillovers may have an impact on portfolio diversification opportunities, which are a fundamental aspect of finance (as discussed in Markowitz 1952). If there were a high correlation between equities, the portfolio would be unlikely to provide diversification, as the equities would offer equal risk and return (Siddiqui and Kaur 2023).1 Notably, on the other hand, low correlation of stock market return and volatility provides an opportunity to diversify the portfolio and minimize the risk accordingly. As suggested by Hwang (2014), it may also be beneficial to consider international diversification. For this reason, modelling stock market volatility in international financial markets is essential for effective portfolio diversification (Spulbar et al. 2020; Trivedi et al. 2021; Natarajan et al. 2014).

Developed countries are known for their high level of financial integration, which is reflected in the correlation of their stock markets. As a result, diversification opportunities in developed stock markets are limited. Several studies, including those by Bekaert et al. (2014); Gérard et al. (2003); Pukthuanthong and Roll (2009), have confirmed the high level of co-movement between developed financial markets. However, emerging markets could offer a portfolio diversification opportunity as they are less correlated with developed markets (Li and Giles 2015). According to Bala and Takimoto (2017), it appears that there is a lower level of stock market integration in emerging markets when compared to developed markets.

It is worth considering that as emerging markets adopt liberal policies, their integration with other markets deepens. Therefore, some emerging markets could not provide options for diversification. It is crucial to estimate the volatility spillover of emerging markets as it is varied through time and markets (Mensah and Alagidede 2017). For instance, India, as an emerging market, has integrated with the US markets (Kumar and Mukhopadyay 2002; Batareddy et al. 2012; Nandy and Chattopadhyay 2019). Bekiros (2014) argued that India’s integration with developed markets has increased following the GFC. As a result, it could be argued that Indian stock markets may no longer be considered a viable diversification alternative.

In this research, our attention is directed towards a selection of emerging markets that have been less examined within the academic discourse: India and the MINT countries. The MINT countries—Mexico, Indonesia, Nigeria, and Turkiye—emerged in the early 2010s as economies with significant growth and investment potential, paralleling the earlier identified BRICS nations (Brazil, Russia, India, China, and South Africa).

Given India’s extensive integration with developed economies, it becomes imperative within the realm of developing nations to investigate the volatility spillovers from India to these emerging markets. The analysis could position these emerging markets as viable alternatives to the Indian stock market, providing a novel perspective on diversification strategies and risk management within the context of global financial interactions.

Our paper investigates the potential for diversification in emerging markets, particularly focusing on the MINT economies (Mexico, Indonesia, Nigeria, and Turkey) as targets for Indian investors. The novelty of the study lies in its focus on this specific group and the time period analyzed. While prior research has explored volatility spillovers between developed and emerging markets, or between individual emerging markets, this study looks at the MINT economies after the COVID-19 pandemic. This period is significant because it saw a surge in tech-focused markets, creating new dynamics to consider. The paper examines the viability of MINT countries as diversification options by analyzing volatility spillovers and DCC-GARCH models to understand the intricate relationships between their stock markets and those of India. The target audience encompasses both financial professionals, such as portfolio managers and investment analysts, seeking portfolio diversification strategies in emerging markets, and researchers in the field of financial markets, volatility spillovers, and portfolio management in emerging economies. The paper’s detailed literature review, methodological approach, and focus on a post-pandemic landscape make it a valuable resource for both groups.

2. Literature Review

The literature of stock market volatility spillovers is both extensive and nuanced, encompassing a wealth of academic inquiry into the mechanisms of market interrelations and their implications for volatility transmission. The advent of this research stream saw the utilization of first-generation models, notably those employing cointegration and VAR methodologies, as foundational tools in the exploration of stock market integration. This approach was instrumental in pioneering studies conducted by Eun and Shim (1989); Mathur and Subrahmanyam (1990); Panda et al. (2019), which aimed to discern the interconnectedness of global stock markets and identify potential indicators of volatility spillovers. However, the static nature of these models highlighted their temporal rigidity, prompting a shift towards more dynamic analytical frameworks (Schwert 1989).

Subsequent studies have increasingly favored ARCH and GARCH methodologies for their enhanced capacity to model the temporal dynamics of market volatility; see Miyakoshi (2003), Cheng and Glascock (2005), Moon and Yu (2010), Kundu and Sarkar (2016), Jebran et al. (2017), and the references therein. Despite the advancements afforded by GARCH models, their analytical scope is limited by a lack of consideration for feedback effects among variables, an aspect crucial to fully comprehending the complexity of market behaviors.

In response to these limitations, the academic discourse has witnessed the ascension of multivariate GARCH (MGARCH) models, heralded for their sophisticated ability to simultaneously capture the volatility of current innovations and assess the enduring impact of lagged volatility. As articulated by Li and Wei (2018), MGARCH models represent a significant methodological advancement, offering a more comprehensive lens through which to examine the intricacies of volatility spillovers. This evolution of modeling techniques underscores the dynamic and evolving nature of financial market research, reflecting a continual quest for analytical precision and depth in understanding the underpinnings of global market volatility.

In addition to these methodologies, Treshold ARCH (TARCH) models incorporate the asymmetry of these volatility spillovers. The addition of TARCH models allows for researchers to explore the asymmetric nature of these transmissions, often revealing that negative shocks tend to have a larger impact on volatility compared to positive ones (Zakoian 1994).

Earlier literature searches for the volatility spillover effect between the developed countries (Hamao et al. 1990; Fratzscher 2002; Bessler and Yang 2003; Tastan 2005; Savva 2009; Xiao and Dhesi 2010; Bekaert et al. 2014; Gérard et al. 2003). Nevertheless, from the perspective of portfolio diversification, emerging markets present a viable alternative. Consequently, in pursuit of identifying those countries characterized by lower volatility spillover and thus suitable for diversification purposes, numerous studies have explored the phenomenon of volatility spillovers from developed to emerging markets. Bubák et al. (2011) searched for the spillover effect from developed European markets to emerging ones. They found a significant integration. Tilfani et al. (2020) also looked for the spillover effect in Central and Eastern countries, and they found that Bosnia, Montenegro, Serbia, and Slovakia are less integrated compared with the other European markets. Živkov et al. (2018) investigated the volatility spillover effect from Germany to Eastern European emerging stock markets and found a high correlation among the markets. Dedi et al. (2016) detected the volatility spillover between Germany, UK, China, and Russia. They found that Germany, China, and Russia are highly integrated, while for UK and Turkiye there is no volatility spillover.2

Asian developing countries also offer good opportunities for diversification. Mensi et al. (2017) used wavelet-based VAR to analyze the integration of Southeast Asian markets, BRIC, and three major stock markets. Southeast Asian and BRIC markets are not highly integrated according to the study. Li and Giles (2015) used the MGARCH approach and an asymmetric BEKK model. They found that the spillover from the US market to the Asian markets is strongly significant. Batareddy et al. (2012) searched for the integration of Asian emerging markets and developed markets (US and Japan). They concluded that India is highly integrated with US and Japan while Taiwan, South Korea, and China are not integrated as much. Zhang and Liu (2018) investigated the integration between China and Southeast Asian stock markets by using dynamic conditional correlation. They found a positive correlation and suggested that this correlation peaked during the Asian financial crisis.3

The body of literature pertaining to the MINT countries—Mexico, Indonesia, Nigeria, and Turkey—is notably limited. While there exists a modicum of research focusing on the individual nations within this grouping, studies that consider the MINT countries as a collective entity remain remarkably scarce. Siddiqui and Kaur (2023) use cointegration and rolling trace statistics in order to search for the integration between MINT countries and developed markets. They suggest that except Nigeria, all MINT countries are highly integrated with US. But in the case of Japan, there is no integration. Yadav and Pandey (2019) used Granger causality and Dynamic Conditional Correlation in order to detect spillover effect from India to MINT countries’ stock markets. They could not find a spillover from India to MINT countries. They argue that Turkiye and Mexico are less integrated with India.

3. Methodology

In our analysis, we combine various methods to obtain more coherent results. Firstly, we employed impulse response analysis and variance decompositions. Impulse response analysis and variance decomposition are two key tools employed within the context of Vector Autoregressions (VARs) to understand the dynamic interactions between variables in a system. They offer insights into how the system responds to unexpected shocks and how the variability of a specific variable can be attributed to these shocks. An impulse response function (IRF) traces the path of a variable in a VAR model following a one-unit shock to another variable, all else held constant. In simpler terms, it allows for us to assess the impact of an unexpected change in one variable on the future values of all variables in the system. IRFs are typically presented graphically, depicting the response over a chosen number of periods.

3.1. VAR and Variance Decomposition

As per the existing literature, the VAR method has been demonstrated as a coherent and credible approach (Stock and Watson 2001). The VAR model can be presented as follows:

$$y_t={\beta }_1{y}_{t-1}+{\beta }_2{y}_{t-2}{+\beta }_3{y}_{t-3}+\cdots +{\beta }_n{y}_{t-n}+{\epsilon }_t$$

(1)

This equation models how an asset’s return Y_t depends on its past performance Y_t−1, Y_t−2, Y_t−3. Coefficients β₁, β₂, β₃ capture the influence of these past returns. In Vector Autoregression (VAR) models, each variable is predicted by its own history, similar to Y_t here. VAR models are best suited for stable data (stationary series) and require choosing the most impactful lags Y_t−1, Y_t−2, etc. Once the VAR model is estimated, variance decomposition can be used to analyze the sources of variability in the dependent variable. This technique helps quantify how much of the future forecast errors (how much the actual values deviate from the predictions) for each variable are explained by its lagged values compared to the influence of other variables in the model. Moreover, impulse response analysis and variance decomposition are often used in conjunction. IRFs reveal the dynamic nature of the response, while variance decomposition helps quantify the relative importance of each shock in explaining the variability. This combined approach provides a comprehensive understanding of how shocks propagate through the system and influence the behavior of individual variables.

Afterwards, to analyze how India’s connection with the MINT economies (Mexico, Indonesia, Nigeria, and Turkey) changes over time, the Dynamic Conditional Correlation (DCC) model is employed. This advanced tool extends beyond static models by capturing the fluctuations in the strength of the relationship between these markets. It helps understand how they influence each other in the short term and how their risk patterns are linked in the long term. To model these changing risk patterns, GARCH techniques are utilized. Unlike traditional models that treat fluctuating volatility (known as heteroskedasticity) as a problem, GARCH recognizes it as an inherent feature of financial data. Markets naturally experience periods of higher and lower risk, meaning the potential size of errors in price movements can vary. GARCH models excel at capturing this volatility, providing a risk assessment tool like standard deviation. This information is crucial for tasks in finance like portfolio construction, pricing derivatives (financial contracts), and risk management.

3.2. Dynamic Conditional Correlation

In this context, the ARCH model assumes that the variance of t u_t in period t σ_t² depends on the square of the error term in the t − 1 period, u_t−1

In this context, the ARCH(q) and GARCH(q) models are as follows:

$$\begin{array}{l}{\alpha }_0 > 0, {\alpha }_i > 0\\ h_t={\alpha }_0+{\alpha }_1{{\epsilon }_{t-1}}^2+{\alpha }_2{{\epsilon }_{t-2}}^2+...+{\alpha }_q{{\epsilon }_{t-q}}^2+{\nu }_t\end{array}$$

(2)

The GARCH models which express the generalized form of ARCH models were developed by Engle (1982) and Bollerslev (1986) to provide reliable estimations and predictions. GARCH models consist of conditional variance, as shown in Equation (3), in addition to conditional mean in Equation (3).

$$h_t={\alpha }_0+{\displaystyle \sum _{i=1}^q{\alpha }_i}{r^2}_{t-i}+{\displaystyle \sum _{i=!}^p{\beta }_j{h}_{t-j}}$$

(3)

In this context, restrictions of the variance model are as follows:

for α_i ≥ 0 and β_i ≥ 0, α_i + β_i < 1:

If α_i + β_i ≥ 1, it is termed as non-stationary invariance.

Financial data often exhibit non-stationary variance, meaning their volatility changes over time. In such cases, traditional forecasts do not accurately predict future volatility as time progresses (as noted by Brooks 2008). This is where ARCH and GARCH models come in handy. They allow for us to estimate the volatility of an asset at a specific point in time, addressing this challenge. It is important to remember that risk and return are intertwined in asset pricing models. The expected return on an investment should reflect the potential risk involved, which is often measured by the variance of the return (as explained by Enders 2004).

Coefficient α_i refers to the ARCH process in the residuals from asset i which depicts the fluctuations of the assets reflecting the impact of external shocks on fluctuations. The ARCH effects measure short-term persistence while the GARCH effect measures long-term persistence which is represented by β_i.

The Dynamic Conditional Correlation (DCC-) GARCH belongs to the class of models of conditional variances and correlations. It was introduced by Engle and Sheppard (2001). The idea of the models in this class is that the covariance matrix, H_t, can be decomposed into conditional standard deviations, D_t, and a correlation matrix, R_t. In the DCC-GARCH model, both D_t and R_t are designed to be time-varying.

Suppose we have returns, at, from n assets with expected value 0 and covariance matrix H_t. Then, the Dynamic Conditional Correlation (DCC-) GARCH model is defined as (4)

$$\begin{array}{c}{r}_t={\mu }_t+{\alpha }_t\\ {\alpha }_t=H_t^{1/2}{z}_t\\ H_t=D_t{R}_t{D}_t\end{array}$$

(4)

r_t: n × 1 vector of log returns of n assets at time t,

α_t: E[α_t] = 0 and Cov[α_t] = H_t n × 1 vector of mean-corrected returns of n assets at time t, i.e.,

µ_t: n × 1 vector of the expected value of the conditional r_t,

H_t: n × n matrix of conditional variances of α_t at time t,

H_t^1/2: any n × n matrix at time t such that H_t is the conditional variance matrix of at. H_t^1/2 may be obtained by a Cholesky factorization of H_t,

D_t: n × n diagonal matrix of conditional standard deviations of α_t at time t,

R_t: n × n conditional correlation matrix of α_t at time t,

Z_t: n × 1 vector of iid errors such that E[Z_t] = 0 and E[Z^T_t].

In addition, Q₀, the starting value of Q_t, has to be positive definite to guarantee H_t to be positive definite. The correlation structure can be extended to the general DCC (M, N)-GARCH model (5):

$$\begin{array}{c}{R}_t={\varrho }_t^{\ast 1}{\varrho }_t{\varrho }_t^{\ast 1}\\ {\varrho }_t=\left(1-{\varrho }_1-{\varrho }_2\right)\overline{\varrho }+{\varrho }_1{\epsilon }_{t-1}{\epsilon }_{t-1}^T+{\varrho }_2{\varrho }_{t-1}\end{array}$$

(5)

In this context, ${\varrho }_t$ can be estimated as mentioned below:

$${\varrho }_t=\frac{1}{T}{\sum }_{t=1}^T{\epsilon }_t{\epsilon }_t^T$$

(6)

Some conditions are imposed on parameters ${\varrho }_1$ and ${\varrho }_2$ to guarantee H_t to be positive definite. In addition to the conditions for the univariate GARCH model to ensure positive unconditional variances, scalars a and b must satisfied

$${\varrho }_1 \ge 0, {\varrho }_2 \ge 0 \mathrm{ve} {\varrho }_1 +{\varrho }_2 < 1$$

4. Data and Empirical Results

For investors looking beyond established markets, the BRICS (Brazil, Russia, India, China, South Africa) nations were once the hot destinations. Today, the focus has shifted to the MINT economies—Mexico, Indonesia, Nigeria, and Turkey.

Therefore, the data set includes the S&P BSE Sensex 504 Index as representative of the Indian stock market, whereas Mexican Bolsa, JKSE, NGSEINDEX, and BIST are the indices of Mexico, Indonesia, Nigeria, and Turkey, respectively. The time period of the data covers between 16.12.2016 and 03.04.2024, and the return of each market is calculated as follows:

$$ln\left(P_t\right)-ln\left(P_{t-1}\right),$$

(7)

where RBSESensex, RShenzhenCSI, RNasdaq, and RDXY refer to the return series of related variables. Figure 1 shows the time series of the daily returns of the markets. The time-varying and volatility clustering characteristics can be observed in Figure 15.

Sensex, launched in 1986, is India’s oldest stock market index. It tracks the performance of 30 large financially stable companies from key industries listed on the Bombay Stock Exchange (BSE). Managed by Standard and Poor’s (S&P), Sensex selects companies based on their freely available shares for trading (free-float market capitalization). Companies with the highest market value (market cap) have a bigger influence on the index’s overall movement. The financial sector, including banks and non-banking financial companies (NBFCs), has the biggest weight in Sensex, followed by the information technology (IT) sector. Mexican Bolsa is mostly driven by banks, retailers, FMCG companies, and telecommunication while banking, mining, energy, and infrastructure drives Indonesia (JSKE). On the other hand, Nigeria stock exchange is driven by energy, food, infrastructure, and insurance. Finally, the driving forces behind Borsa Istanbul over the past five years have likely shifted somewhat. Traditionally, the financial sector, with large banks as key players, held significant influence. However, recent years might show a growing impact from other sectors. Areas like technology, with its potential for high growth, and industrials, which are crucial to the Turkish economy, could be playing an increasingly important role. It is worth noting that external factors like global commodity prices and geopolitical events can also significantly affect Borsa Istanbul’s performance.

Emerging markets like Turkey and Nigeria are enticing for investors seeking high growth due to their young populations. This growth, however, comes with some inherent risk. The Nigerian stock market, NGX, is on a tear. It is the best performing stock market in the world so far this year, with the main index crossing the 100,000 marks for the first time. Several factors are driving the rally, including investor fear of missing out, strategic news on bellwether shares, and sustained interest from new domestic investors. The dominance of a few large companies, like Dangote Cement and BUA Cement, is a key feature of the market. These companies have seen huge gains this year, and Dangote Cement became the first Nigerian company to reach a market valuation of NGN 10 trillion. Institutional investors, both domestic and foreign, are starting to take notice. The expectation of a banking sector recapitalization is driving the buying pressure on Nigerian lenders. And the recent efforts of the Central Bank of Nigeria to clear the foreign exchange backlog and further tighten monetary policy are seen as positive signs for the market. However, some challenges remain. Inflation and fuel price rigidity are persisting issues, and the participation of foreign investors remains low due to FX policy inconsistency. How the government addresses these issues will be key to determining the market’s future course.

On the other hand, Indonesia’s stock market performance can be more susceptible to global commodity price swings, as their economy relies heavily on exports like palm oil and copper. Similarly, Mexico’s close ties to the US economy mean a slowdown there can ripple through to their stock market. India presents a different picture. Its massive and growing domestic population creates a strong internal demand for domestically produced goods and services, potentially leading to robust performance for Indian companies listed on the stock exchange.

We begin by presenting the impulse response functions (IRFs) that capture the dynamic response of the S&P BSE Sensex 50 Index to one-standard-deviation shocks applied to the Mexican Bolsa, Jakarta Stock Exchange Composite Index (JKSE), Nigerian Stock Exchange All-Share Index (NGSEINDEX), and Borsa Istanbul (Figure 2).

The IRFs for the Borsa Istanbul and Bolsa Mexicana de Valores exhibit a statistically significant yet transient negative impact on the linear specification of Indian stock returns. Conversely, the responses of the BSE Sensex 50 Index to shocks from the JKSE and NGSEINDEX are relatively muted, suggesting a weaker competitive influence from these markets on the Indian stock market. Based on these findings, we posit that Turkey and Mexico represent the primary competitors of the Indian stock market (Figure 3).

Empirical evidence from variance decomposition analysis (Figure 4) supports this observation. While internal factors exert the most significant influence on the volatility of the Indian stock exchange, the contribution of external shocks appears to have undergone a potential shift. Contrary to the findings of Yadav and Pandey (2019), external shocks emanating from Turkey and Mexico exhibit a greater impact compared to those originating from Nigeria and Indonesia. This divergence necessitates further investigation, particularly regarding the potential for a structural change in causality triggered by the COVID-19 pandemic.

The COVID-19 pandemic triggered a sharp decline in both the Indian and Turkish stock markets in early 2020. However, their subsequent recoveries diverged significantly. The Indian market, represented by the BSE Sensex, experienced a steeper initial plunge but exhibited a faster and more sustained rebound. This resilience can be attributed to India’s robust economic fundamentals and a younger population, factors that facilitated a quicker adaptation to the new normal characterized by remote work and digital dependence. In contrast, the BIST 100 index in Turkey displayed a more volatile recovery path. This stemmed from pre-existing vulnerabilities like high inflation and weakening currency, which were further amplified by the pandemic’s economic disruptions.

Investors might be worried about inflation hurting the stock market and how well stocks protect against it. To understand this better, let us look at how a particular stock index performed during past periods of inflation. Figure 5 shows monthly returns of the S&P BSE Sensex 50 compared to inflation changes since 2017. Interestingly, even when inflation rose, the index still went up in over half the cases (59%). But during times of very high inflation (above 3%), that success rate dropped to 53%.

Similarly, for the BIST 100 index, even during periods of rising inflation, the index performed well in over half the cases (62.5%). Interestingly, during periods of very high inflation (above 3%), the index performed even better, with a success rate of 77% (Figure 6).

In addition, we analyzed how the index has performed during periods of rising inflation. Figure 7 shows the monthly return of the S&P BMV IPC compared with changes in inflation represented by the Mexico Consumer Price Index (CPI) since 2017. For periods corresponding to rising inflation, the index returns were positive 53% of the time; however, during periods of sharp increases in inflation (over 3%), this figure decreased to 49%.

Finally, we see that BIST 100 and S&P BMV IPC have generally performed well during periods of rising inflation, more so during sharp increases, and have provided significant real returns above inflation over the long run. However, BIST 100 is the stock exchange which benefited from inflation the most.

Moreover, the technology sector emerged as a key driver of growth in both economies during the pandemic. In India, the surge in remote work and digital adoption fueled the success of IT giants like Infosys and Tata Consultancy Services, propelling them to become major forces within Sensex. The Turkish market witnessed similar growth in software, service providers, technology retailers, and telecom, slightly influencing BIST 100. However, the overall influence of Turkish tech firms on the broader index remains less pronounced compared to their Indian counterparts.

Sensex boasts a more diversified composition, featuring established players across various sectors such as IT, banking, and consumer staples. This diversification fosters long-term value creation within the index. Conversely, BIST 100 has a higher concentration of cyclical sectors like banking and construction, which are more susceptible to fluctuations in inflation. While these sectors can benefit from low-interest-rate environments, they are also prone to short-term gains driven by monetary policy adjustments rather than fostering long-term value addition.

The post-pandemic journeys of the Indian and Turkish stock exchanges are distinct. India’s faster recovery and the dominance of IT within Sensex point towards a focus on long-term value creation. Conversely, BIST 100’s reliance on cyclical sectors and its sensitivity to inflation highlight a more short-term, inflation-driven dynamic. These contrasting landscapes offer valuable insights for investors navigating these evolving markets, allowing for them to tailor strategies based on risk tolerance and investment goals.

To sum up, the Indian stock market, Sensex, distinguishes itself from the Turkish market, BIST 100, in two key ways. Firstly, Sensex boasts a more diversified composition across sectors like IT, banking, and consumer staples, fostering long-term value creation. In contrast, BIST 100 relies heavily on cyclical sectors like banking and construction, which are more susceptible to inflation and prone to short-term gains rather than long-term growth. Secondly, the dominance of IT in Sensex, fueled by the pandemic’s surge in remote work, highlights India’s focus on long-term value creation compared to Turkey’s BIST 100, which is more sensitive to inflation and benefits from periods of rising inflation, particularly sharp increases.

To understand the long-term connection between these exchanges, DCC-GARCH graphs provide valuable insights. We generated DCC-GARCH graphs for BSE Sensex 50 compared to both BIST 100 and BMVIPC based on Impulse Response Function (IRF) results. These graphs depict the dynamic conditional correlation between the indexes on the y-axis, with trading days displayed on the x-axis. Positive values indicate that the markets move in tandem, while negative values suggest opposing movement. A value of zero signifies no correlation. The “dynamic” aspect emphasizes that the correlation between the markets can fluctuate over time.

Figure 8 and Figure 9 show that the correlation between the Mexican and Indian stock exchanges has been volatile over the past two years. There have been periods of high correlation, such as in early 2020, as well as periods of low correlation, such as in early 2022.

India’s stock markets are plunged today, with Sensex and Nifty suffering their biggest losses ever. This massive sell-off came amid a surge in coronavirus cases and the lockdown imposed by many states, which spooked investors. Sensex dropped 3934 points to close at 25,981, while Nifty fell 1135 points to 7610. During trading, both indexes dipped even lower, with Sensex reaching 25,880 and Nifty hitting 7583. Every stock in Sensex and Nifty closed in negative territory.

Indian equities experienced a significant downturn in response to the Russian invasion of Ukraine. This military action, the largest state-on-state attack in Europe since World War II, confirmed the worst apprehensions of the West. The benchmark BSE Sensex index plummeted by 2800 points, representing a decline of 4.72%, to close the day at 54,529.91. Similarly, the broader Nifty 50 index fell by 815.30 points, closing at 16,247.95. Since Turkey has a significant role in the Russia–Ukraine war, this correlation increase is quite relevant.

On 18 April 2022, India’s key stock indexes witnessed a significant plunge during opening hours. This drop, amounting to 1280 points or 2.2%, stemmed from a confluence of international worries. Investors grappled with persistent anxieties about global inflation, along with the prospect of a more aggressive interest rate hike by the US Federal Reserve.

Adding to the bearish sentiment were concerns regarding the weakening health of the Chinese economy, particularly in light of rising COVID-19 cases. These anxieties extended beyond the Indian market, impacting the S&P/BMV IPC in Mexico as well. This index, traditionally considered a hedge against inflation, also suffered losses, highlighting the interconnectedness of global financial markets and the vulnerability of even supposedly safe assets during periods of heightened economic uncertainty.

5. Conclusions

The theoretical framework for studying stock market volatility spillovers investigates how volatility fluctuations in one market can influence volatility in another. Initial research employed cointegration and VAR models to assess the interconnectedness of global markets, but these lacked the ability to capture the dynamic nature of volatility. To address this, ARCH and GARCH models were introduced, effectively incorporating time-varying volatility. However, their inability to account for feedback effects between markets led to the adoption of MGARCH models. These advanced models comprehensively analyze volatility spillovers by considering both the current impact of new information and the lasting influence of past volatility. Additionally, TARCH models delve into the asymmetric nature of spillovers, recognizing that negative shocks often have a more pronounced effect on volatility compared to positive ones. Finally, the framework acknowledges the importance of volatility spillovers between developed and emerging markets, particularly for portfolio diversification purposes. Here, research explores the varying degrees of integration between these markets, with some studies revealing a strong correlation between developed and specific emerging markets (e.g., Germany and Eastern Europe), while others find minimal spillover (e.g., UK and Turkey). The framework extends its focus to Asian markets and MINT countries (Mexico, Indonesia, Nigeria, and Turkey), where findings suggest lower integration within Southeast Asia and BRIC markets but a significant influence of the US market on Asian markets as a whole. While research on MINT countries is limited, some evidence points towards their integration with the US market.

Investors looking at emerging markets have two intriguing options: India and MINT economies (Mexico, Indonesia, Nigeria, and Turkey). India’s stock exchanges boast a long history and established regulations, leading to stability and investor confidence. India also boasts a heavyweight market capitalization with a wider range of companies and sectors for diversification. MINT exchanges, on the other hand, are younger and more dynamic, offering the potential for higher growth but also increased volatility. All these exchanges are embracing technology, and while India is a leader in digital trading platforms, MINT exchanges are catching up rapidly.

Each market reflects its domestic economy, so investment goals need to be considered when choosing a market. India is strong in financials, IT, and pharmaceuticals, while MINT economies offer exposure to different sectors. Finally, India has a well-developed regulatory framework, while MINT economies are still evolving theirs, and timezone differences can also be a factor depending on investor location. Ultimately, the best choice depends on investor risk tolerance, investment goals, and preferred sectors. Both regions hold promise for the long term, but it should be remembered that thorough research on individual companies within any market is crucial for success.

The pandemic caused both Indian and Turkish stock markets to plunge in early 2020. While both recovered, their paths differed. India’s market (BSE Sensex) took a bigger initial hit but bounced back faster and stronger. This is likely due to India’s younger population and strong economic base, which helped the country adjust to remote work and digital life quicker. Turkey’s market (BIST 100) had a bumpier recovery because of existing problems like inflation and weak currency, which became worse during the pandemic.

Tech was a bright spot for both economies. India’s tech giants like Infosys thrived thanks to the rise of remote work, becoming major players in the Sensex. Turkey’s tech sector, like e-commerce and internet providers, also grew, but their impact on the overall BIST 100 was not as significant as the Indian tech companies’ influence on Sensex.

Given the increasing adoption of high-frequency data and machine learning tools, future research could delve deeper into several areas. One exciting avenue is exploring how sentiment analysis from social media and news data, incorporated into machine learning models, can predict volatility spillover effects. Additionally, machine learning could be used to identify hidden patterns and non-linear relationships between MINT and developed markets, leading to a more comprehensive understanding of volatility transmission. Furthermore, with high-frequency data, researchers could investigate the time-varying nature of spillover asymmetry (how negative and positive shocks impact volatility differently) across different market pairs. Finally, machine learning could be employed to develop more dynamic portfolio diversification strategies that consider real-time volatility spillovers between developed and emerging markets.