In research literature, the terms “interdependencies” and “comovement” are used interchangeably. The word “comovement” is a financial industry jargon that is not included in ordinary dictionaries. D.Baur (2003) suggests that the word “moving with” or “sharing movement” be used instead of “comovement.” Barberis et al. (2005) describe comovement as a “pattern of positive association.” Interdependencies is the word used by K. J. Forbes & Rigobon (2002) to describe the phenomenon wherein markets show a great deal of comovement. The words “interdependencies” and “comovement” identify the link between two variables under both normal and crisis conditions but they imply no causal relationship between the variables themselves. There is a nuanced difference between a more general terminology like “contagion” and “spillovers.” The word “contagion” is frequently used to describe a highly negative event that spreads throughout a financial market (K. Forbes, 2012).

Another important term is “spillovers”. Spillovers relate to the direction of shock transmission through markets, which implies the presence of dominant markets (net providers of shocks) and dominated markets (net receivers of shocks). There are several theories seeking to explain the root causes of stock market interdependence. According to the first theory, stock market interdependence mimics international commerce and financial relationships. This paper is based on Williams (1938) book on the intrinsic worth of enterprises. While it was widely assumed in the 1930s that financial markets functioned like casinos, Williams believed that the value of a stock should be equivalent to the present value of its future cash flows. With the development of the efficient market theory in the 1960s, Williams’ views gained new significance. According to this theory, stock prices represent rational investors’ anticipation of the underlying worth of businesses.

R. Johnson & Soenen (2003) use data on daily prices of Latin American countries (Argentina, Brazil, Canada, Chile, Mexico, Colombia, Peru, and Venezuela) between 1988 and 1999 to investigate the level of integration between their markets and the US stocks market. The researchers concluded that the US equity market is statistically and substantially related to all eight economies analyzed in that research, with trade between them accounting for a big portion of the relationship.

Cross-border equity flows have expanded as a result of the liberalization of stock markets, providing businesses with access to previously unavailable funding and investors with new opportunities to diversify their portfolios globally. The proportion of GDP invested abroad by foreign shareholders increased from 16% to 87% in developed markets while in developing markets, this proportion increased fourfold, from 4% to 16%, during the same period (Lane & Milesi-Ferretti, 2018). Concurrently, international economic, and financial relations have expanded rapidly in recent years and so it is possible that markets will become more cointegrated and more correlations and comovement will be observed.

Another research on capital flows, published by Ekinci et al 2007 and Lane & Milesi-Ferretti (2018), shows that foreign investors now possess a larger share of the world’s foreign financial assets than they did a decade ago. Previous studies have demonstrated that the ongoing development has led to financial globalization, especially among developed economies. The greatest financial catastrophe since the Great Depression in 1929, the Global Financial Crisis (GFC) of 2008, caused serious doubts about the advantages of liberalizing the financial sector. In its wake, several studies examined the possibility of a widespread contagion (Pyun & An, 2016). Even before that, some have argued that the key transmission mechanism for cross-country shocks during the financial crisis of the 1990s was the presence of financial ties across countries (Baig & Goldfajn, 1999; Caramazza et al., 2004; G. L. Kaminsky & Reinhart, 2000; Van Rijckeghem & Weder, 2003). At the same time, during the GFC, Lane (2013) offered market-specific viewpoints as primary causes of financial integration. The interconnectedness of economies through real-sector and financial linkages acts as a crisis carrier when trouble begins in one national economy and spreads to others (Glick & Rose, 1999; G. Kaminsky et al., 1998; Van Rijckeghem & Weder, 2003).

Global phenomena or common shocks like major economic shifts in industrial countries, significant changes in oil prices, changes in US interest rates, and changes in exchange rates may also have a negative impact on the economic fundamentals of several economies simultaneously, potentially resulting in a crisis (Eichengreen et al., 1996). These effects might be seen as “spillovers” (Masson, 1999), “interdependence” (Forbes and Rigobon, 2002), and “fundamentals-based contagion” (all of which refer to the same phenomenon) (Kaminsky and Reinhart, 1998).

There’s also a theory holding that market flaws or the actions of foreign investors contribute to the international spread of financial crises from one country to another (Diamond & Dybvig, 1983; Dornbusch et al., 2000; King & Wadhwani, 1990; Kodres & Pritsker, 2002; Masson, 1999). When there are gaps in the available information, investors are more likely to be wary of a country’s economic fundamentals and insights. For example, uninformed and less-informed investors may have difficulty extracting information from the signal of falling prices and may instead choose to follow the strategies of better-informed investors, resulting in excess co-movements across markets. This often occurs when a crisis in one country serves as a “wake-up call” to international investors as it forces them to reconsider risks in other countries (Goldstein, 1998; Pasquariello, 2007; Yuan, 2005). The degree of (non)anticipation of a crisis by investors is crucial for the occurrence of contagion because of the allocation of investors’ attention (Mondria & Quintana-Domeque, 2013).

Evidence suggests that market confidence and expectations play a significant role in the propagation of contagion (Masson, 1999; Mondria & Quintana-Domeque, 2013). With the intention of creating an early warning system, the earliest empirical study on financial crises and contagion focused on fundamentals-based processes (Eichengreen et al., 1996; Van Rijckeghem & Weder, 2003). Later empirical research has zeroed emphasis on investor behavior-based mechanisms. There are still significant knowledge gaps in this area, particularly concerning the potential amplifying effect of investor behavior on market contagion and excessive market comovement. This paper seeks to fill one of such lacunae by investigating the potential effects of trust and financial development on market comovement during times of financial crisis and by testing them for homogeneity.

The contagion effect is often thought to be caused by and linked to a sharp increase in the degree to which stock markets move in tandem or in correlation (Duda et al., 2022). When thinking about how to best diversify one’s portfolio it is crucial to consider whether or not the stock markets have an abnormally high or low returns correlation.

Research has shown that financial development leads to more interconnected stock markets. For example, Nikkinen et al. (2011) in their study showed that they witnessed an increased interdependence between the Croatian and Slovenian markets before and during the global financial crisis. This was a result of financial development which led to more interconnected markets. Other research also shows that financial development may lead to increased international investment flows especially in emerging markets. These flows have an overall effect of making markets more integrated (Giofré, 2021; Goyal, 2014; Kant, 2018; Raj & Dhal, 2008). Other factors of financial development which promote stock market integration include but are not limited to, increased liquidity, investor protection and enhanced corporate governance. Gallimberti et al. (2021) explain in their study that stock market liquidity increases following banking deregulation as a form of financial development. This liquidity has the increased effect of making markets more correlated with similarly liquid markets (Ghossoub & Reed, 2012). The International Monetary Fund in their working paper specifically emphasized that the development of local markets in emerging markets had greatly improved the financial landscape over the past 20 years (International Monetary Fund. Monetary and Capital Markets Department., 2016). The IMF states that certain factors related to institutional development like improved corporate governance and investor protection are key elements that promote financial stability and international stock market integration.

Chevallier et al. (2018) also point out that financial development and globalization have led to a higher extent of market interdependence as firms now have access to international capital markets while at the same time providing finance at a lower cost. Financial liberalization has also been shown to reduce the cost of capital by as much as 42% (Errunza & Miller, 1998). This reduction in the cost of capital has the effect of making stock markets more integrated as the pricing of assets will not be much different. It has also been noticed that the international mobility of capital, especially cross-border private funds for investors seeking to diversify their portfolios, has also emerged as another source for financial market integration (Raj & Dhal, 2008). In the same paper, Raj & Dhal explain that financial development is effective in price discovery. This discovery has several benefits including the development of financial markets and institutions which might ultimately make way for stock market integration and comovement. Global computer and telecommunications technology, which forms the backbone of financial infrastructure has improved in recent years. This has contributed to the deepening and widening of financial services and hence to a more integrated financial sector as dissemination of information is fast and transaction costs are greatly reduced (Bhargava et al., 2004).

Most studies agree that financial development has both a direct and indirect role in making stock markets more integrated. Vithessonthi & Kumarasinghe (2016) give a full analysis of how financial development leads to more market integration. In their study, they show that financial development has a positive effect on stock market integration with the global factor or global stock market. This conclusion has been supported by other authors (Ben Rejeb & Boughrara, 2013; Kaneta, 2000).

“Trust is one of the most important synthetic forces within society” (Simmel 1950:326)

Simmel’s (1950) comparison undoubtedly gives us a glimpse of why trust is the cornerstone on which society is formed. Delhey & Newton (2003) also explain that there is a general consensus among contemporary social scientists that indeed social trust, or just trust for short, has gained a lot of attention not only in sociology but in many other academic fields including political science, economics, psychology, anthropology, history, political theory, business management and administration. Most importantly, it is agreed that trust boosts economic growth while also improving the efficiency of market economies. It refers to equitable resource allocation in a society, increases societal integration and peaceful cooperation, enhances personal happiness, political stability and simple things like good health and peace of mind. Finally, trust is the glue that connects social notions like optimism, well-being, health, education welfare, community engagement, and progress.

The concept of trust has its roots in theology, philosophy, socio-political theory, and ethics (Allahyarahmadi, 2013; Misztal, 2013). When there is a high level of danger, uncertainty or ignorance, trust is especially important as it is the willingness of an individual to engage in risky behavior in a social setting. Trust is defined by Fukuyama (1995) as the expectation that arises in members of a social group based on similar norms and beliefs that other members would act according to these norms. Fukuyama goes on to say that trust characterizes a situation in which neither party to a transaction exploits or takes advantage of the other’s flaws; it is an unwritten contract involving social interactions between people, organizations, and civic systems. It also encompasses confidence, expectations, motivation, civic collaboration, meeting obligations, and working together to achieve a unified objective.

Trust affects financial development both directly and indirectly. It promotes larger investments by individuals if they trust that legislation enforcement is adequate. (McCannon et al., 2014). The trusting nations usually have lower corruption and hence better overall financial development as participants are more likely to invest and contribute to the growth of the financial sector (Ben Rejeb & Boughrara, 2013; Kaneta, 2000). Trust in financial institutions along with financial literacy leads to an increase in stock market participation which has a lasting effect of promoting financial development. In their study, Balloch et al (2015) infer that stock market literacy and other behavioural traits, such as trust among them, may explain the level of equity investments. Lachance & Tang, (2012) also show that stock market participation increases with the level of trust in an economy. In countries with substantial social capital and high levels of trust, households often invest more in equity than other assets prompting an increase in financial development (Sapienza & Zingales, 2011; van Raaij, 2016).

In higher-trust nations, trade credit facilitation is often expedited between businesses (Severin et al., 2006). Previous studies have shown that in highly trusting countries businesses that depend on liquidity often receive trade credit and are least affected by declining profits and bank crisis as compared to similarly sized firms in less trusting countries (Levine et al., 2018). Research has also shown that this trade credit ultimately facilitates access to bank debt and this also improves financial development. Therefore, companies that enjoy high levels of trust are able to access bank debt and hence are less likely to suffer from credit constraints (Moro & Kodwani, 2010).

It is also suggested that in highly trusting countries, the influence of trust leads to lower transaction costs in financial operations for firms and investors (Moro & Kodwani, 2010; van Raaij, 2016). To contextualize this, Kwon et al. (2013) in their study found out that companies’ transaction costs can be reduced if partnerships formed on the foundation of trust lead to a collaborative supply chain effort. Trust provides the glue which holds contracts together. This is because it contributes to transparency and contractual safeguards for all parties involved, offering a different form of corporate governance (Thomson, 2011). On the other hand, Schmidt et al. (2003) show that trust has a direct influence on reducing transaction costs associated with handling uncertainty and promoting interpersonal interactions. The author explains that trust fosters growth and economic development by supporting the accumulation and, more importantly, the efficiency of physical and human capital accumulation.

Trust also exerts influence on financial efficiency, social capital, financial infrastructure, and specific demographic clusters (Bossone, 1999; Sangnier, 2011; Yin et al., 2020). In a study conducted by Özen (2019) the author states that there is a positive impact of trust on the growth and development of financial markets. Trust also has the effect of reducing perceived risk and credit risk in financial transactions and credit availability (Moro & Kodwani, 2010) because it is expected that trust can reduce agency cost and the need for collateral (Moro et al., 2012). As long as trust is on the rise or already high there is an increase in household participation in the stock market investment. (Lachance & Tang, 2012). As a result, trust impacts investment and contracting decisions while at the same time reducing the costs associated with investing (Bottazzi et al., 2016).

Calderon et. al. (2002) in their study show that trust is correlated with financial depth and financial efficiency both being characteristics of financial development. Since our task is to determine if trust has a mediating effect on financial development for stock market comovement we have suggested several hypotheses to empirically test our theory. These are the proposed hypotheses:

H1: Financial development does not affect the comovement of a BRICS country’s returns with the global factor proxied with the Dow Jones index.

A1: Financial development influences the comovement of a BRICS country’s returns with the global factor proxied with the Dow Jones index.

The concept of financial development includes two components: institutional development and market development; it is therefore necessary to test if they also influence comovement. So we can expand the first hypothesis to test each of these, leading to the following sub-hypotheses:

H1a: Institutional development does not affect the comovement of a BRICS country’s returns with the global factor proxied with the Dow Jones index.

A1a: Institutional development influences the comovement of a BRICS country’s returns with the global factor proxied with the Dow Jones index.

Then after answering these questions, we can further develop our final hypothesis which aims to determine if trust has any mediating effect on financial development. For this purpose, we develop the following hypotheses:

H2: Trust does not have a mediating effect on financial development for stock market comovement between BRICS and the US Dow Jones.

A2: Trust has a mediating effect on financial development for stock market comovement.

The study aims to investigate the relationship between trust and the comovement of the BRICS stock markets with the US Dow Jones as a global factor.

The primary metric of interest is each nation’s trust index, which can be found primarily on the World Values Survey (WVS) website (Worldvaluessurvey.org, 2019). By answering the poll questions the respondents indicate if they generally see themselves as trusting people. Several studies (Drobetz et al., 2023; Wei & Zhang, 2020) support this approach.

Since its beginning in 1981, the survey has worked to employ the most rigorous and high-quality research designs available in each country to provide a single, composite score representing the level of trust in its culture, in relation to both individuals (Almond & Verba, 1963) and organizations (e.g., Gallup Polls and WVS). Between 1981 and 2020 there have been seven waves of surveys; not all countries participated in each wave, so we will use the average method to fill in the gaps. The index ranges from 0 to 100, with 0 indicating the lowest level of trust and 100 the highest level of trust.

The OECD methodology is utilized by the WVS in order to provide recommendations on how to quantify trust (Murtin et al., 2018). The World Values Survey employs a wide-ranging definition of trust, one that is both easy to understand and useful for parsing out its component parts. For their purposes “trust” is defined as “the idea that another person or institution will act in accordance with one’s expectations of positive behavior.”

Morrone et al (2009) differentiate between conventional trust questions and attempts to gauge trust based on respondents’ anticipations of others’ actions, such as answers to a survey question on whether a lost wallet is likely to be returned. There is plenty of research linking trust survey items to participants’ actual trusting behavior in lab settings (Algan & Cahuc, 2010, 2013; Falk et al., 2023; Fehr & Fischbacher, 2003; Gächter et al., 2010; Glaeser et al., 2000; N. D. Johnson & Mislin, 2012; Lazzarini et al., 2005; Naef & Schupp, 2021). These WVS measures provide a vital insight into the validity of more traditional survey-based results and they have been used to produce better survey questions even though experimental methods of assessing trust are outside the purview of official statistics.

Because the WVS survey is not conducted annually and different nations have different survey periods, we will be using the most up-to-date scores available (from 2015) and will be calculating the average scores for each country using the Microsoft Excel package (Liu, 2019). It should be stressed that generalized trust is stable through generations since it is passed from parents to children, as evidenced by research (Uslaner, 2008). Average trust levels across countries in our study are shown in the table below.

Our analysis also uses DataStream’s comprehensive collection of daily stock prices for the BRICS and US country indexes between 2003 and 2017 to estimate the stock market returns for those specific countries. The period was chosen because the data set had to coincide chronologically and the recent results of the WVS survey were officially released in 2018.

Stocks, stock market indexes, currencies, business fundamentals, fixed-income securities, and important economic indicators are all part of DataStream, a worldwide financial and macroeconomic database that covers more than 175 countries and 110 markets. More than 3.5 million different financial instruments from all over the world, with a combined 60 years of historical time series data, are available for analysis.

Table 2 below shows a summary of the daily stock returns for the US and BRICS nations for the total period under review. We can see that, as a result of the global financial crisis in 2008, the average returns, including those previously on an upward trajectory, were extremely low owing to the recession that wiped them out.

To address the limitations of single indicators as proxies for financial development, the IMF developed a series of indices that summarize the depth, accessibility, and efficiency of developed financial institutions and financial markets, culminating in the final index of financial development shown in Figure 1.

The Financial Development Index is shown below: First, the variables are normalized, then the normalized variables are aggregated into sub-indices that reflect a certain functional dimension; and, third, the sub-indices are aggregated into the final index. The indices range from 0 to 1 with 0 being the least developed while 1 is fully developed.

In order to assess the breadth, depth, and efficiency of financial institutions and markets, a series of factors are employed to generate six subsidiary indices, which are depicted at the base of Figure 1. Sub-indices include those denoted by the letters FID, FIA, FIE, FMD, FMA, and FME, where I and M stand for “institutions” and “markets,” respectively, while D, A, and E stand for “depth,” “access,” and “efficiency,” respectively. The growth of financial institutions and markets is measured by combining these indexes into the FI and FM composite indices.

The FD index, a composite of the FI and FM components, provides a comprehensive measure of financial development. Many facets of the economic system may be summarized by monitoring a select group of these important indicators. The statistical variables are selected only if they provide data for a sufficient number of countries over a sufficient time span. Also the database draws on a set of fundamental proxy variables that are both well-established and available over a wide country-time sample.

Finally, these indexes were developed with ease thanks to the collection’s 33 years of yearly data for 183 developed, emerging, and low-income developing nations from 1980 to 2013. The summary indices for the BRICS nations are given below in Table 3.

The Panel Autoregressive distributed lag (ARDL) and its counterpart Panel Autoregressive distributed lag (ARDL) were used in the subsequent parts of the analysis on financial development and stock market participation. The fully modified ordinary least squares (FMOLS) method was also utilized. Pesaran and Shin pioneered the autoregressive distributed lag (ARDL) method, sometimes known as the Bounds test (Pesaran & Shin, 2012). It is widely regarded as one of the most adaptable econometric approaches. Furthermore, the ARDL method’s ability to accept varied lags in multiple variables makes it particularly appealing, versatile, and flexible.

Econometric Issues: The verification of the presence of the long-run equilibrium relationship between variables is a significant econometric problem. The (Engle & Granger, 1987; Johansen, 1991) methods can be used to keep this from happening. Comparatively, the Engle-Granger (EG) process relies on assessing the stationarity of the regression residuals, whereas the Johansen method makes use of Vector Autoregressive (VAR) models. On the other hand, the Johansen method includes a test of the hypothesis of the long-run equilibrium relationships, whereas the EG process did not include testing of the hypothesis on the co-integrating relationships themselves. The sequence of integration for each series I(d) of variables is also an essential question. The Augmented Dickey-Fuller (ADF) test is a useful tool for doing this (unit root test).

Stationarity Test (Unit Root Test) To determine whether or not to apply OLS, the time series’ stationarity property must be evaluated because most macroeconomic variables are nonstationary; it results in a very high R² when parameters are estimated using OLS, and the emergence of false regression problems may be caused by a non-stationary process. The ADF test (Augmented Dickey-Fuller) is employed. The ADF test is written in the following format:

The significance of the coefficient of (Y_t-1) is tested in the unit root test, and the hypothesis of a unit root cannot be rejected if the ADF test-statistic (t-statistic) is smaller (in absolute value) than the Mackinnon critical values. There is a family of closely comparable statistical tests that includes the Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test, the Phillips-Perron (PP) test, the Ng-Perron test, and the cross-sectional augmented IPS-CIPS test. This evaluation will make use of the ADF as well as one other test.

Co-integration Test: If the ADF results demonstrate that the variables are integrated of order one I(1), then it is important to identify at least one stable and non-spurious linear combination I(0) of these variables. Johansen co-integration was used to determine the total number of co-integrated vectors for a set of “n” nonstationary variables of the same order. Because the Johansen test is hypersensitive to the lag length employed in the VECM, the Akaike Information Criterion (AIC) and the Schwartz Bayesian Criterion (SBC) are employed to determine the optimal lag length.

This study used the FMOLS method to analyze the correlation between the BRICS countries’ stock market returns and the US market, considering both the countries’ levels of financial development and their level of involvement in the OECD. For the purpose of estimating a single co-integrating relationship involving i and j variables the FMOLS is used. (Phillips & Hansen, 1990) first proposed and then refined the FMOLS method.

To get around the inference problem inherent in OLS approaches, the t-test for long-run estimates can be applied when using the FMOLS method instead (Himansu, 2007). Using “Kernal estimators of the Nuisance parameters that alter the asymptotic distribution of the OLS estimator,” FMOLS “fully modifies” the conventional ordinary least squares (OLS) method. This method uses an adjustment to the least squares method to consider the impacts of serial correlation and to test for endogeneity in the repressors that result from the presence of Co-integrating relationships, allowing for asymptotic efficiency to be achieved.

For this analysis the models which will be used for financial development and stock market comovement, the following apply;

Model 1: f(correlation, institutions_index, market_index)

Model 2: f(correlation, institution_access, institution_depth, institution efficiency)

Model 3:f(correlation, market_access, market_depth, market efficiency)

It follows that we can interact the factors of financial development with trust to achieve H2 using the following equations.

Model 4:f(correlation, trust, inst_access, inst_access*trust, inst_depth, inst_depth*trust, inst_efficiency, inst_efficiency*trust, mkts_access, mkts_access*trust, mkts_depth, mkts_depth*trust, mkts_efficiency, mkts_efficiency*trust)

In the above models 1 to 4 the natural logarithms are used to transform the data to remove any impediments.

For this analysis, the study aims to determine the long-term effects of institutional and market development on the stock market comovement for the BRICS markets with the Dow Jones. With only three variables used in this investigation, the methods employed should help overcome some of the difficulties usually experienced in OLS. The suitable models are therefore either the Panel FMOLS or the Panel ARDL depending on the results of the unit root test which will employ both the ADF and PP tests with three trend specifications shown in Table 4 below.

From Table 4 above, it can be inferred that the data is stationary at the first difference and not at levels. This is true for both the PP and the ADF tests which are most significant at the 1% significance level with integration order I(1). This is the main key to determining whether the FMOLS model can be used, as it should only be used when the order of integration is I (1) for all the data. Also, cointegration analysis is performed on the data to verify the appropriate use of the modelling method. In Table 5 below the three main Pedroni cointegration tests are analyzed.

To determine if the FMOLS method is appropriate in this analysis the Pedroni cointegration tests are necessary. Table 5 above shows the results for these tests with the first one being non-deterministic trend statistics. Its result is that there is cointegration: 4 out of 7 tests are significant at both the 1% and 5% levels. The second test is deterministic intercept; its trend model of which 4 out of 7 statistics also show that the data is cointegrated. The last one is the non-deterministic trend and intercept test which has 4 significant statistics, though most are weakly significant at the 10% level. All three tests indicate the existence of a long-term relationship and cointegration. The FMOLS results are presented in Table 6 below.

The FMOLS results in Table 6 above show that it is only financial institution development that is positively associated with the return correlations. This means that, as financial institutions in emerging markets are developing, one can expect that their stock market returns will be integrated with the global factor and comove with the US Dow Jones.

Further analysis of the individual make-up of the financial institution’s development index aims to find out how its basic composition affects the stock markets’ comovement. A similar approach is undertaken with the unit root tests to determine the degree to which the data is integrated. The unit root results are shown in Table 7 below.

Table 7 above shows that the data are mixed, some of the variables being of order I (1) while others integrated to order I (0). This rules out the use of the FMOLS method as the panel ARDL model is suitable for the analysis. The next step is to determine the optimal lag length of the model, which is heterogeneous because the number of variables is less than the number of years, as is shown in Table 8 below.

Table 8, showing the lag length criterion, indicates that the optimal lag length is two. That is why in our model we will only go up to lag 2 as depicted by the criterion. Next, we show the long-run model in Table 9 below.

The chosen model is of type ARDL(1,2,2,2), selected by the software. The optimal long-run estimates for the model in Table 9 above show that institutional access and institutional depth are positively and significantly associated with stock market return correlation in the long run. This means that developing financial institutional access and institutional depth will have a positive effect on the comovement of the stock markets in the long run for the BRICS economies.

This makes sense as greater financial access and improved depth of financial institutions will promote savings and hence increase credit lines for businesses. High-performing stock markets will tend to have their financial returns synchronized with the global factor. Institutional efficiency, on the other hand, is only weakly associated with the stock market correlation. Next, one can find out if there are any short-run relationships for the same system of variables. These are shown below in Table 10.

Table 10 above shows that the model displays error correction as the cointegrating factor is negatively significant at the 1% level. This implies that if there are shocks in the system the model will error correct within 68% of one year. Institutional depth also has a negative short-run effect on the stock market correlations but this is weakly significant at the 10% significance level. Institutional efficiency has a positive short-run effect also at a weak 10% significance level.

These H1a-related results show that financial institutions’ development is a significant parameter in explaining how emerging markets are correlated with the global factor, and, more importantly, institutional depth and access play a vital role in determining the comovement of stock markets.

For financial institutions’ development, a similar analysis using the composition of financial markets development is done. The unit root test is used to determine stationarity (Table 11)

Table 11 above shows that there might be a presence of mixed integration of I(0) and I(1) implying that even though the data is stationary it might not be beneficial to continue along the panel FMOLS and therefore it might be beneficial to use the panel ARDL method. The next step is to determine the lag order structure of the data shown in Table 12 below.

In the above Table 12 the optimal lag length structure is shown to be 1. The long-run results are given below in Table 13.

The optimal model selected by the software was the ARDL (1,1,1,1). From the long-run model in Table 13 above, one can observe that market access is positively related to the stock market correlation at the 5% significance level. This implies that in the long run when financial market access improves, it will tend to lead to the stock returns of emerging markets moving together with the global factor being the US Dow Jones. This is the same for market depth, which is positively and significantly related to the comovement in the long run.

Financial markets make funds available for businesses; this improves the business environment in which companies are operating causing stock markets to have high returns. Financial market efficiency seems to have a negative effect on stock market comovement in the long run. The study analyzes the short-run effects of the market factors on the correlation shown in Table 14.

The short-run model in Table 14 shows that there is a presence of error correction in the system as the cointegrating factor is negatively and statistically significant at 1% level. This means that if there is a disturbance in the system it will self-correct 51% of the time. Financial markets efficiency is positively weakly significant at the 10% level implying that in the short term, any sudden positive changes in market efficiency will also lead to more correlated stock returns. It has been observed that financial market development, in the long run, is positively associated with the integration of stock markets into the global financial system. Therefore, as markets deepen and get more accessible, they might be at risk of financial contagion as they move in together with other financial markets, especially the US market. This is consistent with the H1b hypothesis.

The trust index used for each country is an average of the country-specific scores. The score for each country still ranges from 0 to 100, with 0 being the least trusting and 100 as the most trusting. Also, the focus shall be on the interacting terms.

In Table 15 above the first interaction term is with trust and institutional access; however, the term is not significant. The second term is the interaction between institutional depth and trust which is positively significant at 5% level. This indicates that, since trust is a stationary variable, any positive change in the depth of financial institutions will be associated with a positive increase in the comovement of emerging stock markets with the global factor proxied as the Dow Jones.

Institutional efficiency is also positively moderated by trust. One can see in the table that the interaction term is positively and significantly associated with the correlation variable at 1% level. Therefore, any positive change in the institutional efficiency along with high trust will lead to an increased comovement of the emerging markets with the US market.

If investors’ trust levels are high they are willing to deal and cooperate with financial institutions: transactions are cheap and there is no need for expenses like litigation costs. Financial institutions that have sufficient depth and function smoothly and efficiently can also drive capital to viable businesses facilitating their expansion. This contributes to building a business environment where financial markets thrive and wealth is created. In such an economy trust plays a crucial role, becoming another “invisible hand”. This can be achieved through financial development.

Overall, this wealth creation process is what makes stock markets thrive. When two countries enjoy high trust levels the wealth creation process in both of them tends to cause their stock markets to move together. This is also true for financial market efficiency.

It is thus possible to posit the existence of evidence suggesting that financial development is positively associated with stock market comovement answering our H1. Our hypothesis sought to test whether trust is a mediating factor for financial development in market comovement and we have found out that, indeed, it plays a role in promoting comovement answering to our H2 hypothesis.