Can Human Capital Drive Sustainable International Trade? Evidence from BRICS Countries

Abstract

This paper examines the causal relationship between human capital and economic factors in BRICS countries using a panel vector autoregressive model and data from 1997 to 2020. The economic factors considered include foreign direct investment (FDI), imports, exports, and gross domestic product (GDP). The study conducts a comparative analysis of Brazil, India, China, Russia, and South Africa by adopting a vector autoregressive (VAR) model. The findings indicate a bidirectional causality between human capital and FDI in China, while a unidirectional causality from FDI to human capital is observed in Brazil. Moreover, a unidirectional causality exists from human capital to GDP in Brazil, Russia, India, and South Africa. Additionally, a unidirectional causality is found from human capital to imports and exports in South Africa. Overall, the results suggest the pivotal role of human capital in achieving sustainable economic development in BRICS countries. Policymakers should ensure sustained investment in human capital, focusing on economic growth, FDI, and international trade.

1. Introduction

The education of the labor force and the accumulation of human capital significantly influence a country’s ability to generate new ideas and adapt existing ones. According to endogenous economic growth theory, investing in human capital, innovation, and knowledge are key drivers of economic growth.

Beyond its direct economic impact, human capital enhances domestic productivity through channels of international technological spillovers, such as foreign direct investment (FDI) and international trade. FDI brings advanced technologies to the host country, and foreign firms can contribute significantly, directly or indirectly, to the host country’s innovative activities. Simultaneously, international exchanges in tangible intermediate inputs, manufactured goods, and capital equipment lead to the efficient utilization of domestic resources, thus boosting domestic productivity. Furthermore, it facilitates open communication among trade partners, fostering “cross-border” learning about foreign technologies, production processes, and materials, contributing to economic growth. In turn, economic growth can spur an expansion of exports, fostering knowledge spillovers and other externalities, creating virtuous circles of cumulative causation. Particularly, further economic growth generates new needs that cannot be met by domestic production alone, resulting in a further increase in the level of imports.

The economic role of human capital has garnered considerable attention. Previous research suggests various relationships, including long-run relationships [1,2], dynamic links [3,4], and one-way or two-way causal relationships [5,6] between human capital and economic growth. Other studies focus on the impact of human capital on FDI or international trade [7,8,9]. Meanwhile, some studies investigate whether FDI or international trade stimulates human capital accumulation [10,11,12,13,14]. Additionally, a few studies identify two-way causal links between human capital and FDI or international trade [15,16,17].

However, very few studies have integrated human capital and economic factors (e.g., GDP, FDI, and international trade) into a panel vector autoregressive (VAR) model with a specific focus on BRICS countries. Notably, the BRICS countries encompass 40% of the world’s population and contribute to more than 25% of the global gross domestic product (GDP). Their participation in world GDP is anticipated to rise significantly in the coming years. To achieve sustainable economic development, understanding how to enhance the economic role of human capital in BRICS countries becomes crucial.

The purpose of this paper is to uncover the causal relations between human capital, FDI, imports, exports, and GDP in BRICS countries—Brazil, Russia, India, China, and South Africa—utilizing a VAR model. To the best of our knowledge, this study represents the first comparative analysis of these five BRICS countries.

The remainder of this paper is organized as follows: Section 2 provides a summary of the existing literature and presents the hypotheses. Section 3 describes the methodology and data. Section 4 discusses our empirical results. Finally, Section 5 presents the conclusions and policy implications of this study.

2. Literature Review

2.1. Human Capital and FDI

Numerous studies have underscored the positive impact of human capital on Foreign Direct Investment (FDI). Miyamoto [18] and Noorbakhsh [19] highlighted human capital as a crucial factor in attracting FDI. Thangavelu and Narjoko [20], analyzing data from the Association of Southeast Asian Nations (ASEAN) from 2000 to 2009, and Dorożyńska et al. [21], using Polish data, both confirmed the positive effect of human capital on FDI. Cleeve et al. [22], employing various human capital indicators, found that human capital positively affected FDI across education levels in Sub-Saharan African countries from 1980 to 2012. Kim et al. [23] demonstrated, using bilateral FDI and foreign student data from 63 countries over 1963–1998, that a host country’s foreign-educated workforce positively influences FDI inflows. Kheng et al. [15] identified a positive bidirectional causality between FDI and human capital based on data from 55 developing countries from 1980 to 2011.

Conversely, some studies have explored the impact of FDI on human capital accumulation. Wang [13] found, using state-level data from the United States over 1997–2004, that informational FDI increases human capital while manufacturing FDI has the opposite effect. Zhuang [14] studied East Asia and discovered that FDI positively affects secondary schooling but is negatively related to tertiary schooling.

2.2. Human Capital and International Trade

Some studies have recognized the role of human capital in comparative advantages and export performance. Andersson and Johansson [7] concluded, using firm-level data from Sweden, that the spatial distribution of human capital influences both the quantity and unit price of exported products. Zhao et al. [9], using Chinese firm data, found that investments in salesperson human capital positively affect export performance.

Conversely, research has shown that trade also influences human capital. Lee and Wie [24] and Kasahara et al. [12] provided evidence that the use of foreign intermediate goods in production leads to skill-based technological changes, widening wage inequality, and increasing the demand for skilled labor.

Recently, scholars have examined how changes in trade patterns influence human capital accumulation. Blanchard and Olney [25] found that an increase in skill-intensive export volume enhances investment in human capital, while the expansion in agriculture and unskilled-intensive exports decreases human capital formation, a finding supported by Hou and Karayalcin [11] and Li et al. [26].

2.3. Human Capital and Economic Growth

Many scholars affirm that human capital positively affects economic growth by improving labor productivity and facilitating the adoption of new technologies. Fleisher et al. [27] reported in their study of China that human capital has both a direct and indirect positive effect on economic growth through innovation activity and technology spillovers. Ahmad and Khan [3], Ha and Manh [28], and Matousek and Tzeremes [29] provided additional evidence supporting the positive economic effect of human capital.

From a causality perspective, Tsen [17] and Anoruo and Elike [5] identified a bidirectional positive relationship between human capital and economic growth based on data from China and 29 African countries, respectively. However, Kalaitzi [30] reported a unidirectional causality from human capital to economic growth in their study of the United Arab Emirates (UAE).

The role of foreign technology transfers and the absorption of foreign technology on productivity has garnered attention. Teixeira and Fortuna [2] found, using imports as a channel of diffusion of foreign technology, that human capital and imports are complementary in improving productivity. Soukiazis and Antunes [31] reported positive direct and interactive effects of trade and human capital on economic growth using data from 14 European Union countries. Furthermore, studies by Ali et al. [32], Agbola [1], and Su and Liu [33] supported the idea that a sufficient stock of human capital enhances the positive economic effect of FDI through international technological spillovers.

This thesis investigates the influence of digitalization, mobile technology, ICT, and cross-border e-commerce on global economic variables and gender disparities in labor markets. It uncovers that, in high-income countries, digitalization has a positive impact on the gender gap, whereas in low- and middle-income nations, this impact is negative. Additionally, mobile technology and ICT play significant roles in fostering economic growth, particularly in developing economies. The study also finds that cross-border e-commerce has a substantial positive effect on trade in services. The findings underscore the importance of implementing technology-related policies to support both economic growth and gender equality [13,34,35,36].

3. Empirical Analysis

3.1. Modeling Framework

The research question is: to what extent does the quality and investment in human capital influence the sustainability and patterns of international trade in BRICS countries, and how do variations in human capital across these nations contribute to differences in their trade competitiveness and resilience amid global uncertainties?

To examine the relationship between human capital and economic factors in BRICS countries, this paper adopts a vector autoregressive (VAR) model. The economic factors include FDI, import, export, and GDP. The majority of studies have considered a one-way relationship of human capital with one or two economic factors. Therefore, the biggest difference of this model from other approaches is that it considers a two-way relationship of human capital with four economic factors, which allows us to examine the interactions among these five indicators.

The VAR model is widely used for examining the dynamics as well as the causality relationship on economic issues [34,37,38,39,40]. The interactions among human capital, FDI, import, export, and GDP can be examined by developing five VAR models from the prior paper as follows:

Model (1):

$$\mathrm{g}{\left(HC\right)}_{i,t}={\alpha }_1+\sum _{j=1}^P{\beta }_{1,t} \mathrm{g}({HC)}_{i,t-k}+\sum _{j=1}^P{\gamma }_{1,t} \mathrm{g}{\left(FDI\right)}_{i,t-k}+\sum _{j=1}^P{\delta }_{1,t} \mathrm{g}({IM)}_{i,t-k}+\sum _{j=1}^P{\theta }_{1,i} \mathrm{g}{\left(EX\right)}_{i,t-k}+\sum _{j=1}^P{\mu }_{1,t} \mathrm{g}{(GDP)}_{i,t-k}+{\epsilon }_{1i,t}$$

In Model 1, human capital (HC) is used as the dependent variable, while FDI (FDI), import (IM), export (EX), and GDP (GDP) are used as independent variables.

Model (2):

$$\mathrm{g}{(FDI)}_{i,t}={\alpha }_2+\sum _{j=1}^P{\beta }_{2,t} \mathrm{g}{(HC)}_{i,t-k}+\sum _{j=1}^P{\gamma }_{2,t} \mathrm{g}{(FDI)}_{i,t-k}+\sum _{j=1}^P{\delta }_{2,t} \mathrm{g}{(IM)}_{i,t-k}+\sum _{j=1}^P{\theta }_{2,t} \mathrm{g}{(EX)}_{i,t-k}+\sum _{j=1}^P{\mu }_{2,t} \mathrm{g}{(GDP)}_{i,t-k}+{\epsilon }_{2i,t}$$

In Model 2, the dependent variable is FDI (FDI), and the independent variables are human capital (HC), import (IM), export (EX), and GDP (GDP).

Model (3):

$$\mathrm{g}{(IM)}_{i,t}={\alpha }_3+\sum _{j=1}^P{\beta }_{3,t} \mathrm{g}{(HC)}_{i,t-k}+\sum _{j=1}^P{\gamma }_{3,t} \mathrm{g}{(FDI)}_{i,t-k}+\sum _{j=1}^P{\delta }_{3,t} \mathrm{g}{(IM)}_{i,t-k}+\sum _{j=1}^P{\theta }_{3,t} \mathrm{g}{(EX)}_{i,t-k}+\sum _{j=1}^P{\mu }_{3,t} \mathrm{g}{(GDP)}_{i,t-k}+{\epsilon }_{3i,t}$$

Model 3 presents import (IM) as the dependent variable, and human capital (HC), FDI (FDI), export (EX), and GDP (GDP) as the independent variables.

Model (4):

$$\mathrm{g}{(EX)}_{i,t}={\alpha }_4+\sum _{j=1}^P{\beta }_{4,t} \mathrm{g}{(HC)}_{i,t-k}+\sum _{j=1}^P{\gamma }_{4,t} \mathrm{g}{(FDI)}_{i,t-k}+\sum _{j=1}^P{\delta }_{4,t} \mathrm{g}{(IM)}_{i,t-k}+\sum _{j=1}^P{\theta }_{4,t} \mathrm{g}{(EX)}_{i,t-k}+\sum _{j=1}^P{\mu }_{4,t} \mathrm{g}{(GDP)}_{i,t-k}+{\epsilon }_{4i,t}$$

Model 4 uses export (EX) as the dependent variable, and human capital (HC), FDI (FDI), import (IM), and GDP (GDP) as the independent variables.

Model (5):

$$\mathrm{g}{(GDP)}_{i,t}={\alpha }_5+\sum _{j=1}^P{\beta }_{5,t} \mathrm{g}{(HC)}_{i,t-k}+\sum _{j=1}^P{\gamma }_{5,t} \mathrm{g}{(FDI)}_{i,t-k}+\sum _{j=1}^P{\delta }_{5,t} \mathrm{g}{(IM)}_{i,t-k}+\sum _{j=1}^P{\theta }_{5,t} \mathrm{g}{(EX)}_{i,t-k}+\sum _{j=1}^P{\mu }_{5,t} \mathrm{g}{(GDP)}_{i,t-k}+{\epsilon }_{5i,t}$$

In Model 5, the dependent variable is GDP (GDP), while the independent variables are human capital (HC), FDI (FDI), import (IM), and export (EX).

$\alpha ,\beta ,\gamma ,\delta ,\theta ,\mathrm{a}\mathrm{n}\mathrm{d}\mu$ are the coefficients of the lagged regressors; these parameters represent the impacts of the explanatory series on the dependent series. $\mathrm{g}\left(HC\right),\mathrm{g}\left(FDI\right),\mathrm{g}\left(IM\right),\mathrm{g}\left(EX\right),\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{g}\left(GDP\right)$ represent the growth rate of human capital, FDI, import, export, and GDP, respectively. i denotes country, while t indicates time. P is the optimal lag, and ε is the error term.

Before employing a VAR model, some steps should be taken. First, the variables should be stationary. Therefore, a stationary analysis should be carried out. If the variables are stationary, the unconstrained VAR model can be conducted directly. If the variables are not stationary, then the stationary forms should be used for estimation. Next, the optimal lag length (P) should be determined in advance by considering the information criteria.

3.2. Data Sources

The current study uses annual time series data from Brazil, Russia, India, China, and South Africa over the period of 1990–2020.

Our variables consist of human capital, FDI, GDP per capita, and international trade comprising imports and exports.

Following Blanchard and Olney [25], and Li et al. [26], we use average years of schooling to measure human capital, which is derived from the Human Development Report. The selection of the average years of schooling as an indicator aims to capture a quantitative representation of the educational attainment within each country. This measure is widely employed in empirical studies and international comparisons due to its availability in global datasets and its reflection of educational quantity.

The data on FDI inflows are taken as the percentage of GDP, which is obtained from the United Nations Conference on Trade and Development (UNCTAD) data set. The data on trade and GDP per capita are from the World Development Indicators (WDI) database. Imports and exports are used as the percentage of import or export value of total trade to GDP.

Table 1. Summary of descriptive statistics.

Variables	Mean	Median	Max	Min	Std. Dev.	Obs.
HC	0.844	0.845	1.079	0.471	0.153	137
FDI	0.086	0.215	0.777	−2.589	0.519	137
IM	1.266	1.313	1.684	0.843	0.172	137
EX	1.297	1.339	1.795	0.828	0.203	137
GDP	3.429	3.499	4.204	2.479	0.473	137

presents the descriptive statistics.

4. Results

4.1. Panel Unit Root Tests

To check the stationarity of the variables, we employ the panel unit root tests of Levin, Lin, and Chu (LLC), and Im, Pesaran, and Shin (IPS). The null hypothesis of these two panel unit root tests is that there is a unit root (non-stationary).

Table 2. Panel unit root tests.

Variables	LLC Test		IPS Test
	Level		Level
	Statistics	p Value	Statistics	p Value
g(HC)	−8.75067	0.0000 ***	−10.2339	0.0000 ***
g(FDI)	−6.28389	0.0000 ***	−7.77192	0.0000 ***
g(IM)	−9.17016	0.0000 ***	−9.32210	0.0000 ***
g(EX)	−11.5880	0.0000 ***	−11.0823	0.0000 ***
g(GDP)	−6.82058	0.0000 ***	−5.78060	0.0000 ***

Note: ***: significant at 1%.

reports the results of unit root tests for level variables. For all of our variables, the null hypothesis can be rejected for level variables with a 1% significant level. It means all level variables are stationary, I (0).

4.2. Optimal Lag Order Analysis

Next, the number of optimal lags should be defined in advance. We use the Likelihood Ratio Criterion (LR), Final Prediction Error Criterion (FPE), Akaike Information Criterion (AIC), Final Prediction Error Criterion, and Hannan–Quinn Information Criterion (HQ) for the optimal lag selection. In

Table 3. VAR Lag Order Selection Criteria.

Lag	LogL	LR	FPE	AIC	SC	HQ
0	923.6354	NA	5.17 × 10−14	−16.40420	−16.28284 *	−16.35496
1	967.9566	83.89369 *	3.66 × 10−14 *	−16.74922 *	−16.02105	−16.45378 *
2	983.0912	27.29649	4.38 × 10−14	−16.57306	−15.23808	−16.03142
3	1003.845	35.57775	4.76 × 10−14	−16.49723	−14.55545	−15.70939
4	1016.351	20.32288	6.02 × 10−14	−16.27413	−13.72554	−15.24009

Note: * indicates lag order selected by the criteria.

, most of the criteria, namely LR, FPE, AIC, and HQ, suggest that one lag in our VAR model is the most reasonable. Therefore, we use VAR (1) for estimations.

4.3. Panel VAR Estimates for BRICS Countries

After determining the optimal number of lags, our VAR models were estimated to investigate the links between human capital, FDI, imports, exports, and GDP. All the variables, in their growth form, were found to be stationary; therefore, level time series were used for estimation.

Table 4. Regression results for panel data of BRICS countries.

	Mode (1)	Mode (2)	Mode (3)	Mode (4)	Mode (5)
	Dependent Variable
Independent Variable	HC	FDI	IM	EX	GDP
HC (−1)	0.027	0.587	−0.474	−0.639 **	0.771
FDI (−1)	0.007 ***	−0.255 ***	0.005	−0.0004	−0.005
IM (−1)	0.018	2.111 **	0.213 *	0.173	−0.315 *
EX (−1)	0.001	−1.443	−0.073	0.078	0.269
GDP (−1)	−0.017	−0.323	0.004	−0.083	0.377 ***
Causality direction	FDI→HC	IM→FDI	-	HC→EX	IM→GDP

Note: *: significant at 10%; **: significant at 5%; ***: significant at 1%.

presents the estimates of five VAR models for BRICS countries as a whole.

In Model 1, the dependent variable is human capital. FDI lagged by one period has a positive impact on human capital at a 1% significance level. Specifically, a 1% increase in FDI can lead to a rise in human capital by 0.007% in the one-period lag term. International trade and GDP with a one-period lag have no impact on human capital.

In Model 2, FDI is used as the dependent variable. Imports lagged by one period positively affect FDI at a 5% significance level: a 1% increase in imports causes FDI to rise by 2.111%. Additionally, FDI, exports, and GDP have no effect on FDI in the one-period lag term.

In Model 3, we use imports as the dependent variable. However, the results show that human capital, exports, and GDP with a one-period lag have no impact on imports.

In Model 4, the dependent variable is exports. Human capital lagged by one period has a negative effect on exports, significant at a 5% level. Specifically, a rise in human capital can lead to a decline in export volume by 0.639% in the one-period lag term. Additionally, FDI, imports, and GDP lagged by one period have no impact on exports.

In Model 5, GDP is used as the dependent variable. Imports with a one-period lag are negatively related to GDP at a 10% significance level: a 1% growth in import volume causes a decline in GDP by 0.315%. Human capital, FDI, exports, and GDP, with a one-period lag, show no impact on GDP.

From the above results, it can be concluded that, firstly, there is a unidirectional causality from FDI to human capital in the one-period lag term. A rise in FDI inflows will increase human capital. Secondly, the existence of a unidirectional relationship is found between human capital and exports in the one-period lag term, indicating that the rising level of human capital will lead to a decline in export volume. Additionally, there are unidirectional causalities among economic factors: from imports to FDI, and from imports to GDP.

4.4. Comparison Analysis for Brazil, Russia, India, China, and South Africa

To examine whether the impact differs across countries, we undertook a detailed causality analysis in Brazil, Russia, India, China, and South Africa. Before establishing the VAR models, we employed unit root tests on each country to check the stationarity of all the variables. The results of the unit root tests show that all the variables in the growth form are stationary; therefore, we use the level time series for estimations.

Table 5. Regression results for Brazil, Russia, India, China, and South Africa.

	Brazil	Russia	India	China	South Africa
		Dependent variable: HC
FDI (−1)	0.015 **	0.005	−0.0009	0.023 *	0.005
IM (−1)	0.061	−0.005	0.110	−0.021	0.062
EX (−1)	−0.027	0.012	−0.092	0.009	−0.061
GDP (−1)	−0.016	0.0001	−0.047	−0.011	−0.049
		Dependent variable: FDI
HC (−1)	−3.534	−0.037	10.796	−5.016 ***	−0.657
IM (−1)	0.742	0.745	−1.751	0.514	8.640
EX (−1)	−2.809 ***	−0.678	3.195	−0.387	−4.868
GDP (−1)	−0.580	−0.447	1.267	−0.274	−0.028
		Dependent variable: IM
HC (−1)	0.380	2.868	−1.332	−0.643	−0.686 *
FDI (−1)	0.041	−0.068 **	−0.054	−0.005	0.015
EX (−1)	−0.105	−0.264	0.468	−0.012	−0.177
GDP (−1)	−0.048	0.030	0.500 *	−0.623	0.282 *
		Dependent variable: EX
HC (−1)	−2.310	3.837	−0.625	0.027	−0.757 **
FDI (−1)	0.039	−0.026	−0.067 *	−0.057	0.005
IM (−1)	−0.282	0.843 **	0.249	0.277	0.077
GDP (−1)	−0.100	−0.111	0.303	−0.553 *	0.105
		Dependent variable: GDP
HC (−1)	4.491 *	−10.330 *	−2.513 **	−0.251	1.066 *
FDI (−1)	−0.068	0.069	0.019	0.012	−0.003
IM (−1)	−0.734	−1.044	−0.420	0.080	−0.381
EX (−1)	0.251	0.637	0.332	0.067	0.081

Note: *: significant at 10%; **: significant at 5%; ***: significant at 1%.

presents the results for the five BRICS countries.

For Brazil, FDI, with a one-period lag, shows a positive impact on human capital (b = 0.015, t = 2.213). However, human capital lagged by one-period has no impact on FDI (b = −3.534, t = −0.846), suggesting a unidirectional causality running from FDI to human capital in the one-period lag term. Moreover, human capital lagged by one period positively affects GDP (b = 4.491, t = 1.975), but GDP lagged by one period has no impact on human capital (b = −0.016, t = −0.751). It suggests the existence of unidirectional causality from human capital to GDP in the one-period lag term. In addition, we find no causality between human capital (b = 0.380, t = 0.317) and imports (b = 0.061, t = 1.558), and between human capital (b = −2.310, t = −1.513) and exports (b = −0.027, t = −0.715).

For Russia, human capital lagged by one period has a negative impact on GDP (b = −10.330, t = −1.734), but GDP lagged by one period has no impact on human capital (b = 0.0001, t = 0.027). It suggests that there is a unidirectional causality from human capital to GDP in the one-period lag term. Moreover, there is no causality between human capital (b = −0.037, t = −0.003) and FDI (b = 0.005, t = 1.650), between human capital (b = 2.868, t = 1.413) and imports (b = −0.005, t = −0.228), and between exports (b = 0.012, t = 0.697) and human capital (b = 3.837, t = 1.287).

For India, human capital with a one-period lag is negatively related to GDP (b = −2.513, t = −2.161). However, GDP lagged by one period has no impact on human capital (b = −0.047, t = −1.073), suggesting a unidirectional causality from human capital to GDP in the one-period lag term. In addition, we find no causality between human capital (b = 10.796, t = 1.561) and FDI (b = −0.0009, t = −0.137), between human capital (b = −1.332, t = −0.959) and imports (b = 0.110, t = 1.660), and between exports (b = −0.092, t = −1.385) and human capital (b = −0.625, t = −0.468).

For China, FDI lagged by one period has a positive impact on human capital (b = 0.023, t = 1.930), whereas human capital lagged by one period has a negative impact on FDI (b = −5.016, t = −3.260). It suggests that there is bidirectional causality between human capital and FDI in the one-period lag term. In addition, no causalities are found between human capital (b = −0.643, t = −0.525) and imports (b = −0.021, t = −0.654), between exports (b = 0.009, t = 0.228) and human capital (b = 0.027, t = 0.030), and between human capital (b = −0.251, t = −0.381) and GDP (b = −0.011, t = −0.279).

For South Africa, human capital with a one-period lag is negatively related to imports (b = −0.686, t = −1.761) and exports (b = −0.757, t = −2.080). However, imports and exports with a one-period lag have no impact on human capital (imports: b = 0.062, t = 0.272; exports: b = −0.061, t = −0.218), suggesting the existence of unidirectional causality from human capital to international trade in the one-period lag term. Moreover, human capital lagged by one period lag has a positive impact on GDP (b = 1.066, t = 1.946), but GDP has no impact on human capital (b = −0.049, t = −0.532), suggesting a unidirectional causality from human capital to GDP. In addition, there is no causality between human capital (b = −0.657, t = −0.096) and FDI (b = 0.005, t = 0.841).

The following observations emerge from a comparison of Brazil, Russia, India, China, and South Africa. First, FDI with a one-period lag has a greater positive impact on human capital in China than in Brazil. It indicates that the growth in FDI inflows leads to a greater accumulation of human capital in China. On the contrary, human capital with a one-period lag has a negative impact on FDI in China but has no impact on FDI in Brazil. As a result, there is a bidirectional causality between human capital and FDI in a one-period lag term in China but a unidirectional causality in Brazil. Second, human capital lagged by one period has a more positive impact on GDP in Brazil than in South Africa, while human capital lagged by one period has a more negative impact on GDP in Russia than in India. However, GDP lagged by one period has no impact on human capital across countries. This indicates that there is unidirectional causality from human capital to GDP in a one-period lag term in Brazil, Russia, India, and South Africa.

Table 6. Summary of the causality direction for BRICS countries.

	Dependent Variable: HC	Dependent Variable: FDI	Dependent Variable: IM	Dependent Variable: EX	Dependent Variable: GDP
	Causality direction
Brazil	FDI→HC	EX→FDI	-	-	HC→GDP
Russia	-	-	FDI→IM	IM→EX	HC→GDP
India	-	-	GDP→IM	FDI→EX	HC→GDP
China	FDI→HC	HC→FDI	-	GDP→EX	-
South Africa	-	-	HC→IM GDP→IM	HC→EX	HC→GDP

presents causality relationship.

5. Conclusions and Implications

This study, employing a VAR approach, meticulously examined causal relationships among human capital, FDI, imports, exports, and GDP across BRICS countries from 1990 to 2020. The overarching findings highlighted a discernible unidirectional causality from FDI to human capital and from human capital to exports across the BRICS nations.

Upon closer scrutiny of individual BRICS countries, nuanced patterns emerged. Notably, China exhibited a bidirectional causality between human capital and FDI, while Brazil demonstrated a unidirectional causality from FDI to human capital. Furthermore, a unidirectional causality from human capital to GDP was identified in Brazil, Russia, India, and South Africa. In South Africa, a unidirectional causality from human capital to international trade was observed.

• Policy Implications

These findings carry vital policy implications. Firstly, fostering FDI inflows can contribute significantly to human capital accumulation in BRICS countries, especially in Brazil and China. However, the negative impact of human capital on FDI in China underscores the need for policies aligning FDI and human capital strategies, ensuring maximum societal benefits. Secondly, the positive unidirectional causality from human capital to economic growth in Brazil and South Africa underscores the pivotal role of human capital in stimulating sustained economic growth. Redirecting investments, particularly in education, is recommended. Conversely, the negative relationship between human capital and economic growth in Russia and India necessitates complementary initiatives addressing institutional factors and regional development. Thirdly, the observed negative impact of human capital on both imports and exports in South Africa emphasizes the need for measures to enhance technological progress. Encouraging competition, stimulating exports, and promoting technology adoption can enhance competitiveness and economic growth.

• Broader Significance

This study underscores the pivotal role of human capital in shaping economic dynamics. Beyond immediate economic impacts, the sustainable development of BRICS nations hinges on strategic investments in human capital. Policymakers are urged to prioritize sustained efforts to enhance education, skills, and knowledge, fostering an environment conducive to economic growth, FDI attraction, and international trade. Moreover, the identified unidirectional causality from human capital to exports highlights the importance of aligning human capital development with strategies that enhance global competitiveness and sustainability. Countries with a high level of human capital should leverage this advantage to contribute positively to the global economy, fostering sustainable trade practices and responsible economic growth.

In conclusion, this study not only provides valuable insights into the economic dynamics of BRICS nations but also emphasizes the critical link between human capital development and sustainability. As these countries continue to play a significant role in the global landscape, nurturing human capital emerges as a key driver for achieving sustainable economic development, aligning with broader goals of environmental, social, and economic sustainability. Recognizing the significance of up-to-date scientific information, we acknowledge the need to address this limitation in the next study by incorporating more recent research and ensuring a higher percentage of sources from the last five years (2019 onward). This adjustment aims to enhance the relevance and contemporaneity of our literature review, thereby reinforcing the robustness of our research in this rapidly evolving field.