The economy of Bangladesh is experiencing sustained economic growth. Thus, the purpose of this study is to analyze how human capital and material, or physical capital interacts to influence the economic growth of Bangladesh. As a response to analyzing mechanisms of economic growth from 2000 to 2019, the present paper conducts quantitative research into the link between human capital, material capital, and economic growth in Bangladesh with the use of a Vector Autoregression Model (VAR). The study discovered that both human capital and physical capital can contribute to economic growth in different time dimensions, but their respective efficiency varies. As a result of the research, recommendations are made for policy development.
Bangladesh is considered a market economy in the process of developing. The Bangladeshi economy is one of the worlds fastest growing. Bangladesh averaged 4.5% GDP growth in the decade following 2004. According to the World Bank, Bangladeshs economy grew 8.3% in the first quarter of 2019 making it the worlds seventh fastest. The country has the 37th largest economy by nominal terms and is the 31st in purchasing power parity. Bangladesh is classified as a frontier market and a ‘Next Eleven emerging market economy (Absar et al., 2021). Exports of ready-made garments, remittances from abroad, and domestic agricultural growth were the major factors driving its growth. In terms of exports, the country mainly exports textiles, shipbuilding, fish, and seafood and leather goods.
The pharmaceutical industry, the steel industry, and the food industry are other self-sufficient industries. With high foreign investment in Bangladeshs telecommunications sector, it has grown rapidly throughout the years. The government is promoting Digital Bangladesh as a way to develop Bangladeshs information technology sector. Bangladesh, which ranks seventh among Asias gas producers, also holds substantial gas reserves. The Bay of Bengal is full of limestone, where a lot of exploration is taking place offshore (Ladd et al., 2019; Pomi, 2019). Bangladeshs seaports enable Nepal and Bhutan to reach the sea via the sea. As well as Bangladesh, Chinas land-locked southwest, which includes Tibet, Sichuan, and Yunnan, is considered a possible entry point from Bangladesh (Alam et al., 2021). Economic growth may be hampered by a lack of investment in infrastructure and human capital (Ahmed et al., 2020).
The economic development of Bangladesh has achieved remarkable results, but human capital and physical capital investments have yet to be coordinated. Numerous studies have been done on economic growth domestically and abroad, most of which use mathematical models to study the factors that influence growth. Is economic growth more likely to occur when human capital investments are coordinated with physical capital investments? Human capital and material capital have also assumed a prominent role at this time. Extensive economic development is no longer an effective model in recent years. Therefore, it is important to know the key roles that physical infrastructure, human capital, and financial development play in the context of Bangladesh in enabling economic growth. In this study, VAR models are used to quantitatively analyze economic growth in Bangladesh in a mathematical and physical way, thereby filling up academic gaps in this area. Using the panel data from 2000-2019 to conduct theoretical and empirical research and then provide policy recommendations to promote economic growth.
The World Bank created a human capital index to quantify the economic impact of health and edu-cation. The relevance of human capital to economic growth has been emphasized by several relevant theories. By investing in education, health services, and other social services, the government is promoting the acquisition of human capital in developing countries. A skilled and educated work-force is a key component of economic growth, according to the human capital theory. It assumes that the formal education of the workforce determines their production capacity, that is, the higher the education, the higher the productivity. There are also several other growth theories, for example, endogenous theory, that emphasize the relative importance of endogenous and exogenous factors for economic growth. In addition to physical capital, countries around the world consider human capital as an independent factor of production, indispensable to achieve high levels of sustainable economic growth even during the pandemic (Absar et al., 2021; Ayittey et al., 2020; Dhar et al., 2021). By investing in health and education, rich human capital is created, which is essential for economic development. Academic analysis can be performed at two levels based on the current research results. Economic growth is a function of two different factors, one of which is human capital and the other of which is physical capital (Amin, 2020; Islam and Alam, 2019). Many studies have examined human capital and economic growth interrelationships and made a complete analysis of them currently. According to many other researchers, health, training, education, and social services mean greater skills and capabilities of the work force, which improves economic growth. In several studies, the positive contribution of human capital formation to economic growth was demonstrated (Islam and Khan, 2019). Developing the concept of human capital and spurring a major research boom in this field, Schultz, (1961) was the first to analyze the relationship between human capital and economic development in an academic setting. Uzawa, (1965) examined regional differences in economic growth through human capital. According to Lucas (1988), human capital has both an external and internal spillover effect on growth. According to Mankiw et al. (1992), human capital has always played an essential role in economic growth, and the importance of economic differences between regions was also discussed. Musila and Belassi, (2004) found that either directly or indirectly, education has a tremendous effect on economic growth. By analyzing data from different provinces in China, Hongyi and Huang, (2009) demonstrate that education and health are positively related to economic growth. Recent researches have already discovered the relationship between human capital, performance and sustainable economic growth (Chowdhury and Dhar, 2012; Dhar 2014, 2017, 2018, 2019, 2020; Dhar, 2019; Dhar, 2020; Dhar and Mutalib, 2020; Masruki et al., 2020; Aktar, 2021; Sarkar et al., 2018).
Among recent research, Roopchund, (2017) discovered a direct relationship between human capital and economic growth, and further investment should be made in developing human capital to improve economic growth. According to the result, Sarkar et al. (2021) and Ho, (2018) found that except ageing issue human capital has a positive and significant impact on economic growth, both in the short and long term. The study concluded that policymakers should focus on policies that will foster sustainable economic growth by enhancing human capital. Second, physical capital plays a role in economic growth. The ratio of material capital versus human capital is higher than the global average, according to Heckman, (2005). Model econometric methods are used to establish the vector auto-regression model, the long-term equilibrium model, and the short-term equilibrium model in the research of Sun and Dong, (2007). According to the result, human capital, material capital, and economic growth are in long-term equilibrium. Gong et al. (2012) found that the advancement of capital stock was the dominant driving force for economic growth. A regression analysis by Garzarelli and Limam, (2019) showed that physical capital was a leading factor in economic growth in the central region, with a contribution rate of 80%. In the absence of proper allocation, it will prevent material capital from functioning. The alternative is that it promotes economic growth and eradicates social inequality if it is allocated properly. In the results, Xu and Xie, (2018) indicated that physical capital continues to play a key role in economic growth and its contribution rate is as high as 4/5. Human capital quality, however, cannot be ignored when the stock of human capital is decreasing. Previous studies show a positive relation among exports, imports, capital formation and economic growth while the study of Shah et al. (2020) shows that in the long run capital formation and economic growth has no effect. In the existing literature, most scholars research focuses on national issues, but there are relatively few studies focused on specific countries. Further, the current literature review focuses on the development of a single-equation theoretical framework for research, but the research results cannot explain the interaction between two variables, but only the influence of one sequence on another. Based on the above problems, physical capital and human capital data are used to develop an analysis frame-work for economic growth in Bangladesh. Using data ranging from 2000 to 2019, we have developed an autoregressive vector model (VAR) to analyze the relationship between human capital, physical capital, and economic growth in Bangladesh.
Source of Data
A number of factors influence human capital, including education and medical care, but education plays an important role. In order to measure human capital, university students are selected as an indicator. Physical capital refers to the actual total amount of fixed capital input, while GDP per capita represents real economic growth. Empirical research is founded on a VAR model using these three variables. In view of the availability and validity of the data, the data used in Bangladesh from 2000 to 2019 are from the official website of the National Bureau of Statistics and the Bangladesh Statistical Yearbook.
Theoretical model
Stationary test - Statistically, a spurious correlation or relationship is an association between two or more events, but one that is not causally related because of either coincidence or the presence of another unknown factor (Ali et al., 2018). Using time series data models that are non-stationary can cause an erroneous test result because spurious relationships may appear (Mele and Magazzino, 2021). Augmented Dickey - Fuller (ADF) test measures the stability of our original variables to reduce the appearance of "spurious relationships". The test formula is as follows:
ΔY_t=a_1+a_2t+(p-1) Y_(t-1)+ε_t
A unit root sequence has been determined to be unstable if H0: ρ = 1, Y represents the standard error, and ε_t is standard error. When calculating the F statistic in the sample data, if the F value is higher than the critical value, the empirical conclusion negates the null hypothesis, indicating that the original series Y is a stationary series, otherwise it means that the original series is a non-stationary series.
Granger causality test
Granger causality test can empirically test the direction of influence between variables (Chang et al., 2014). The following formula shows the general Granger causality test formula. In the following formula, t represents time; k represents the maximum lag of the variable, and represents the random disturbance term in the model. Based on this model, the Granger causality test model of Y, lnX1, lnX2, lnX3 and lnX4 is established (Pradhan et al., 2018).
ln〖Y_t 〗=A_10+∑_(i=1)^k▒〖A_Ii ln〖Y_(t-1) 〗 〗+∑_(i=1)^k▒〖B_Ii ln〖X_(t-1)+ε_1t 〗 〗
ln〖X_t 〗=A_20+∑_(i=1)^k▒〖A_2i ln〖X_(t-1) 〗 〗+∑_(i=1)^k▒〖B_2i ln〖Y_(t-1)+ε_1t 〗 〗
VAR model
The VAR model embeds the variables to be studied and determines its lag value to analyze the dynamic relationship between the variables (Dungey & Pagan, 2000; Dinh, 2020). This model can reflect the impact of univariate changes on other variables in the system, so as to accurately describe the relationship between variables (Giordano et al., 2007). The VAR model formula is as follows:
Y_t=Φ_1 Y_(t-1)+⋯Φ_p Y_(t-p)+Dx_t+ε_t
Here, t=1, 2,..., T, Y_tare the k-dimensional endogenous column vector,x_t is the n-dimensional exogenous variable, p is the lag order, and Φ_1,Φ_(2,) Φ_3,…..,Φ_pand D need to be estimated The coefficient matrix, εt is k as the disturbance term. Before establishing a VAR analysis, the best lag order should be specified using the AIC and SC rules, and stability analysis should be performed again to determine whether it is suitable to construct a VAR model. Suppose there is a VAR (2) model:
Xt =A1Xt-1 + A2Xt-2 + B1Yt-1 + B2Yt-2+ε_1t
Yt=C1Xt-1 + C2X2-1 + D1YYt-1 + D2Yt-2 + ε_1t(t=1, 2… T)
The impulse response analysis is mainly realized by the iterative method. When ε10=1 and ε20=0, X0, X1, X2… and Y0, Y1, Y2… and Y0, Y1, Y2,… can be obtained through iteration. The above data are all X caused by X shock And the response sequence of Y. At the same time, when ε10=0 and ε20=1, the corresponding shock reflection function can be obtained.
Stationary test
To reduce the effect of random error over the empirical regression to some extent, you should first logarithmize the data, in order to avoid the interference of potential heteroscedasticity from the objective factors. Because non-stationary data modeling may result in spurious relationships that can lead to incorrect test conclusions, HC, PC, and GDP are recorded as LNHC, LNPC, and LNGDP. The Augmented Dickey - Fuller (ADF) test is used to measure the stability of the original variables in order to minimize the occurrence of spurious relationships. In the article, empirical research is conducted using Eviews 9.0 software. The test formula is as follows:
Yt = 1 + 2t + (-1) Yt – 1 + t
Here, ε_t is standard error, the original hypothesis H0: ρ=1, which means that the sequence Y has a unit root, and the conclusion obtained is that it is unstable. By calculating the F statistic in the sample data, if the F value is higher than the critical value, then the empirical conclusion indicates that the original sequence Y is a stationary process, otherwise, it indicates that the original sequence is a non-stationary process. The three sets of data were tested for stationary. Table 1 shows the results of the tests:
Table 1: Stationary test.
Variable |
Inspection (C, T, K) |
ADF Statistics |
Critical value |
P value |
Result |
|
1% |
5% |
|||||
lnHC |
(C, T,5) |
-4.780495* |
-4.309395 |
-3.921156 |
0.0102 |
Stationary |
InPC |
(C, T, 6) |
5.462764** |
-4.345414 |
-3.203362 |
0.0004 |
Stationary |
lnGDP |
(C, T, 3) |
-0.997401 |
-3.145771 |
-2.038981 |
0.8876 |
Not Stationary |
DlnGDP |
(C, T, 0) |
-3.896101** |
-3.932267 |
-2.896229 |
0.0249 |
Stationary |
Note: ** and * respectively mean to negate the original conjecture and reflect stability to the extent of 1% and 5%. Among them, C, T, K represents the intercept part, trend part and lag order part of the ADF test, respectively. The lag order is determined by the AIC, SC, and HQ criteria. As shown in Table 1, the LNGDP obtained after the logarithm analysis of the variables negates the original guess by 5%, indicating that the original sequence is not stable, but the stable sequence is obtained after the first-order difference processing. Additionally, LNHC and LN-PC are stationary series. The original sequence data cannot be used in co-integration studies since they are not a single integration. VAR models are instead constructed using the Granger causality test.
Determine the optimal lag order
The FPE, AIC, SC, and HQ criteria indicate that the optimal lag order is 5, while the LR indicates that the optimal lag order is 4. Using the principle of deter-mining the rule at most, this paper establishes the VAR (5) model. Grangers method is applied to the VAR (5) model to analyze the relationship between the sequences to investigate whether there is a causal connection between them.
Table 2: Optimal Lag Order.
Lag |
LogL |
LR |
FPE |
AIC |
SC |
HQ |
0 |
80.05128 |
NA |
7.34E-09 |
-6.028361 |
-6.632147 |
-6.63204 |
1 |
182.4793 |
207.4177 |
1.01E-10 |
-12.18065 |
-12.33921 |
-13.55793 |
2 |
224.1720 |
34.16633 |
2.90E-11 |
-14.19696 |
-15.72363 |
-16.12214 |
3 |
238.0937 |
23.80382 |
1.84E-11 |
-17.63115 |
-16.01064 |
-17.71543 |
4 |
240.6002 |
22.42683* |
4.27E-12 |
-18.01712 |
-16.10587 |
-18. 61843 |
5 |
247.2043 |
15.1703 |
2.51e-12* |
-19.77660* |
-17.73336* |
-19.85022* |
Granger causality test
Table 3 shows that the Granger causality test concludes that human capital and physical capital are both two-way Granger factors for economic growth, and human capital is also a two-way Granger factor for physical capital, which can be further tested for stability.
Table 3: Granger causality test.
Null hypothesis |
Observations |
Chi-sq |
P Value |
Relationship |
LNHC ≠ LNPC |
20 |
28.02033 |
0.0045 |
No |
LNHC ≠ DLNGDP |
20 |
20.27531 |
0.0005 |
No |
DLNGDP ≠ LNHC |
20 |
16.97485 |
0.0067 |
No |
LNPC ≠ LNGDP |
20 |
15.58564 |
0.0041 |
No |
LNPC ≠ LNHC |
20 |
31.67854 |
0.0034 |
No |
DLNGDP ≠ LNPC |
20 |
19.67854 |
0.0043 |
No |
This paper uses the vector auto regression (VAR) model to analyze the effect of human capital and material capital on regional economic growth, and locks the scope to Bangladesh. Based on an analysis of Bangladesh data from 2000 to 2019, it is found that physical capital investment has a negative impact on economic growth in the short run, but in the long run, these two factors promote economic development and are the leading factors behind economic development in Bangladesh. It is imperative to organically combine the two to achieve the best possible economic growth in Bangladesh. Mathematical research shows that human capital investment has a relatively long return period compared to physical capital investment, but its return rate is often much higher. Finally, human capital has a very strong positive externality in comparison to physical capital investment. Investing in human capital is more beneficial in the long run. Bangladesh still lags behind other developed countries in terms of education level. Rural development must be accelerated, the training level of labor in underdeveloped countries must be improved, and the total amount of human capital must be improved across the board. Invest in education and higher education as part of a planned change in direction and proportion. It is absolutely necessary to invest in physical capital in the short term, such as new infrastructure and innovative technologies. Promote the integration of technology and industry development, as well as developing emer-ging industries vigorously. The policy trap we need to avoid is emphasizing one part of the input while under-estimating the other.
Many thanks to the co-authors supported with proper assistance and help for data analysis to conduct successful research work.
The authors have no conflicts of interest in publishing this research study.
Academic Editor
Dr. Liiza Gie, Head of the Department, Human Resources Management, Cape Peninsula University of Technology, Cape Town, South Africa.
YWCA, Chittagong, Bangladesh.
Pomi SS, Sarkar SM, and Dhar BK. (2021). Human or physical capital, which influences sustainable economic growth most? A study on Bangladesh, Can. J. Bus. Inf. Stud., 3(5), 101-108. https://doi.org/10.34104/cjbis.021.01010108