Economic Journal of Emerging Markets Institutional factors, entrepreneurship capital types, and economic growth in Asian countries

This paper investigates the relationship between institutional factors, entrepreneurial types, and economic growth. The analysis is based on an unbalanced panel data of 18 Asian countries over 2006-2018 using a 3SLS estimation method. It extends the neoclassical growth model with entrepreneurship capital types as an endogenous variable to the economic growth function. Findings/Originality : The results show that new business density and productive entrepreneurship significantly affect GDP per capita. Additionally, a reverse impact of economic growth on entrepreneurship is revealed. Institution's constructs, namely corruption control and the rule of law, are crucial to entrepreneurship, which in turn stimulate economic growth. The results also confirm the significant role of human capital, accumulating domestic investment, economic openness, and controlling inflation in the economic growth model of the Asian countries.


Introduction
Entrepreneurship is a crucial engine of economic growth and other social aspects, playing a role as is a channel of transmission between institutions and economic development (Acs, Estrin, Mickiewicz, & Szerb, 2018;Baumol, 1990;Ivanović-Djukić, Lepojević, Stefanović, van Stel, & Petrović, 2018). The positive effect of entrepreneurial factors on economic growth is significant in Eastern or Western developed countries (Aparicio, Urbano, & Audretsch, 2016;Urbano, Audretsch, Aparicio, & Noguera, 2020). Meanwhile, entrepreneurial impacts on economic growth are vague, with mixed effects in developing and emerging economies (Hessels & van Stel, 2011). North (1990) provides a significant contribution, in theory, showing the importance of institutional factors in promoting growth, shaping the progressive intention of individuals in each society, and differences in the institution might lead to differences in national economic development. Institutional factors can influence economic growth through endogenous variables, including entrepreneurship and industrial development (Acemoglu, Gallego, & Robinson, 2014;Bjørnskov & Foss, 2016). The institutional factors are particularly useful in understanding how entrepreneurship is shaped and how entrepreneurs make decisions to improve economic development. However, new research questions about what institutional dimensions are conducive to entrepreneurship, which enhances economic growth and opens for new empirical studies (Urbano et al., 2020).
The mechanism of influence of the institutional environment needs to be clarified to understand why entrepreneurship types' impact on growth varies across regions and countries (Bjørnskov & Foss, 2016). David (2012) concludes that understanding the implications of the relationship between institutions, entrepreneurship, and economic growth can better encourage the dynamics of these sectors at the micro and macro levels, bringing about useful for strategic Sobel, 2008). In particular, informal institutional factors influence entrepreneurship stronger and more positively than formal institutional factors (Thornton, Ribeiro-Soriano, & Urbano, 2011).
Previous studies adopted the neo-classical economic growth theory to estimate the impact of entrepreneurship on economic growth, assumed in production decisions (Aparicio et al., 2016). Schumpeter (1934) asserts that entrepreneurship encourages the process of innovation and thereby impacts economic development. Empirical studies confirmed the presence and role of entrepreneurship in growth models of countries, regions, and industries (Bjørnskov & Foss, 2016;Bosma, Content, Sanders, & Stam, 2018). Other studies continue this approach, demonstrating the link between entrepreneurship and economic growth based on the endogenous growth theory (Noseleit, 2013). Audretsch, Bönte, and Keilbach (2008) adopt both neoclassical growth theory and endogenous growth theory, which have affirmed the importance of linking entrepreneurship with economic growth and the role of the institutional environment in which this relationship takes place. The institutional economic theory with formal institutional factors is adopted to explain the difference in the outcome of entrepreneurship impacts on economic growth across cultures (Aparicio et al., 2016;Baumol & Strom, 2007).
The study examines the relationship between formal institutions, entrepreneurship capital types, and the economic growth of Asian countries. The findings might contribute to the literature of development economics by extending the neoclassical growth model with the addition of new business density and productive entrepreneurship as endogenous variables to the economic growth function of countries in Asia. Entrepreneurship is a significant mechanism for contribution to economic growth through sufficient institution bases, namely corruption control and the rule of law.

Methods
The study approaches the unbalanced panel data set of 18 countries in Asia over the period 2006-2018, which is classified according to World Bank's per capita income criteria and the availability of entrepreneurial activities, including: 1. Lower middle-income countries (India, Indonesia, Philippines, and Vietnam); 2. Upper middle-income countries (China, Iran, Kazakhstan, Malaysia, Thailand, and Turkey); 3. High-income countries (Cyprus, Israel, Japan, Qatar, Saudi Arabia, Singapore, South Korea, and the United Arab Emirates). Detailed measurement variables and data sources are described in Table 1. Reflects perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as "capture" of the state by elites and private interests, Worldwide Governance Indicators (WGI), World Bank. (+/-) Agriculture (ARG) Agriculture, forestry, and fishing, value added (per cent of GDP), World Bank national accounts data, and OECD National Accounts data files.

(-)
Economic openness (XOG) Exports of goods and services (per cent of GDP), World Bank national accounts data, and OECD National Accounts data files.

Neoclassical Economic Growth Model Extension
Cobb and Douglas (1928) have built a production function, which is named Cobb -Douglas function: In which: represents the total production in an economy (GDP) in some years, is the capital of the economy, is labor in the economy.  is the partial elasticity of output with respect to capital and  is the partial elasticity of output with respect to capital. When + = 1 represents a constant rate of return on the scale, it implies that doubling capital and labor doubles output. When + < 1 shows the rate of return decreases with scale and when + > 1 shows the rate of return increases with scale. Solow (1956) and Swan (1956) input technological progress (A) into a production function, which reflects long-term economic growth in the form of: In which: represents total factor productivity (TFP). Increase two sides by the partial derivative with: In which: is the growth of output (GDP); is the growth of capital; is the growth of labour; is the increase in the total factor productivity.
Contributions of Solow (1956) and Swan (1956) with the development of the neoclassical growth model have led to a theoretical framework where economic growth is primarily explained by the accumulation of physical capital and labor. The technological progress sources are not explained by the neoclassical growth model, which is frequently known as the Solow-Swan residual or TFP. Jones and Romer (2010) identify two approaches in explaining the residuals in the neoclassical growth model. The first approach focuses on the role of human capital, suggesting that real GDP per capita growth is positively related to a country's initial reserves or investment in human capital. The endogenous growth theory explains the role of technological progress, knowledge accumulation in growth, with technology as an endogenous variable, a leading determinant of growth (Romer, 1990). However, the explanation level is only above 50 percent of the variation of the TFP growth, and the residual in the neoclassical growth model is still very large (Barro & Lee, 1993). Thus, new factors are added to the endogenous growth model, including ideas, institutions, population, and human capital. In particular, the difference in institutions could be the underlying source of significant differences in countries' growth rates, helping to allocate resources and results effectively (North, 1990;Robinson & Acemoglu, 2012). Entrepreneurial activities in a productive institutional environment provide a transmission mechanism from innovation and technological progress to economic growth (Acs et al., 2018).
Based on the theoretical background analyzed, this study proposes an expanded neoclassical growth model with the addition of entrepreneurship as equation 5. The function reflects the dynamic model: In which: = − −1 is the first-order difference of , a derivative for real GDP per capita growth (GDPGR). −1 : The logarithm of real GDP per capita is one term late, deriving the initial income level to control production capacity in the theory of neoclassical growth.
is a set of variables in the neoclassical growth model, including: Gross capital formation ( ) is the reserve of investment capital. The relation between investment and growth has been summarized in the growth theory of Domar (1946).
Entrepreneurial activities are assessed through quantity and quality criteria. The quantity of entrepreneurship is measured by the number of newly registered limited liability companies per 1,000 people aged 15-64 (Chambers & Munemo, 2019).
The quality of entrepreneurship is assessed by productive and unproductive entrepreneurial activities (Baumol, 1990). Productive entrepreneurship is measured by the TEA rate (Chowdhury et al., 2018). Unproductive entrepreneurship activities are measured by the necessity-driven TEA rate because there is no better career choice (Sobel, 2008).

Effect Model of Institutional Factors on Entrepreneurial Types
The link between entrepreneurship development and economic growth are reciprocal (Galindo & Méndez, 2014). Additionally, entrepreneurial development is influenced by formal and informal institutional factors (Acs et al., 2018;Aparicio et al., 2016;Berdiev & Saunoris, 2018;Chambers & Munemo, 2019;Fuentelsaz, González, & Maicas, 2019;Omri, 2020). Therefore, the equation examining the impact of institutional factors on entrepreneurial development is described as equations 6, 7, and 8: Where: is institutional factors. The institutional environment in each country is derived by six-global public governance indicators set by the World Bank, including political instability and absence of violence, control of corruption, regulatory quality, the rule of law, government effectiveness, and voice and accountability. These six indicators cover three dimensions that fully reflect a country's institutional quality (North, 1990). This set of indicators is compiled from the survey results of businesses, people, and experts in developed and developing countries with a scale of approximately -2.5 to 2.5 corresponding to the lowest and highest quality for each indicator.
: Controlled variable in the model is the GDP growth rate.

Econometric Model
Zellner and Theil (1962) proposed using the three-stage least squares (3SLS) method to apply to the system of simultaneous equations. The advantage of 3SLS is asymptotically more efficient since it takes into account the correlation among the errors of each of the simultaneous equations of interest (Wooldridge, 2010;Zellner & Theil, 1962). The method also adjusts the weighting matrix for potential heteroskedasticity of the errors by estimating the coefficients within a generalized least square (GLS) framework, an approach outlined by Wooldridge (2010). This method is particularly suitable when an equation is overly identified. In the first stage, each reduced form equation is estimated, and the predicted values of endogenous values are stored. These values then are substituted for endogenous variables and the estimated structural equation. However, in calculating residuals and standard errors, real endogenous values are used instead of predicted values. Once structural estimates have been obtained, we can use them to obtain hidden short-form estimates by resolving each endogenous variable through exogenous variables. Because 3SLS estimates take extreme identification limits, hidden short-form estimates derived from them are more effective than direct reduction estimates. Besides, the 3SLS method is more effective than the two-stage least squares (2SLS) method if the system of equations is adequately specified (Baltagi, 1998). Breusch and Pagan (1980) test of independence is applied to evaluate whether cross-equation disturbances are indeed correlated, which verifies the system estimation method's requirements. Besides, Hausman's test (1978) is adopted to determine if the system is adequately assigned and selects the appropriate estimation method between 3SLS and 2SLS. Hausman test method is performed with the null hypothesis of the differences in the regression coefficients among the 2SLS, and 3SLS estimation methods are not systematic. If this standard Hausman test rejects the null hypothesis that the conditional mean of the regressors' disturbances is zero, the applied researcher reports the 2SLS estimator. Otherwise, the researcher reports the 3SLS estimator, see (Hausman, 1978). Table 2 presents the data's descriptive statistics that give all the essential information about the primary and control variables we have used in our empirics analysis.    Gross capital formation increased by one percent, leading to growth from 0.006 percent to 0.008 percent ( < 0.001). New business density increased by one percent, leading to a growth of 0.003 percent ( < 0.01). Productive entrepreneurship increased by one percent, resulting in a growth of 0.004 percent ( < 0.01). Meanwhile, unproductive entrepreneurship harms growth, and this entrepreneurial activity does not seem to contribute to the growth of countries ( = 0.00003, > 0.1). An average total of years of schooling for the adult population increased by one percent may, in turn, lead to growth from 0.057 percent to 0.072 percent ( < 0.001). Economic openness increased by one percent, positively impacting economic growth, 0.0001 percent -0.0002 percent ( < 0.01). An inflation rate of one percent would increase the growth by 0.0009 percent to 0.001 percent ( < 0.01). The impact of maintaining the state budget balance on economic growth is not significant. Meanwhile, the role of agriculture in Asian countries is not revealed in this study.

Results and Discussion
The estimation results in Table 4 show the critical role of corruption control and the rule of law for types of entrepreneurship development in countries. Economic growth provides an essential foundation for developing entrepreneurship in countries. The effectiveness of corruption control increased by one unit, and new business density increased to 0.439 percent ( < 0.05), and the unproductive entrepreneurship increased to 0.938 percent < 0.01). The effectiveness of the rule of law increased by one unit, new business density increased to 0.995 percent ( < 0.001), productive entrepreneurship increased to 0.759 percent ( < 0.01), and the unproductive entrepreneurship increased to 0.952 percent ( < 0.01). Economic growth in countries is tending to support entrepreneurship activities in nations. If economic growth increased by one percent, new business density increased by 0.281 percent < 0.001), the productive entrepreneurship increased by 0.333 percent ( < 0.01). The paper finds that entrepreneurship capital types have an essential impact on economic growth, based on creating new jobs, expanding tax bases for governments (Bosma et al., 2018;Sobel, 2008). However, low-quality entrepreneurial activities are driven by the need only to create jobs for owners and not bring spillover benefits to society, negatively impacting economic growth (Baumol, 1990).
The impact and contribution of entrepreneurship to national economic growth depend on the type of entrepreneurship with economic development (Adusei, 2016;Antony, Klarl, & Lehmann, 2017). Numerous studies have demonstrated that the impact of entrepreneurial activities depends on each country's economic development and has a U-shaped relationship (Amorós, Fernández, & Tapia, 2012). The impact of entrepreneurship on economic growth is different across countries and geographies, depending on the political structure and stage of economic development, the capacity and effectiveness of laws and regulations. TEA tends to be the highest among resource-based economies, declines for higher economic development levels, and is negatively correlated with economic development, economic growth, economic freedom, and global competitiveness (Szabo & Herman, 2014). The impact of entrepreneurship development on economic growth is not well promoted in developing countries and even shows adverse effects (Sautet, 2013).
Resource-based growth economies are dominated by low value-added goods and services production. During this period, non-agricultural self-employment rates are high, creating no knowledge or innovation, and the impact of these factors on economic growth is limited (Acs et al., 2018). However, entrepreneurship still plays an essential role in the growth of these countries (Pogodaeva & Senchenko, 2017). There is a transition from self-employment to wage employment in efficiency-based countries because of the substitution between capital and labor generated during this period, which increases the benefits of wage employment and reduces the profits from selfemployment (Acs et al., 2018). Innovation-based economies experience a decline in production and an increase in service provision, and information technology development provides more opportunities for entrepreneurship (Jorgenson, 2001).
The findings emphasize the role of corruption control and the rule of law in entrepreneurship development in countries and have similarities with recent studies (Acs et al., 2018;Aparicio et al., 2016;Berdiev & Saunoris, 2018;Fuentelsaz et al., 2019). Baumol (1990) argues that when the institutional structure discourages creative entrepreneurial talent and encourages redistribution and seeking rents, economic growth may be lower than its potential. Acemoglu et al. (2014) consider economic institutions critical because they affect the economic incentive structure in society. The quality of good political institutions reduces corruption and increases the government's governance efficiency, ensures political stability, and enhances the democratic freedom of the people, thereby improving the efficiency of investment capital and the increase in human capital accumulation, which indicate growth (Samadi, 2019). The quality of good economic institutions also improves the business environment, secure property rights, and transparent and consistent policies, thereby reducing transaction costs and increasing the scale of production in the economy (Redford, 2020). The high quality of economic and political institutions might positively affect entrepreneurial activities' quality and inhibit the unproductive entrepreneurship types (Chambers & Munemo, 2019;Omri, 2020).
The lag of GDP per capita (-1) harms economic growth in all research models, supporting the hypothesis of a conditional convergence of per capita income in the long run (Barro & Lee, 1993). As a result, countries are moving toward a normal long-term average income level; lowincome countries likely grow faster than high-income countries. The findings also confirm the crucial role of capital and labor, economic openness, government budget balance, and inflation in the growth model among Asian economies.
The study has certain methodological limitations, including the size and nature of its sample and measurements. Further studies can cover other areas and different types of institutional factors, entrepreneurship capital types, and economic development outcomes.

Conclusion
Economic growth throughout history is only achieved by creating an appropriate institutional structure that leads to productivity-enhancing economic activities. The result reveals that entrepreneurial types significantly affect GDP per capita. Besides, economic growth has a reverse impact on entrepreneurship. The institutional aspect represented by corruption control and the rule of law plays a critical role in developing entrepreneurial types in Asian economies with an essential economic growth foundation. Macroeconomic intervention policies improve institutional quality, invest in human capital, accumulate domestic investment, promote comprehensive international economic integration, and control inflation, which plays a decisive role in promoting economic development.