“Role of institutional quality in economic development: A case study of Asian countries”

The paper investigates the impact of institutional quality on economic growth by tak- ing 48 countries in Asia between 2005 and 2018. By using the quantile regression methods with panel data, institutional quality is found to be a key factor of economic de- velopment. However, in the lower-income Asian countries, the institution with better quality appears to promote the growth more effectively than in the higher-income ones. Moreover, the paper also finds out a nonlinear relationship between institutions and economic growth. The results show that there is an institutional threshold for econom- ic growth to reach its highest level. If the institution indicator exceeds the threshold, it causes the reverse effect on the growth. Moreover, the economic growth of Asian coun- tries is also affected by inflation (INF), labor force (LABO), trade openness (OPEN), and infrastructure (TELE). From that, the study suggests some policy implications for Asian countries and Vietnam, in particular, in order to improve institutions contributing to economic growth.


INTRODUCTION
Economic growth is understood as an increase in the output (or income) of an economy in a certain period (Dung & Phuong, 2009). In the domestic and international socio-economic context with certain advantages and challenges, economic growth is an important goal of each country's macroeconomic policy. The economic growth theory, therefore, aims to explain long-term economic growth and reveal a slight impact of institutions on economic growth, the biggest disadvantage of traditional growth theories . New institutional economics has been studied as a theory of economic growth. North (2020) commented that factors such as expanding markets, improving technology, and increasing human investment are all partly contributing to productivity, and a social institutional framework helps to create incentives, directly influencing economic and political activities and foundations for economic development. Bernabe (2017) studied sustainable economic development goals in Asian countries and pointed out that one of the economic development concerns is institutional and legal quality.
Aware of the significance of institutional improvement, leaders of the Ministry of Finance regularly pay attention and engage series of activities for institutional improvement in Vietnam, which consolidates to the national economic development targets in 2016-2020 . Recognizing the national economic development goals and the impor-tant role of the institution in relation to economic growth. In recent years there are many domestic and foreign scientific studies in this field. Nakabashi Ngo and Nguyen (2020) concluded that institutions have positive impacts on economic growth. However, Dias and Tebaldi (2012) and Acemoglu et al. (2015) did not find a relationship between institutions and economic growth. Barro (1996) found a non-linear relationship between democracy and economic growth in which democracy can promote and negatively affect economic growth if it crosses a certain threshold. Law et al. (2013) supposed whether there is an institutional threshold for optimum economic growth, which is concerned by further studies. Marakbi and Turcu (2016), Ndjokou and Tsopmo (2017), and Ouoba and Sawadogo (2019) studied the role of institutions and concluded that there exists a threshold of institutions affecting economic growth while Chong (2020) found a non-linear relationship between institutional quality and economic growth. Thus, there is not complete research on the impact of institutions on economic growth, especially in Vietnam where institution related topic is relatively new.
Economic development management plays the primary role in any governmental commissions of countries and institutions, because it is one of the key factors influencing the growth. Therefore, it cannot be denied that studying the influence of institutional quality on economic growth is the vital target of all countries in the current context with the COVID-19 pandemic affecting the world's economic stability. The study results are the scientific basis to suggest some policy implications for improving the quality of institutions to promote economic growth in Asian countries including Vietnam.

LITERATURE REVIEW AND HYPOTHESES
According to North (1990), institutions are social rules of the game and their main role in a society is to reduce uncertainty by providing a structure for everyday life activities via human-human interaction guidelines such as greeting friends on streets, borrowing money, starting a company, etc. Without institutions, rules, or interaction guidelines, people can neither respond to one another nor follow their established agreements that allow individuals and legal entities to purchase, recruit people, and proceed with investments with the belief that contractual arrangements between parties are fulfilled (Kasper & Streit, 1999).
Recently, several studies have been done on the relationship between institutions and economic growth following new institutional economic theory (NIE), state theory, institutional change theory, property rights theory, transaction cost theory, and Porter's theory; new institutional economic theory is a theoretical model, built on the old institutional economics theory that emphasizes that institutions are important factors for economic growth (Hodgson, 2004). By examining the relationship between economic institutions, political institutions, trade openness, and per capita in-come with OLS estimation techniques, Rigobon and Rodrik (2005) showed that both democracy and compliance with the law are good for economic performance and strongly impact income in the long run. It was indicated that countries with higher incomes have better institutions. Gani (2011) studied the effect of public governance and institutional quality framework on economic growth with a sample of 84 developing countries and showed that political stability and efficiency of the government have positive and significant effects on growth. Aisen   In general, studies on the impacts of institutions on economic growth have not reached a common conclusion. The existence of an institutional threshold affecting economic growth and a non-linear relationship between institutions and economic growth has been identified. However, there are not many studies on institutional thresholds and it is not clear whether there exist institutional thresholds affecting economic growth. Moreover, the institutionally related topic is quite new in Vietnam and none of the complete studies on institutions uses the quantile regression method. Hence, there are two primary objectives of this study. Firstly, it is crucial to determine the impacts of institutions on the economic growth of 48 Asian countries. Additionally, the difference of institution effects on economic development is explored using the quantile regression method compared with standard regression methods of panel data. Secondly, a non-linear relationship between institutions and economic growth is examined to identify the existence of an institutional threshold affecting economic growth in Asia. Consequently, based on the aims of the study, two hypotheses are suggested as follows: H1: Economic growth is positively influenced by institutions.
H2: There are non-linear relationships between institutions and economic growth.

METHODOLOGY
To determine the impact of institutions on economic growth, the study uses some standard regression methods for panel data such as Pooled OLS, FEM, and REM regression, as well as carries out the Chow test and the Hausman test to choose the most appropriate estimation method. However, the outcomes revealed that there are models with heteroscedasticity and autocorrelation. Therefore, the FGLS method is executed, as proposed by Beck and Katz (1995), to improve the effectiveness of the estimation. In addition, the Breusch-Pagan test showed heterogeneous error variance of the FEM model is inconsistent, so the use of the quantile regression method, introduced by Koenker and Bassett (1978), to study the economic situation for several quantiles is suitable. Moreover, the sample consists of Asian countries with different levels of economic, financial, social as well as political development. Consequently, the quantile regression is the more suitable method to explore the different impacts of institutions on economic growth relating to several quantiles of economic performance in the sample. The results from quantile regression are compared with the FGLS method showing that whether the effect of institutional quality on quantiles of economic performance is different from the mean value or not.
Although quantile regression can be performed in detail from 0.01 to 0.99, basic quantiles of the ef-ficiency variable distribution function are chosen for analysis, including the quantiles of 0.1, 0.25, 0.5, and 0.75 due to the scope of study and relevant sample features.
The study further tests the non-linear relationship between institutions and economic growth by adding squared institutional variables to the model and using panel data regression methods. However, the results of the model defect tests have heteroscedasticity and autocorrelation. Therefore, general least squares regression (FGLS) for nonlines on the panel data is used, as proposed by Beck and Katz (1995).
Based on empirical studies and theories, study model factors include institutional quality, expenditure of the government, total domestic in-   Barro (1996) and Chong (2020) found a nonlinear relationship between institutional quality and economic growth. This means that institutional quality can promote economic growth until it gets a certain threshold, continuing to improve the quality of institutions might reduce economic growth. Therefore, an "inverted U-shaped" relationship between institutional quality and economic growth is found by Chong (2020

RESULTS
Firstly, after using Pooled, FEM, and REM methods to carry out regression data of 48 Asian countries in the period 2005-2018 with 672 observations and tests, the FGLS regression method is applied to improve the efficiency of the estimation. To show the relationship between institutions and economic growth, the study examines the impact of institutions at different quantiles of income per capita (GDP) and its results from five models corresponding to institutional variables of INS, CPIA_PUBS, CPIA_SOCI, CPIA_STRC, CPIA_ECON summarized in Table 2  impacts on economic growth. Particularly, negative impacts of the variable CPIA_STRC on GDP are found with a negative regression coefficient and statistically significant at the expected value, different from the hypothesis but consistent with Ngo and Nguyen (2020). However, it is noted that the level and direction of impact of these factors are different at different quantiles.
As summarized in Table 2, the effect of variable INS on economic growth has significant meaning at most of the quantiles, except for the quantile 0.5, which coefficient INS is not statistically significant while positive impacts of public and institutional management quality variables (CPIA_PUBS) on economic growth with significance level at only low quantiles of 0.1 and 0.25. The effects of CPIA_STRC and CPIA_ECON on the GDP are both positive and have statistically significant levels at a quantile of 0.1 while the opposite trend is found at the other higher quantiles. The impact of CPIA_SOCI on economic growth is the strongest because of its largest coefficient compared to other variables representing institutions affecting the GDP. It is indicated that the direction of the impact of social inclusion/equity policy quality on economic growth is positive and significant across all GDP quantiles, but its level decreases at higher quantiles. Therefore, the impact of institution quality on economic performance is not the same at different quantiles. In general, the impact of institutions is usually positive at low quantiles of the GDP while that is reversed when the GDP gets higher quantiles.
Secondly, the FGLS regression method is used to examine the non-relationship between institutional quality and economic growth. The results of regression models (6); (7); (8); (9) and (10) are presented in Table 3.

Table 2. Synthesis of results on the impact of institutions on economic growth in Asian countries
Source: Authors' elaboration.   (10). It is shown that there is a non-linear relationship between the institution and economic growth for institutional variables of INS, CPIA_ PUBS, CPIA_SOCI, and CPIA_STRC, implying the existence of a non-linear relationship between institutions and economic growth, consistent with Chong (2020). From that, the study determines the institutional threshold for maximizing average per capita GDP income based on the regres-sion coefficients that are statistically significant in models (6), (7), (8), and (9).

Expectations
The threshold of INS is found considering the first derivative of both sides with INS as 12 2.

ββ = +⋅
To find the highest value of the GDP, the first derivative of the GDP is 0. Therefore, the highest value of the GDP is Similarly, thresholds of variables CPIA_PUBS, CPIA_SOCI, and CPIA_STRC are also found by considering the first derivative of both sides with CPIA_PUBS, CPIA_SOCI, and CPIA_STRC: with the corresponding regression coefficients in Table 3, the thresholds of CPIA_PUBS, CPIA_SOCI, and CPIA_STRC are 3.428, 3.894, and 3.326 respectively for the highest GDP value. Coefficients of CPIA_ECON and CPIA_ECON 2 are found not statistically significant in model (10). It is concluded that there is no existence of an institutional threshold in relation to economic growth with the institution measured by the variable of CPIA_ECON. Therefore, Table 4 summarizes the results on non-linear relationships between institutions and economic growth in five models corresponding to five different institutional variables measured by CPIA indicators. The results show that there are non-linear relationships in four cases, in which institutional variables measured by INS, CPIA_ PUBS, CPIA_STRC, CPIA_SOCI indicators as well as the value of institutional thresholds are found corresponding to four cases. Table 3. Results of FGLS method for models (6), (7), (8), (9) and (10) Source: Authors' elaboration. Note: Standard errors in parentheses. *** means p < 0.01; ** means p < 0.05; and * means p < 0.1.

DISCUSSION
The results shown in Table 2 indicate that economic growth in Asia is affected by institutions as all of regression coefficients in the the models are statistically significant. However, only institutional quality about structural policies has negative impact on economic development while the others support the first hypothesis (H1). These inventions prove that not all aspects of institutions always promote economic growth in Asia. On the other hand, the results in Table 3 agree with the second hypothesis (H2), except the case of institutional quality measured by CPIA_ECON variable.
There is no evidence of a nonlinear relationship between institutions about economic management and economic growth in Asia because the regression coefficients of both CPIA_ECON and CPIA_ECON2 variables in model (10) are not statistically significant.
As shown in Figure Table 5. Consequently, the results show that Vietnam needs to improve institutional quality in terms of social inclusion/equity as well as public and institutional management policies to promote economic growth because variables of CPIA_PUBS and CPIA_SOCI are statistically significant and positive (> 0). Coefficients of the CPIA_ECON variable as well as CPIA_STRC variable are not statistically significant at the quantile of 0.25 following the regression results of the impacts of institutions on Asian economic growth at the different distribution of economic performance.
CPIA scores of Vietnam in 2018 in Table 6 reveals that the score CPIA_PUBS of 3.50 exceeds the threshold of CPIA_PUBS (3.428) and the score of CPIA_SOCI of Vietnam in 2018 at 3.70 is lower than the CPIA_SOCI threshold of 3.894.

Table 4. Institutional thresholds affecting economic growth in Asia
Source: Authors' elaboration.
Source: Authors' elaboration. Regarding improvement on institutions, Vietnam needs to continue to enhance social inclusion/ equity policy to reach the optimal threshold of CPIA_SOCI and maintain CPIA_PUBS at the threshold of 3.428 for maximum growth. It is noted that when public and institutional management CPIA_PUBS score exceeds the optimal threshold, institutions negatively affect economic growth, meaning that the improvement of institutional policy does not promote economic development. Vietnam's overall CPIA of 3.57, measured by INS variable, which is lower than the INS threshold of 3.62 found in the study, implying that Vietnam needs to improve appropriately its institutional quality in all respects to upgrade the overall av-erage CPIA score and remain at the threshold for maximum economic growth. Therefore, the study proposes several solutions to help improve the quality of institutions in Vietnam and contribute to economic growth relating to social inclusion/ equity as follows: Firstly, gender equality in Vietnam should be addressed in the process of improving social inclusion/equity policies. The whole society needs to realize gender equality and Vietnamese women empowerment are objective demands for the whole country's development, in line with the requirements and responsibilities as a United Nations member to the international and regional com-  munity. From that, Vietnamese women need to practice self-confidence to overcome gender stereotypes and upgrade to self-assertion. Secondly, Vietman is suggested to effectively manage budget by a clear financial mechanism so that it can intervene in time and avoid situations beyond control.
Finally, Vietman should issue policies to support initial investment, encourage effective investment in public services with various forms to increase capital mobilization, and provide favorable conditions for non-public establishments following the socio-oriented development.

CONCLUSION
Data was collected from 48 Asian countries between 2005 and 2018. The study investigates the relationship between institutions and economic growth to determine the factors that affect the expected level of economic growth using panel data regression (FGLS). The quantile regression method is applied for further analysis of the institutional impacts on economic growth at the different distribution of quantiles. The results show that expectation of economic growth is influenced by institutions via five different measurable variables of the CPIA indicator set. It is indicated that the level and direction of impacts of a single variable are different across various quantiles in which the degree of institutional impacts on economic growth decreases or reverses at higher percentiles while it is higher at countries with low quantiles of 0.1 and 0.25, and its direction is always positive at 0.1 percentile.
In addition, the study examines the non-linear relationship between institutions and economic growth, thereby identifies the institutional thresholds affecting economic growth in Asia. The results indicate that there is an institutional threshold for economic growth to reach its highest level and exceeding it causes the switching impacts on economic growth. Vietnam has a growth rate of 0.25 quantile compared to 48 Asian countries during the study period, so Vietnam must focus on improving the quality of institutions in the field of social inclusion/equity as well as public sector management and institutions policies to promote economic development. With CPIA scores in 2018, Vietnam has not reached the optimal threshold at CPIA_SOCI or INS (overall CPIA) scores yet but exceeded the optimal threshold of the CPIA_PUBS index.
The results of the quantile regression method show that there is a strong classification of the level of institutional impact on single groups of countries subject to its relevant quantile of economic growth. Therefore, to achieve the desired growth rate of the country, policymakers should focus on the overall and individual impacts of the institutional quality on economic growth relevant to its economic performance so that appropriate improvements on institutional aspects, goals, and volumes are given to promote the country optimal economic growth. In the case of Vietnam, the government needs to maintain the quality of institutions in two aspects including (i) public sector management and institutions; and (ii) social inclusion/equity policy at the optimal threshold. Moreover, the Vietnamese government should keep upgrading the total average CPIA score as well as maintain at the threshold level so that maximum economic growth can be reached.