Abstract
The empirical limits of this study lie in investigating the nexus between trade openness, institutional quality and economic growth in emerging BRICS countries. Having nascent institutions, the study aims to analyse their role in determining the trade–growth relationship. Using endogeneity expunging GMM technique, the empirical evidence shows—first, imports play a crucial part in augmenting growth. While exports and overall trade openness do show a positive impact, they lack the statistical significance. Second, institutional infrastructure shows an indirect effect by augmenting the economic performance when complemented with higher imports. These countries stand to gain more from imports with improved institutional infrastructure.
Introduction
The proximate determinants of growth have been a bone of contention in economic literature. While neoclassicals suggested that growth is determined by capital accumulation and technology, the modern economists criticised and called it a narrower view. They postulate that growth may not be independently determined by factor accumulation (Acemoglu & Robinson, 2010). This has shifted the focus of economists to analyse the role of non-economic determinants like institutional quality in accelerating economic growth (Acemoglu & Johnson, 2005; Aron, 2000; Djankov et al., 2006; Glaeser et al., 2004; Henisz, 2000; Knack & Keefer, 1995; Scully, 1988). According to Farole et al. (2011), the countries are witnessing differential economic trajectory in accordance to their economic, political and social institutional quality. In this context, it is important to understand that the institutional framework is not a substitute to growth determinants rather an adopted complement.
Another parallel strand of literature has grown documenting the role of trade openness in economic prosperity—Balassa (1986); Burange et al. (2019); Dollar (1992); Frankel and Romer (1999); Makun (2017); and Redding (1999), to name a few. They postulated that trade leads to static gains through efficient reallocation of resources. Moreover, it ushers dynamic advantages by expanding market for domestic products, allows learning by doing and knowledge transmission, enhances productivity and brings in healthy competition (Coe & Helpman, 1995; Kraay, 1999; Lee, 1995). Exposure to foreign externalities also enhances the performance of non-export sectors, accelerating overall economic growth (Dollar, 1992; Froning, 2000; Lucas, 1988; Romer, 1986; and Stokey, 1988). Increasing competition by trade liberalisation also decreases the losses of deadweight incurred by domestic monopolies and oligopolies, and thereby bringing additional gains (Krishna & Mitra, 1998).
While, the predominant message suggests positive impact of trade on growth, many studies contradict (see Jebran et al., 2018; Redding, 1999; Rodriguez & Rodrik, 2000; Sheikh et al., 2020). The studies such as Aghion et al. (2005), Borrmann et al. (2006), Dollar and Kraay (2003), Freund and Bolaky (2008), Rodrik et al. (2004), and Winters (2004), have recognised institutions as a deeper determinant of economic growth and development while trade as a facilitator. This raises a fundamental question which remains largely undertested—what happens to the trade–growth relationship in developing countries with only nascent institutional structure?
It is in this backdrop, this study complements the existing literature by empirically analysing the impact of trade openness and institutional quality on economic growth in BRICS economies. BRICS provides a compelling case to study this aspect in some detail, because over the years the economic reforms in BRICS have led to the monumental change in their outlook. These countries have substantially integrated with the global economy and are no longer closed economies. Consequently, the external sector has undergone a metamorphosis due to escalating competition, paradigm shifts in policy and operational environment. In somewhat similar vein, the institutional structure of these countries has also undergone a huge change during the last three decades. Despite their progression at a fast pace, these countries are grappling with low technology and poor institutional quality.
The study proceeds with the next section discussing the data which is followed by model specification. The third section presents the analysis and interpretation and the last section provides the concluding remark.
Data and Research Methodology
The data for gross domestic product, gross capital formation as a percentage of GDP, inflation and trade openness denoted by yit, GCFit, INFit and TOit, respectively, in Equations 1 and 2, is extracted from the World Bank database. Data for institutional quality index IQit was drawn from Heritage Foundation. Our two baseline models are:
and
where yit, GCFit, INFit and TOit are same as defined above. μit (ui + ε it ) is the error term combining the individual country specific effect and omitted variable effect. β1, β2, β3 … … β6 represents the effect of their respective corresponding variables on growth when all other variables are zero. Equation (2) considers the mediating effect of institutional quality in trade–growth relationship captured through an interaction term of trade openness and institutional quality (INTER).
We applied endogeneity expunging technique called GMM developed by Arellano and Bover (1995), and Blundell and Bond (1998) on an annual data for the period of 1996–2017. GMM is preferred for its various statistical properties as it can leverage the weakness, where fixed and random effect models produce biased results.
Analysis and Interpretation
We began our analysis with cross-sectional dependence test indicating if error terms are cross correlated or not. Accordingly, we applied Pesaran’s (2004) cross-sectional dependence (CD) test. The results given in Table 1, does not reject the null of uncorrelated variables across countries. This indicates cross border effect of domestic events which reflects the interconnectedness of BRICS countries.
Cross-sectional Dependence Test.
This is followed by verification of stationarity of data through unit root testing. Precisely, it means a flat looking series with constant variance, without trend and no periodic fluctuations (Enders, 2008). Accordingly, we applied Maddala and Wu (MW) (1999), and Pesaran’s (2007) panel unit root tests both at levels as well as at first difference. The results established in Table 2 indicates that the data is not stationary at levels I(0) as the p-values stood above the threshold of 0.05 except for lngc which is stationary at levels. The confirmation of first order integration of variables comes from the estimated statistic values and their associated p-values in Table 3. The p-values given in the parenthesis are significantly below the threshold p-value of 0.05, thereby rejecting the null hypothesis of panel containing unit-roots.
Panel Unit Root Tests at Levels.
Panel Unit Root Tests at Difference.
GMM Estimation Output
After the pre-testing procedures, the results fetched by the baseline models employed for the present study are necessarily to be discussed vis-à-vis economic propositions and past literature. Therefore, this section is committed to comprehensively produce the results generated through dynamic panel GMM approach using logarithm of GDP as a dependent variable. The regression coefficients, standard errors and p-values respective to each explanatory variable regressed against the log of GDP are tabulated in Table 4.
GMM Estimation Output.
Table 4 encapsulates the statistical values for different parameters which are instrumental in investigating the nature, direction and magnitude of relationships between trade, institutions and economic growth. The statistics indicating the nature of relationship between the log GDP and five explanatory variables assumed to have important bearing are briefly discussed below.
As established by Table 4, the first column shows the independent impact of trade openness (measured by trade intensity) and institutional quality on economic growth. The contingency of trade–growth relationship is captured by an interaction of trade and institutional quality index duly reported in the second column.
Beginning with the lagged GDP, whose significance validates the dynamic GMM as an appropriate estimator for statistical inference. Previous growth has an important bearing as indicated by the coefficient values of yit–1 established in Table 4, which explains the major chunk of current economic growth. The estimated coefficients of capital formation (lngcf), in all specification (first to sixth column) suggests positive and significant effect on economic growth. Historically, BRICS countries have been capital scarce but labour surplus countries. Any increase in capital per worker reflects significantly in economic performance.
Contrary to this, FDI and government expenditure are shown to be negatively associated with economic growth and significant at conventional levels of significance. This phenomenon is alternatively explained in the previous literature. In developing countries, FDI is said to be either crowding out the domestic investment or does not have its independent effect on economic growth; rather it augments trade (Adams, 2009). Moreover, exploitation of the spillover effects of FDI requires appropriate institutional infrastructure, which is lacking as reflected by insignificant coefficient of IQit in Table 4. Similarly, for government expenditure, though many associate it with higher growth, yet dominant is the message for it being unproductive and therefore detrimental to long term growth. Our results indicate the prevalence of unproductive government spending and lends support to the findings of Easterly and Rebelo (1993). To our surprise is the coefficient of (lninf) showing significantly positive correlation with economic growth. Khan and Ssnhadji (2001) estimated threshold level of inflation for a panel of 140 countries over the period of 1960–1998 and argued that in developing countries, inflation rate beyond 11%–12% exerts negative impact on growth. This finding is plausible, because BRICS has only witnessed mild inflation (with occasional spikes in Brazil and Russia in early 1990s), which encourages the production than adversely affecting it. Fundamental to the analysis of this study is the coefficient pertaining to the institutional quality (lninq) and trade openness (lnto). lninq shows a positive but statistically insignificant impact on economic growth. The results indicating the importance of quality institutions is both plausible and sensible. Such outcome may be due to any or both of the following reasons. First, BRICS economies are only emerging countries where the institutional infrastructure is in its nascent stage. Developing institutional infrastructure may not be able to augment growth significantly. Further, as argued by many studies, the indirect effect of institutional infrastructure on economic growth is stronger than its direct effect. The results pertaining to the interaction of trade openness with institutional quality in the second column supports the later view. Similarly, trade openness (lnto) is positive but insignificant in first column but turns significant with the inclusion of interaction term in second column. While trade openness and institutional quality show positive impact, their interaction (Lnto*lniq) indicates otherwise. The significant and negative coefficient of interaction term indicate that along with the development of institutional setup, the trade-led growth will lessen. As suggested by the previous literature, short term slowdown in growth rate during transitionary phase occurs due to the strengthening of competition by institutional improvements (Nguyen et al., 2018). Since BRICS have large underutilised capacity, which invites considerable foreign competition through trade openness, any improvements in institutional infrastructure might temporarily dent economic growth rate. The established results also confirm the indirect influence of institutional quality on economic growth via trade openness. Our results reinforce the findings of Baliamoune and Ndikumana (2007).
The coefficient of import intensity ratio in third column is both positive and significant at 10% level of significance. This indicates that imports are growth augmenting rather than growth retarding as claimed by mercantilists. These results lend support to the previous studies arguing developing countries benefit themselves through importation of intermediate goods which exposes them to new technologies unavailable locally. Since BRICS countries heavily import intermediate goods, the results are indicative of the technology spillover effect. These results are in line with Malik et al. (2021). The inclusion of interaction term (Lnimp*lniq) in fourth column raises considerably the significance and magnitude of the coefficient of imports (Lnimp). The significant and positive coefficient of interaction term indicate that along with the development of institutional setup the technology spillovers will be larger. These results are plausible because improved institutions nurture the absorption capacity of developing countries, thereby enabling them to exploit the technological spillovers to the maximum extent.
Much likely to that of trade intensity, the estimated coefficient of export intensity in fifth column also portrays similar effects. Exports though having positive growth effects lack statistical significance and remain unchanged with the inclusion of interaction term (Lnexp*lniq) in sixth column. Though the previous literature suggests varied set of reasons, the ‘composition of export’ seems more plausible. Many studies like Nnebe (1994) have stressed that it is not just exports but the composition of export basket which matters for growth. Since all of BRICS except China export mostly primary and low technology goods, the results seem to be driven by low technology export basket.
Conclusion
This study investigates the impact of trade openness and institutional quality on economic growth for a panel of five emerging countries over a period of 22 years from 1996–2017. Using dynamic panel GMM model and taking different measures of trade openness, the results are suggestive of the positive effects of both trade openness as well as institutional quality. Further, we attempted to analyse if trade–growth relationship is contingent upon the level of institutional quality? To this, though the outcome produced certain evidences, the effects are largely negative except in case of imports. Overall, the outcome of the study suggests that further integration with global economy could be decisive to enhance economic growth. Any improvements in institutional quality should be made in consideration of their impact on domestic firms. A carefully designed strategy of institutional improvement should reduce the vulnerability of domestic firms to external competition.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.