An exploratory analysis of economic growth and governance indicators of Bangladesh, 1996-2018

Abstract

Bangladesh has been experiencing a steady economic growth around seven percent for the last few years. Although the economy is growing faster but the governance indicators are showing a different scenario in six dimensions of governance: voice and accountability, political stability, government effectiveness, regulatory quality, rule of law and control of corruption. Thus, there has been a wrangle regarding the relationship between economic growth and governance. This thesis examines the relationship of World Governance Indicators (WGI) and GDP growth of Bangladesh. Both exploratory and econometric analyses of the relationship have been undertaken. The study estimates regression model using data for the period, 1996-2018. The regression results find that control of corruption (CC) is the most important governance variable that influences positively on economic growth. Similarly, rule of law (RL) shows significant positive relationship to economic growth. Although the estimated coefficient for voice and accountability (VA) is statistically significant, the negative magnitude of the coefficient does not really reflect the expected impact on GDP growth, perhaps due to small sample size of the study. However, the remaining three variables such as government effectiveness (GE), regulatory quality (Reg) and political stability (PS) are not found statistically significant and have not really impacted economic growth in Bangladesh. Therefore, if corruption can be curbed and ensure rule of law, we will able to attain a steady higher economic growth. In this paper, we have carried out tests to determine the presence of serial correlation. The results show no presence of serial correlation, hence, no corrective measures required. As the sample size is small there are the issues of stationarity, causality, multicollinearity, endogeneity that may arise in time series model. Within the purview of this analysis, we refrain from carrying out further tests of perfect collinearity, heteroskedasticity, stationarity, etc for the purpose of simplicity. However, such tests are highly recommended for more reliable and robust results.