Impact of Urbanization, Development Level, and Openness on Institutions in Bangladesh: An ARDL Investigation Using Two Measures of Institutional Quality

Abstract

Future development policies require a better understanding of the various factors that affect the quality of institutions in Bangladesh. In this paper, we estimate empirical ARDL models to investigate the roles of urbanization, openness, and the level of development as determinants of institutional quality. We use two measures of institutional quality, both derived from the World Bank’s Worldwide Governance Indicators (WGIs). The first proxy, INQ, is a composite index developed from the six WGIs using Principal Component Analysis (PCA). The second indicator, AvgINQ, is the average of the six WGIs. For each indicator of institutional quality, three models were estimated, resulting in six models in total. The impact of urbanization on institutional quality is consistent in terms of sign and significance across all six models, a claim that cannot be made for the other variables in the model. The error correction term reflects dampening convergence to equilibrium for the first proxy for institutional quality but regular convergence when the second proxy is used. In terms of ARDL diagnostics, models with AvgINQ as the dependent variable perform better. The Johansen cointegration test confirms the existence of a long-term relationship among the variables in all the models. Pairwise Granger causality tests are also conducted. The findings emphasize the importance of managing urbanization pressure for better institutional quality.