Can public administration explain economic performance: the case of sixty three provinces in Vietnam

Abstract

Economic growth can be explained using fundamental (De facto) cause and proximate (De jure) ones. This study attempt to explore proximate nature which is of elements of economic activities of Vietnam including institutional quality based on cross-province dataset. The methodology adopted is informal growth regression, which allow to account variables that represent the underlying as well as proxy for proximate causes of growth. Multiple imputation (MI) technique is employed to deal with missing values of regressors since some observations are measured each couple of years and missing values. Problem of model uncertainty is addressed using approach for model selection named BMA (Bayesian model averaging) to evaluate probability of including of predictors in growth model. Variables flagged as being important by the above procedures then used in formulating the final models. Using dataset collected from 63 provincial administrative level units of Vietnam with 15 proximate determinants, which are proxy for public administrative reform in Vietnamese context. Panel data difference GMM (DGMM) is applied to examine in some detail and to deal with multicolinearity and endogenous nature of data. The results indicate that elucidating economic growth variations of provinces in context of public administration would rather choose the fundamental group of variables than group of proximate variables. For determinants of growth, an increase of public expenditure and a high level of lagged GDP will help to significantly raise provincial economic health. In context of institutional change, control of corruption is proved to remarkably affect current economic well-beings. Moreover, with additional models of estimations to quest for possible effects of all variables, the results show that the result of focused model of this study are consistent and robust.