Assessing the validity of structural reform aggregations in economic growth models of 24 transition economies

Abstract

During the early 1990s a trend of aggregating policy variables emerged out of econometric examinations of growth in transition countries, where more conventional growth models proved inadequate or incomplete. This thesis examines the use of one aggregation in particular, the cumulative liberalization index (CLI) representing structural reforms, and how different constructions of the index based on its conceptual structure can lead to different results. Further, the implications of different conceptual structures can lead to far different conclusions in applications of policy. This analysis shows, that although the different aggregations have similar capacities to predict growth, there are important differences in how the results are interpreted and applied. Namely, different conclusions about the effect of liberalization policy can be formed or obscured based on the outcomes of economic models that use different aggregations of policy indicators. A direct line can be drawn from the concept structuring through to the results and interpretation. Additionally, a conceptually simplified model based on the substantive results of the analysis is proposed that eliminates the need for a structural reforms construct entirely.