Working Paper: CEPR ID: DP954
Authors: Danny Quah
Abstract: This paper reinterprets a simple model of growth and fluctuations across many economies to allow for the explicit characterization of the dynamically-evolving cross-economy distribution of income. Such a framework provides a more natural, revealing study of the convergence hypothesis. The data show limited intra-distribution mobility in incomes across economies and thus, little convergence. The analysis uncovers some `convergence club'-like dynamics, and reveals the wide diversity in growth experiences across countries. Conditioning on physical capital investment, secondary school enrolment, and a dummy for the African continent fails to overturn these characterizations.
Keywords: convergence; education; evolving distribution; first passage time; growth; investment; polarization; stochastic kernel
JEL Codes: C23; F43; 047
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Traditional empirical analyses of convergence yield misleading results (O47) | Poor economies are catching up with richer ones at a rate of approximately 2% per year (F62) |
| Conditioning on standard variables such as physical capital investment and education does not significantly alter findings (D29) | Observed persistence in income disparities (D31) |
| The dynamics of income distribution are characterized by significant diversity in growth experiences (D31) | Majority remain either rich or poor (D31) |
| Standard regression techniques fail to capture intricate behaviors of economies at different income levels (C51) | Understanding the dynamics within the distribution is crucial (D39) |