Doppelhofer, G. and Weeks, M.

Robust Growth Determinants

CWPE1117

Abstract: This paper investigates the robustness of determinants of economic growth in the presence of model uncertainty, parameter heterogeneity and outliers. The robust model averaging approach introduced in the paper uses a flexible and parsimonious mixture modeling that allows for fat-tailed errors compared to the normal benchmark case. Applying robust model averaging to growth determinants, the paper finds that eight of eighteen variables found to be significantly related to economic growth by Sala-i-Martin et al. (2004) are sensitive to deviations from benchmark model averaging. For example, the GDP shares of mining or government consumption, are no longer robust or economically significant once deviations from the normal benchmark assumptions are allowed. The paper identifies outlying observations - most notably Botswana - in explaining economic growth in a cross-section of countries.

Keywords: Determinants of Economic Growth, Robust Model Averaging, Heteroscedasticity, Outliers, Mixture models

JEL Codes: C11 C21 C52 O20 O47 O50

Author links: Melvyn Weeks

PDF: https://www.econ.cam.ac.uk/research-files/repec/cam/pdf/cwpe1117.pdf