How Robust Is the Growth-Openness Relationship to the Inclusion of Standard Control Variables? A Cross-Country Study over the Period 1970-2000

Abstract

The objective of this work is to investigate the robustness of the relationship between openness and growth to the inclusion of standard control variables. To this end, we use data on a set of countries over the period 1970-2000 to estimate an augmented Solow regression model, in which a measure of openness, due to Sachs and Warner (1995a), along with other variables, found to be robust in growth regressions based on different statistical criteria, are included. Among the indicators used to control the relationship between growth and openness in our regressions, we mention the black market premium, the share of primary goods exports to real , the terms of trade volatility, and governance and institutions. The results obtained show that openness is generally positively linked with growth. The magnitude of the effect of openness on per capita growth, and its statistical significance, depend, however, crucially on the controlling variable(s) being used