This paper re-examines the evidence linking poor growth during the era of import substituting industialization with trade restrictions. Recent work, notably Rodriguez and Rodrik (2000), asserts that all the evidence is fragile, implying that economists who believe that trade restrictions played a role in this poor performance do so largely on faith. This paper argues that this criticism ignores crucial evidence and tests regression specifications that are either not relevant for the question or are bound to have low statistical power. After revising and updating measures of trade openness used in Sachs and War (1995), the weight of the evidence argues that trade restrictions were indeed harmful to growth during ner this period. While there is certainly room for debate about how best to use the available data on trade restrictions, a large number of variants yield strong and consistent results. Furthermore, although some assert that it is impossible to empirically distinguish measures of trade restriction from other policies or institutions, the findings are not diminished even after controlling for many alternative economic policies or institutions.