Why did the tariff-growth correlation change after 1950?

This article uses a new database to establish a key finding: high tariffs were associated with fast growth before World War II, while they have been associated with slow growth thereafter. The paper offers explanations for the sign switch by controlling for novel measures of the changing world economic environment. Rejecting alternative explanations based on changing export market growth or transportation cost declines, it shows how the modem negative correlation could be reversed in a world environment characterized by a moderately higher generalized tariff protection such as that which prevailed before 1914. We show that an increase in average tariff rates among trading partners by just one third might suffice to reverse any negative relationship between an average country's tariffs and its growth. An increase in own tariffs after 1950 hurt or at least didn't help growth, but it might have helped growth in a world where average trading partners' tariffs were moderately higher and retaliation was the best strategy. The world environment matters. Leader-country reaction to big world events matters.