While previous studies of the impact of trade reforms on economic growth have used a single binary variable to represent the dating of a significant liberalisation in policy, the authors use several new measures of trade policy reform, focusing in particular on a time-varying indicator of current account openness that reflects the depth and speed of those episodes, as well as policy reversals. These measures lead to more precise estimates of how changes in trade policy affect the path of economic growth. The results show that going from closed to open trade policies leads to a 10-20% increase in per capital GDP after 10-15 years. The results also reveal an interesting asymmetry: the costs of closing to trade are significantly larger than the gains from opening of trade.