Going back to school takes time: Evidence from a negative trade shock

We estimate the impact of a negative trade shock on labour market outcomes and educational choices of workers. We exploit the Canadian lumber exports crisis beginning in 2007 in a quasi-experimental design. We find that the employment probability of forestry industry workers decreased by 4.1 percentage points following the crisis relative to other workers in comparable industries. While one would expect younger forestry workers to return to school in such circumstances, we find that in the first two years following the crisis, unemployed workers did not go back to school. But going back to school takes time, and after 3 to 4 years, we find that education enrollment increases by 2.5 percentage points (p=0.083). This confirms the idea that adjustments towards an increase in education enrollment are gradual, as it is easier to drop out than to enroll. In time of crisis, facilitating a return to education might be a valuable policy intervention.