Import competition is known to cause lasting income losses for manufacturing workers, but its effects on their children have been largely overlooked. One might expect these losses to tie children's income more closely to their parents' income, since poorer parents have little to fall back on, while richer ones can keep investing in their children. Using matched employer-employee data that link children to their parents, I find the opposite: import competition from China increases intergenerational income mobility among children of Canadian manufacturing workers. In heavily exposed families, a child’s position in the income distribution depends about 20% less on where their parents stood. Children from poorer families climb the income ladder; they move out of their parents' industries into ones that offer similar pay on average but where family background matters less. Children from richer families fall; they stood to follow their parents into high-paying manufacturing industries, and the shock takes those jobs away.

This paper exploits quasi-experimental variation in local exposure to the shock of China's accession to the World Trade Organization (WTO) to evaluate its impact on high school enrolment, a key margin of human capital accumulation. I construct a measure of prefecture-level trade exposure using initial industrial composition, industry-level tariff reductions, and spatially determined geographic trade costs, and I estimate both the short- and long-run impact of the shock on enrolment. Using detailed population census data on successive cohorts of 16-year-olds from 1995 to 2014, I find that a one-standard-deviation increase in exposure reduces high school enrolment by 8% relative to the baseline rate of the last pre-accession cohort. Surprisingly, this drop does not fade over time; it persists and even grows across successive cohorts. These results suggest that China's WTO accession created an export-driven labour demand shock biased toward low-skilled workers, which was in turn magnified by subsequent complementary investments.

Production today is split into sequential stages located in different countries. A tariff on one input therefore moves the prices of the goods made from it downstream and of the inputs used to make it upstream, and part of its effect returns to the country that imposed it. Where along a supply chain should a country levy a tariff, and how high should it be? I develop a two-country model of sequential production in which the government chooses both. The optimal tariff at a given stage extends the classic inverse-elasticity rule with two new forces. The first, a re-export rent borne by foreign buyers of downstream output, raises it. The second, a cost boomerang, is the domestic value that returns through the price of the taxed import. A tariff on final goods forgoes the re-export rent entirely, so taxing further upstream can raise welfare by more.