Most trade economists have come to believe that the gains from trade liberalisation go beyond lower prices for imported goods. Inspired by the ‘new’ trade models of Krugman (1980) and Melitz (2003), they now usually add that there are also new import variety and domestic productivity gains. The basic idea is that consumers benefit from having access to a wider range of imported products and average productivity grows as competition forces the weakest firms to shut down. For example, trade liberalisation allows consumers to enjoy an increasing variety of imported food items while at the same time shaking out unproductive domestic food firms.
These ideas gave rise to an empirical trade literature which generally concludes that there are large new gains from trade. For example, Broda and Weinstein (2006) assess the magnitude of the import variety gains and document that US consumers benefitted substantially from the increase in the range of imported products available to them between 1972 and 2001. Moreover, Trefler (2004) argues that Canadian manufacturing productivity rose significantly following the Canada-US free trade agreement as a result of the exit of low-productivity firms. However, Arkolakis et al. (2012) caution that under certain conditions, traditional and new trade models predict the same gains from trade which seems at odds with the notion of large new gains from trade.
In recent research, we reconsider the Canada-US free trade agreement (CUSFTA) and carefully account for the new gains from trade (Hsieh et al. 2016). Our accounting is based on an exact decomposition of the gains from trade in a generalised Melitz (2003) model, which we can link to confidential micro data from Canada and the US. Our decomposition reveals that the prior empirical literature has only provided an incomplete and selective account of the new gains from trade. Our punchline result is that the new gains from trade reaped by Canada were actually negative since Canadian consumers lost strongly from the exit of Canadian firms.
To understand this, it is instructive to consider the effects of CUSFTA on the set of firms serving the Canadian market. On the one hand, additional US firms enter into exporting that tend to be less productive than incumbent US exporters since it is now easier to export to Canada. On the other hand, some Canadian firms exit out of production that tend to be less productive than surviving Canadian firms since they now face tougher import competition from the US. Intuitively, selection into production and exporting follows the Darwinian principle of ‘survival of the fittest’ so that adjustments at these margins tend to involve relatively unproductive firms.
This means that to accurately measure the new gains from trade, we must account for both variety and productivity effects for both domestic firms and foreign firms. We must measure domestic variety losses and not just import variety gains. We must measure decreases in the average productivity of foreign firms that export to Canada and not just the increases in the average productivity of domestic producers. Our analysis also emphasises that Canadian consumers always gain from new foreign varieties and always lose from the exit of domestic varieties, regardless of productivity. The fact that, in practice, entering and exiting firms tend to be smaller and less productive than continuing firms only attenuates these effects on welfare. So, for Canada to reap positive new gains from CUSFTA, the gains from additional US varieties would have to dominate the losses from fewer Canadian varieties.
We estimate that in the eight years following CUSFTA, Canadian consumers experienced a welfare loss equivalent to 2.1% of real income due to the reduction of domestic varieties available to consumers. The increase in choice due to entry of new foreign varieties was only worth a 0.4% increase in Canadian real income. The fact that exiting domestic firms tended to be smaller and less valuable to consumers raised average productivity among domestic firms, but this was only worth 0.3% of real income during this period and was not enough to offset the loss of domestic varieties. Likewise, the new foreign entrants were less productive, which lowered their contribution to Canadian welfare by 0.2%. Combining all of these domestic and foreign variety and productivity effects, the new welfare effects of CUSFTA amounted to a 1.5% reduction in Canada’s real income eight years after CUSFTA came into effect.
The fact that the reduction in domestic variety far outweighed the increase in foreign variety does not mean that Canada lost from CUSFTA overall. On the contrary, we find that Canada’s welfare actually rose substantially as a result of this trade liberalisation with the overall gains amounting to 4.4% of real income. That means that the traditional effect of trade – cheaper imports due to tariff cuts – far outweighed the negative net variety and net productivity effects we document. The average tariff imposed by Canada against manufacturing imports from the US fell from over 8% to below 2% as a result of CUSFTA, thereby substantially reducing the prices Canadian consumers paid for US goods.
Our findings do not challenge the common belief among economists that trade liberalisation brings about welfare gains. However, they do challenge the notion that these gains arise because of an increase in the variety of available products or the average productivity of firms. While such variety and productivity adjustments have to be taken into consideration, the traditional point that trade liberalisation reduces import prices is still the most important argument for free trade quantitatively.
Arkolakis, C, A Costinot and A Rodriguez-Clare (2012) "New trade models, same old gains?", American Economic Review, 102(1): 94-130.
Broda, C and D Weinstein (2006) "Globalization and the gains from variety", Quarterly Journal of Economics, 121(2): 541-585.
Hsieh, C, N Li, R Ossa and M Yang (2016) “Accounting for the new gains from trade liberalization”, NBER, Working Paper 22069.
Krugman, P (1980) “Scale economies, product differentiation, and the pattern of trade”, American Economic Review, 70(5): 950–959.
Melitz, M (2003) "The impact of trade on intra-industry reallocations and aggregate industry productivity", Econometrica, 71(6): 1695-1725.
Trefler, D (2004) "The long and short of the Canada-US Free Trade Agreement", American Economic Review, 94: 870-895.