This paragraph from the Council on Foreign Relations blog is disappointing:

The yearly U.S. trade deficit peaked at 6.4% of GDP in August 2006. It improved significantly after the financial crisis, bottoming out at 3.6% in January 2010. This swing provided a boost to GDP and nudged the U.S. external balance toward a more sustainable level. The deficit then resumed an upward march, reaching 4.3% by November. A closer look at America’s bilateral trading relationships since the deficit high-point in 2006 reveals a significant improvement with many countries, and only a small deterioration with a few others. China – with which the U.S. has its largest deficit – is the conspicuous exception, as the figure shows. 2011 looks set to be a year of yet further-rising trade tensions between the two countries.

First, the discussion of the relationship between the trade deficit and the business cycle is rather unclear. The second and fourth sentences suggest that the business cycle might be driving the trade deficit (the deficit is smaller when growth is down), but the third sentence says that the shrinking trade deficit boosted GDP.

The third sentence is certainly misleading. While it is true in an accounting sense that a smaller trade deficit raises GDP (Y = C + I + G + NX is an accounting identity), that says nothing about a causal relationship. The trade deficit and national income are part of a general-equilibrium system — it’s folly to describe one as causing the other without a clear exogenous shock or a proper GE model. Accounting identities are not behavioral relationships. One could just as easily write NX = Y – C – I – G.

The claim that the dampening of the trade deficit in January 2010 represented a move towards a more sustainable level seems unlikely. Since the income elasticity of trade is greater than one, the drop in global GDP and associated plunge in global trade did flatten out global imbalances. As soon as growth resumed, the imbalances began growing too. This describes an equilibrium comovement, not a direction of causation, but that’s enough to suggest that the January 2010 “improvement” was business as usual, not a move towards smaller long-run imbalances.

Second, the discussion of bilateral trade deficits is misleading. Bilateral deficits do not matter (economically, though politicians seem to care). The canonical analogy for undergraduates is that the professor will have a bilateral deficit with his/her local grocer for the rest of his/her life without any problems (CEA 2004, pdf). Moreover, trade statistics measure gross flows, so China’s role as an assembler of products made by “Factory Asia” destines it to have a bilateral surplus with countries importing finished products. Value-added measures of China’s surplus are substantially smaller. For these reasons, and many more, the discussion should focus on multilateral imbalances.