This explainer from Mike Feroli at JPMorgan does a pretty good job showing why a focus on trade deficits as a measure of US economic vitality is wrongheaded (bold is by me):
When each worker in the economy does a narrow task efficiently, the overall economy is more productive than if each of us tried to grow our own food, sew our own clothes, and produce our own economic research. The basic case for free trade among nations rests on a similar argument—when each country can specialize in the activities at which it is most efficient, the overall size of the global pie will be larger. If trade were to be substantially restricted, we would expect to see lower levels of productivity and GDP in the long run, as workers and countries are forced to make things they are not especially good at making.
Of course, the same textbooks would caution that trade can have important effects on the distribution of income. As one example, low-skilled workers in rich countries can lose from being thrust into competition with workers in low-wage countries. But, although theory suggests that the overall gains from trade should be more than enough to compensate for these losses, nothing guarantees that this redistribution will occur.
President Trump brought the labor market effects of trade to the forefront of the policy debate during his campaign and has maintained focus on them after the election. But we note that his rhetoric often differs from the textbook concerns about losers from trade. In particular, he and his advisors focus on trade deficits as a measure of the nation’s losses from trade. Economists view the trade deficit differently: as a measure of our borrowing from the rest of the world. Every dollar by which US purchases from abroad exceed sales to abroad is matched by a dollar that is borrowed by someone in the US from someone overseas, who has chosen to invest that dollar in the US. Thus, the US trade deficit also reflects how eager foreigners are to invest their savings in the US.
It is possible that a large trade deficit could signal problems if such borrowing and lending become excessive, at which point it might make sense to reduce government deficits or encourage households to consume less and save more. But, in general, the trade deficit is not a measure of the losses created by trade. More relevant measures of the costs of trade would include unemployment, wage stagnation, or low labor force participation among workers affected by trade (which also receive attention, of course).And in our view efficient policy responses to these concerns would address them directly rather than by attempting to reduce trade and its many benefits.
We also note that bilateral trade deficits are a particularly poor measure of losses from trade. Bilateral tradedeficits are a natural outcome of specialization—just as each of us runs a trade surplus with our employers and a trade deficit with the stores where we shop, any given country might sell things on net to certain countries and buy things on net from others.
I guess the big question is whether President Trump and protectionist advisers such as economist Peter Navarro are educable or persuadable on this issue. Anyway, the alternative is not to ignore the losers from trade or hand-wave away harmful trade practices. And perhaps “trade deficit” is used as simple political shorthand for those issues. It’s pretty easy to make a chart showing big trade deficits.
But a focus on trade deficits leads one to propose the wrong fixes and reforms. Like broad tariffs. And they provide a poor metric of success and failure. Indeed, these supposed fixes and reforms can make the economy and workers worse off.