Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.
Donald Trump has chosen Peter Navarro, one of the most ardent critics of trade with China, to head his newly created National Trade Council -- a move many consider to be the opening shot in a trade war. Though I think the China trade boom of the 2000s did have some harmful effects, especially on certain U.S. workers, I worry that Navarro’s overall approach to trade doesn’t make economic sense.
The Cato Institute’s Dan Ikenson recently took Navarro to task for his views on trade. Ikenson says Navarro is making an elementary error when he writes:
When net exports are negative, that is, when a country runs a trade deficit by importing more than it exports, this subtracts from growth.
It’s definitely not true that trade deficits always subtract from growth. This is a common error that I see people making, so it’s important to explain why it’s wrong.
Everyone who’s taken an economics class knows that gross domestic product, which represents the total value of the stuff a country produces, can be broken down into four parts:
GDP = Consumption Investment Government Spending Net Exports
That last term, “net exports,” represents the difference between exports and imports. So the equation is actually:
GDP = Consumption Investment Government Spending Exports - Imports
Because imports have a negative sign in this equation, lots of people think that if imports go up, GDP (and therefore growth) goes down. But that’s not right. The reason, which every econ student ought to learn, is that imports also add to consumption, investment or government spending.
Suppose you’re living in Cleveland, and you buy a video game console made in Japan. Consumption goes up. But we don’t want to count that in GDP, since the console was made outside the U.S. So we subtract the value of the console. That’s why the “imports” term in the GDP component equation is negative. It’s just to avoid bad accounting. Similarly, if Intel Corp. imports a German machine tool, investment goes up, and we need to subtract the import to make sure GDP stays the same.In other words, imports don’t count negatively in GDP. They amount to zero.
This means that a higher trade deficit doesn’t have to make the U.S. poorer. Imagine a situation where the U.S. keeps consuming and investing the same amount of domestically made goods, but starts exporting $1 billion more to foreign countries and importing $3 billion more. The trade deficit just went up by $2 billion dollars, but GDP went up by $1 billion! Growth was positive! The new imports didn’t change U.S. GDP, but the new exports added to it.
Similarly, cutting the trade deficit doesn’t necessarily make the U.S. richer. Suppose the country enacted trade restrictions that cut imports by $10 billion and cut exports by $1 billion, but (as in the previous example) left consumption and investment of domestically made goods unchanged. The trade deficit just shrunk by $9 billion, but GDP declined by $1 billion. That’s negative growth. That’s a recession! That’s going to cost American jobs.
So when Navarro writes that Trump will boost growth by cutting the trade deficit, is he making this mistake? It sure looks like he is. Later in his white paper on Trump’s economic plan, Navarro writes:
Suppose the US had been able to completely eliminate its roughly $500 billion 2015 trade deficit through a combination of increased exports and decreased imports rather than simply closing its borders to trade. This would have resulted in…a real GDP growth rate that year of 5.97%.
This is just wrong. Imagine in 2015 the U.S. had closed the $500 billion trade deficit by exporting $1 more and importing $499,999,999 less. GDP would have gone up by $1, not by $500 billion. The increased exports would have added a single dollar amount to GDP, while the decreased imports would have left it unchanged. In that scenario, which fits Navarro’s description perfectly, GDP growth in 2015 would have been exactly the same as it actually was -- 2.6 percent.
So it does look like Navarro is making a mistake that an econ professor like him really shouldn’t be making. To really figure out the impact of trade changes on GDP, we need to know a lot more than just the simple accounting equation written above. We need to know how trade interacts with domestic production.
If I buy a Japanese-made laptop, does that mean I decide not to buy one made in the U.S.? If Intel buys a German-made machine tool, does that mean it chooses not to buy a different tool made domestically? The answers to these questions will determine how trade affects the economy.
If imports substitute for domestic-made goods and services, then reducing trade deficits really might boost the economy -- not via the simple accounting fallacy described above, but by boosting domestic producers. If tariffs on Chinese electronics caused companies like Apple Inc. to start making their products in the U.S., without spurring China to block U.S. goods, it really could give the American economy a shot in the arm.
But if imports are complements to U.S. production, then trade restrictions could backfire severely. Foreign-made parts are often essential to U.S. manufacturing. Foreign call centers and other business services help many U.S. businesses concentrate on doing what they do best. Foreign-made industrial tools are essential for many U.S. manufacturers. If these imports prove very expensive to replace with domestic equivalents, then trade restrictions could create huge costs for many American businesses, forcing those businesses to be less profitable, shrink or shut.
The moral of this story is that we can’t just do economics by looking at simple accounting relationships. The world is too complicated for that. Because it’s very hard to forecast the precise effects on the U.S. economy, trade restriction is an inherently risky policy.