Do Tariffs Reduce Trade Deficits?
International Economics as I Learned it
Noah Smith writes,
The first point is that as far as I can tell, nobody really understands international economics. It’s basically macroeconomics on steroids. There are a huge number of factors that make issues of tariffs, trade surpluses, and the effect of trade on consumption vs. investment very complex.
When I was in grad school, international economics did not seem so hard. I mean, apart from the fact that in the diagrams Dornbusch drew on the board, a currency appreciation was when e went down, and I kept thinking it was supposed to be up. But even though every time he called on me in class I got the answer wrong, I think I got the basic ideas right in the end. I didn’t flunk the class.
The way I learned it, a trade surplus was the mirror image of a surplus of national savings over national investment. National savings is equal to the government budget surplus (haha) plus household saving plus business saving (retained earnings).
When Chinese national saving exceeds what they need to fund their investment and their budget deficit, China has to accumulate foreign assets. They can obtain U.S. Treasury bills, for example.
Noah writes,
a Chinese trade surplus is just countries writing China IOUs in exchange for physical goods. Countries don’t really have a comparative advantage in writing IOUs.
But if your country does not save enough to cover its government deficit and its business investment, it does have a comparative advantage in writing IOUs. In the nineteenth century, as the U.S. was financing its railroads and other capital investment, we had a comparative advantage in writing IOUs. Poorer countries today that are building infrastructure would have a similar comparative advantage. Today, the U.S. has a comparative advantage in writing IOUs because we have a reputation of paying back.
As China saves (and the U.S. borrows), China buys Treasury bills, which drives up the value of the dollar. That makes U.S. goods expensive and Chinese goods cheap, so that China runs a trade surplus. That trade surplus is the mirror image of their national savings surplus.
Now, suppose we impose a broad tariff on Chinese goods.1 What happens to the Chinese trade surplus?
To a first approximation, nothing. We still have a comparative advantage in writing IOUs. The dollar will appreciate, so that the price of Chinese goods in dollars will not rise by as much as the tariff. It might not rise at all.
That is not quite the entire story. The tariff revenues add to our national saving and subtract from China’s. So we get a bit of a reduction in the trade deficit from that. But that is not a first-order effect.
That is international economics as I learned it. With flexible exchange rates, a tariff leads to a currency appreciation, which mostly offsets the tariff.2
The primary determinant of net trade flows is the difference in national saving rates. If we do not do anything to increase our national saving, we are going to keep running trade deficits.
You may want to attack me in a comment. But I know that
(S-I) + (T-G) = (X-M)
That may not be much to know, but if it’s more than what you know, then don’t expect me to be impressed by your analysis.
I am NOT talking about a tariff on one particular good, such Chinese cars. If we charge a tariff on cars but not on anything else, then the result will be that we buy fewer Chinese cars and more of something else from China. By a broad tariff I am talking about a tariff on all Chinese goods. And as I will argue, this does not reduce imports of Chinese goods, because the exchange rate will move to offset the tariff.
Actually, not all of the adjustment will be between the U.S. and China. We may import more from other countries and other countries may import more from China.
The remarkable thing about the equation (S-I) + (T-G) = (X-M) is that it is an accounting identity or, put another way, a simple matter of the arithmetic adding up. Thus, it must hold for every country whether capitalist or socialist, whether under free trade or controlled trade (tariffs, quotas) and regardless of the monetary regime (capital controls, gold standard, fiat money, managed or free-floating currencies). And yet, this simple accounting identity drives real-time economic phenomena as Arnold described: whether a country is in a trade or current account deficit, whether its currency appreciates or depreciates, and whether it suffers rapid inflation or enjoys stable inflation.
A related point is that managing international trade and finance (tariffs, capital controls, exchange rates), as a first order effect, has no influence on the level of employment. Such management mainly affects the mix of employment - the types of jobs that the labor force performs, and hence whether labor is allocated efficiently to its highest and best use or inefficiently to other jobs.
The second order effect of such management is worse because incomes suffer, which affects household consumption, investment, and labor demand, almost surely for the worse.
I think this is right, but would generalize it to say
1. A tariff is just a special case of a tax. And a quota is just a special case of a regulation.
2. In most cases, there's enough other things going in the real world that the immediate, theoretical effects may be more than offset by everything else. Does every study show negative employment effects from raising the minimum wage? Nope. Am I reasonably certain that raising the minimum wage does have a negative effect? Yep. Should we focus our time on the the small theoretical effects or the other factors that tend to dominate them? Obviously the latter.