March 25, 2016
Everyone knows that reasonable people are supposed to hate protectionism, that is of course unless it’s for doctors and lawyers, who lack the skills necessary to compete in the world economy (or drug patents). But that shouldn’t mean that an ostensibly serious newspaper (I’m feeling generous today) gets to say whatever it wants to trash the policy.
Today we have the spectacle of the Washington Post telling us that Donald Trump’s plan to impose 45 percent tariffs on imports from China coupled with his plan to impose 35 percent tariffs on imports from Mexico would cost us 7 million jobs if the countries retaliate and 3.5 million if they don’t. This is supposedly the output that Mark Zandi got, the chief economist of Moody’s Analytics, when he plugged these tariffs into their model. That seems more than a bit high to me. The logic of the tariffs is that they make it more expensive to import items from these countries, but the extent to which they raise prices here depends both on the extent to which we can substitute domestic production or can find other foreign sources.
The latter is likely to be especially important, since many of the items produced by both countries can be readily found elsewhere. In fact an analysis by the Peterson Institute of tariffs the U.S. imposed on imports of tires from China found that the tires were almost entirely replaced by imports from other countries. For this reason, the impact on consumers from tariffs imposed on these countries is likely to be substantially limited by the availability of imports from other countries and/or our ability to produce these items domestically.
But just to get a crude idea, let’s assume that the price of our imports rise by half of the amount of the tariff. This is almost certainly a huge overstatement since for many imports the price rise will be just a small fraction of the size of the tariff, since there are alternative sources and even in the extreme cases the suppliers will almost certainly have to eat some of the tariff in the form of lower profit margins.
Last year, we imported $482 billion in goods and services from China. If the price of these items were to rise by 22.5 percent, that would imply an increase in costs of roughly $106 billion. We imported $295 billion worth of goods and services from Mexico. If the price of these items were to rise by 17.5 percent that would increase costs to consumers by $51 billion. That brings the total hit to an increase in import prices of $157 billion, or a bit more than 0.8 percent of GDP. This is supposed to cost 3.5 million jobs.
Let’s compare this to the rise of the price of oil imports in the last decade. Petroleum imports cost us $263 billion in 2005. This rose to $476 billion in 2008. That’s an increase of $213 billion, which was roughly 1.5 percent of GDP at the time. If paying more for imported oil works in Zandi’s model the same way as paying more for imports from China and Mexico, then we would have expected to see a hit of around 7 million jobs from the oil price hikes. We did see a hit that large, but most people, including Zandi, attributed it to the collapse of the housing bubble and the resulting financial crisis, not the runup in oil prices. (In fairness, this run-up was partially reversed with the recession.) It also seems that Zandi’s model doesn’t find any increase in domestic production in areas where consumers spend less money on imported items.
Perhaps even more peculiar is the idea that the impact is doubled if China and Mexico retaliate in kind. Really?
Our exports to China last year were $116 billion. We exported $236 billion to Mexico. Do exports to both countries go to zero with 45 percent tariffs and 35 percent tariffs, respectively? And if we can’t export to China and Mexico, does Zandi’s model show that we can’t export these items anywhere, or consume them in the United States? If so, that sounds like a very interesting model.
The Post really hates protectionism except when it benefits its friends, we get that. But it should still try to be a little serious when it talks about this stuff.
Note: Some additional comments were added to this post.
Second note: I should say that I think the author of this piece, Jim Tankersley, tried to approach the question with an open mind. I don’t think he got a serious answer from Mark Zandi, which unfortunately took central place in the article. I think many others at the Post, in both the news and opinion sections, do not approach the trade issue with an open mind.
Comments