Recently, I tweeted this about the trade balance:
Ro Khanna's Twitter bio says "Trade surplus again." I wish people would think more deeply about the trade balance. Here are some good pieces to read: cato.org/blog/trade-def cato.org/testimony/amer cato.org/free-trade-bul
My basic point was that, in my view, the trade deficit is widely misunderstood, and I linked to several articles that I think provide useful corrections to various misunderstandings.
I didn't expect anything to come of the tweet except a few retweets/likes, but Rob Atkinson of ITIF replied, and then Ro Khanna (D-CA) himself weighed in!
Eventually, the Twitter exchange devolved, as they often do, into a bunch of threads that were hard to follow, and I had actual work to do at that moment, so I said I'd be happy to respond to all this in a blog post if they had papers that outlined their views in more detail. Rob then pointed me to a paper of his. I'm going to make this blog post somewhat of a response to Rob's piece and to a Foreign Affairs article by Ro, although there will be some other things thrown in as well.
Let me start with some big picture points. First, with regard to the causes of U.S. trade deficits, I think Gary Hufbauer's explanation here is a good one:
The size of the trade deficit is primarily determined by four macroeconomic forces:
- The difference between US household and business savings on the one hand and spending by US residential construction and business investment on the other hand;
- The combined deficit of federal, state, and local governments;
- The level of economic activity in the rest of the world, particularly in countries that are close trading partners of the United States;
- The trade-weighted exchange rate of the US dollar.
If that is all correct, and I think that it is, the particular focus of a country's economy on one sector or another is not a key long-term factor in determining the trade balance, although there are circumstances in which sectoral changes could lead to short-term shifts towards a deficit or a surplus.
Second, the existence of a trade deficit does not mean your economy is doing badly. As has been pointed out many times, U.S. trade deficits tend to go up during economic booms, and down during recessions. That’s because when times are bad, we consume less, which includes importing fewer things. Conversely, when times are good, we consume more. (See, e.g., this piece by Christine McDaniel from April 2020: "Today, we are importing less because Americans are consuming less during an economic shut down.") Thus, too much emphasis on the trade balance as a metric of how the economy is doing is a mistake. I think we can all agree that a trade surplus because of a recession is not desirable.
Bet let's now get into the issue of manufacturing, because I think that's at the heart of the arguments from Rob and Ro. They want the U.S. to do more manufacturing and they see that as closely tied to the trade balance. In their view, if I understand it correctly, the decline in U.S. manufacturing led to the trade deficit, and an increase in manufacturing would eliminate the trade deficit. For example, Ro says: "The deficit reflects the decline in domestic industry." If we can do enough manufacturing, they believe, the trade deficit will turn into a trade surplus, and the economy will be better off. (If that's not the claim, I apologize, and perhaps then Rob and Ro can clarify what their claim is.)
So, while I wasn't saying anything about manufacturing in my original tweet, I think that's where they want the conversation to go, and I'm going to focus on that now.
The practical issue, as I see it, is whether government intervention with the goal of promoting more manufacturing in the U.S. would eliminate the trade deficit (and also whether that would be a good idea more broadly).
One general point worth noting at the outset is that if you do more of one thing in your economy, you will almost certainly do less of something else. The human and financial resources that will be shifted to manufacturing have to come from somewhere. Other key sectors of the economy are services, agriculture, and natural resource extraction. If we are going to do more manufacturing, we will do less of one or more of those.
Now, the government can certainly use measures such as tariffs and subsidies to encourage more manufacturing in the U.S. I don't think anyone disagrees that this is possible. But should it? To answer that, it's worth considering the following effects that such measures would have:
-- fewer manufactured imports (and less overall competition and higher prices in the manufacturing sector)
-- unclear impact on manufacturing exports (depends on if these new manufactured goods are competitive in international markets and if the companies are going to try to export them)
-- shift of resources (such as capital and labor) from other sectors to manufacturing, so presumably less production and fewer exports of those products/services
-- retaliatory measures by trading partners
What would this mean for the overall trade balance? It's hard to say. It really depends on the degree of each of these specific effects. If any economists have written about this sort of thing, I'd love to hear about it.
My overall point here is that for people who want the U.S. to do more manufacturing, the overall trade balance is of very little relevance. The trade balance is mainly dictated by the macroeconomic factors noted above, not some specific new factories that were opened. The "seen" part of the equation is the new factory, but the "unseen" has to be taken into account too, as the government efforts to shift resources around have an impact on other sectors and the overall economy.
To respond to Rob and Ro more directly, a key passage from Rob's paper says:
If the United States is to reduce the trade deficit in goods, it will need to find a way to produce more here, in part by significantly increasing manufacturing productivity growth rates. If it can do that and eliminate the manufactured trade deficit, ITIF estimates the nation would gain an additional 1.3 million manufacturing jobs.
This makes me think we are talking past each other a bit. Rob has in mind the manufacturing trade deficit, whereas I'm talking about the overall trade deficit. Would more U.S. manufacturing lower the U.S. manufacturing trade deficit due to fewer imports? Quite possibly. But the impact on the overall trade deficit is, for the reasons noted above, less clear.
And Ro seems to believe that manufacturing is inherently more important than other sectors, stating: "But innovation is intrinsically linked to production." That's a subject worthy of debate, but it is not related to the overall trade balance.
Summing up, there is an important debate to be had about whether it makes sense to use government intervention to shift the U.S. economy towards more manufacturing. But the overall trade balance isn't particularly relevant to this question. If the overall trade deficit is the main concern, the only solution is to address the macroeconomic issues.
