National Defense
as an argument for tariffs
In Less Bad Arguments for Protectionism I offered a number of arguments for tariffs that, unlike the more common ones, are consistent with a correct understanding1 of the economics of trade; I do not find any of then convincing but someone else might. One had to do with national defense.
Suppose we get into a war with China. It would be inconvenient if some of the things we needed for the war, ammunition, computer chips, drones, or something else, were things we did not produce because we had been importing them from China. So it might be prudent to use protective tariffs to keep critical industries going even if they could not compete with foreign competitors. It is not an absurd argument, but it has several problems, especially as a defense of the tariffs Trump actually imposed.
Trump’s tariffs were imposed on a wide variety of countries and goods, to first approximation tariffs on everything. A supporter might claim that protecting everything was fine since the more industries we have the more we can produce if there is a war, but he would be wrong; while a protective tariff benefits an import-competing industry, it harms export industries. If foreigners get dollars by selling us things they have to do something with them and the alternatives are either to buy US goods and take them out of the country, our exports, or to buy US assets, T-Bills or shares of stock, and leave them in the country. The less they sell to us, the less they will buy from us. As explained in an earlier post, the result of a tariff is to shift the exchange rate, the price on the dollar/yuan (or dollar /EU or dollar/£) market, making dollars, hence American goods which are bought with dollars, more expensive. So a tariff that benefits an import-competing industry also harms all export industries; we are not getting more industries, just different ones. To argue that Trump’s tariffs make us better prepared for a future war with China you need some reason to believe that the import-competing industries helped by tariffs will be more important than the export industries harmed.
A possible response is that if we reduce imports and exports, in the limiting case abolish trade, we will be self-sufficient, produce as much of everything as we consume, no longer dependent on any foreign producers. The problem with that argument is that it assumes that what we consume in peacetime is a good proxy for what we will need to win a war.
It isn’t. The result of abolishing trade is that we are producing everything we use in peacetime and only what we use in peacetime. We will have more factories producing clothing and toys — but not producing enough clothing and toys would not be a problem in war time unless it is a very long war. We will have enough factories producing aircraft or weapons for domestic peacetime use but that will probably much less than we need for a war; if a war starts we would be better off if we still had the factories that had been producing those things for export. Other export goods that might be useful in wartime include integrated circuits, petroleum, chemicals, machine tools, computers, automotive vehicles, and cell phones. The fewer of these things we export the fewer we produce, the fewer we produce the less our ability to produce lots more if there is a war and we need them.
How To Have More Factories
I have so far argued as if a reduction in imports is matched by an equal reduction of exports. That would be the case if trade always balanced, imports equal to exports, if the trade deficit was zero, but it isn’t. The reason, as I explained in the earlier post, is that the dollars foreigners get by selling goods to us do not have to be spent on buying goods from us; they can also be spent to purchase capital assets, shares of stock or government bonds. A decrease in the capital inflow will reduce the trade deficit, lower the price of the dollar, increase exports, decrease imports, increase the production of both export and import-competing industries.
There is no obvious reason to expect tariffs to decrease the capital inflow, hence no reason to expect them to decrease the deficit or increase the number of factories in the US. There is, however, another relevant deficit: the budget deficit. One reason the US imports capital is that its government borrows nearly two trillion dollars a year, some of it from foreigners. If we reduced the budget deficit, borrowed less, US interest rates would fall, making both government securities and other capital assets less attractive to foreigners. The result would be to reduce the inflow of capital, hence the trade deficit, benefiting both export and import-competing industries. More factories.
Currently the budget deficit is about 1.8 trillion dollars, the trade deficit about .9 trillion, so if the US stopped borrowing money and the reduction in foreign purchases of government securities2 plus the reduction in foreign purchases of corporate stocks and bonds due to the lower interest they would pay if they did not have to compete with the government for capital came to half the reduction in government borrowing, the trade deficit would disappear.
The Other Thing Wrong With The Argument
I started this by imagining the problem of getting into a war with China and discovering that some of what we wanted for the war was made there instead of here. That is an argument for tariffs against China, countries that might ally with China and countries whose trade with us China would be in a position to block. It is not an argument for tariffs against the EU, probably not for tariffs against India, unless one assumes that the Chinese navy ends up in control of most of the oceans of the world. It is certainly not an argument for tariffs against Canada and Mexico. Currently only about 13% of US imports are from China, about the same fraction from Mexico and a little less from Canada.
Trump’s tariffs cover most of the countries we import from so are mostly against countries we could still trade with if we were at war . The argument I started this with might be a legitimate argument for tariffs on China covering only militarily useful goods but it makes no sense as a defense of the tariffs Trump has actually imposed.
The Improved Argument And What Is Wrong With It
Suppose someone, perhaps the next president, perhaps Congress after the Supreme Court has ruled against Trump imposing tariffs on his own, proposes a pattern of tariffs based on the national defense argument, tariffs against China, its allies Russia and North Korea, perhaps a few other countries, and restricted to goods, such as drones, believed likely to matter in a future war. That would be much better than what we now have. The cost to Americans would be less, since production of those goods would shift from China to whatever country, not necessarily America, could best produce them. We would be better prepared for a war with China since we could base our wartime consumption on the production of most of the world economy, get our drones from whatever friendly country, perhaps Ukraine, was best at producing them, have our fighters built at home in factories that had been producing airplanes, military or civilian, for export.
There are still two arguments against it. One is that we do not know what goods will matter in a future war; prior to the current Ukrainian war, drones would probably not have been on the list, tanks would have been. The other is that, whatever the public justification for the tariffs, what goods tariffs actually get imposed will be determined by the same political pressures as in the past. A standard justification for tariffs has long been the infant industry argument but in practice it has been senile industries, not infant industries, that got protection.3
There is, however, a compromise on tariffs I would be willing to support. Impose tariffs or import bans on militarily significant goods from China and its allies. Abolish all tariffs, institute unilateral free trade, with everyone else. If that feels too wimpy, make an open offer to most of the world: Free trade with anyone willing to have free trade with us.
My web page, with the full text of multiple books and articles and much else
Past posts, sorted by topic
A search bar for past posts and much of my other writing
Should I do an AMA?
A commenter suggest either an open thread or an Ask Me Anything, a post in which commenters can put questions on anything they want for me to answer. Do other readers think that is a good idea?
For what that is, see Ptolemaic Trade Theory.
For an explanation of why see Chapter 19 of my Price Theory. The relevant passage starts with the subhead “Tariffs in the Real World.”

I would benefit from an Ask Me Anything with David Friedman. For me, that is the best reason to have one. However, I fear that for David Friedman, most of the answers would be, "see Chapter x of "Hidden Order" (or "Price Theory")", or, "See my post at *link*," which would be boring for the respondent. How about an Ask Me (and an editorial assistant who posts links) Anything?
This is brilliant but isn't there another reason for not having tariffs against China, even on military goods? Namely, the more we trade, the friendlier our relations and the less chance of war.