Less Bad Arguments for Protectionism

In a recent post I sketched why the standard argument for protectionism, that it protects American workers from losing their jobs to Chinese workers and that tariffs prevent a negative (“unfavorable”) balance of trade or produce a positive (“favorable”) balance, are wrong. Tariffs protect some American workers, producers of goods that compete with imports, from the competition of other American workers, producers of export goods. A negative balance of trade means the country is importing capital, as the US did through much of the 19^th century, building railroads and digging canals with the money of European investors. It is not an inherently bad thing and, whether bad or good, tariffs do not prevent it, as becomes clear once one takes account of their effect on exchange rates.

There are, however, better arguments for tariffs, trade restrictions in general, arguments that do not depend on ignorance of things economists have known for two hundred years. Some were raised by commenters on my post, others have been raised by protectionists who have not only read the economic arguments but understood them, possibly including the current Republican candidate for vice president. This post is about those arguments.

Efficiency vs Utility

My previous post sketched the argument for thinking that trade benefited us, that we can grow cars at a lower cost than we can build them, but did not explain what that meant. Either imposing or removing a tariff will make some people better off, some worse off. How do you add up those effects to say whether a change harms or benefits “us”?

One philosophical answer is provided by utilitarianism: A change is an improvement if it increases total utility,¹ makes people happier by more than it makes other people less happy. But, while we can tell whether one thing makes you happier than another — given the choice, which do you choose — we have no way of comparing how much something makes me happier with how much it makes you less happy.

The standard economic answer, originated by Alfred Marshal more than a century ago, is a concept that he called “economic improvement” and modern economists call “economic efficiency.” Define the value of a change, positive or negative, to an individual as the largest amount he would pay to get the change (positive) or prevent it (negative). Add up the values. If the sum is positive, the change is an improvement.

The advantage over the philosophical answer is that value, unlike utility, is revealed in behavior. If, to take the simplest case, I buy something from you, the fact that I was willing to pay a price that you were willing to accept means the value to me was more than to you, hence the transaction was an economic improvement. Marshal, who was a utilitarian, offered the concept as a proxy for utility increase, arguing that in most cases differences between people in the relation between value and utility would average out, since most economically interesting cases involved effects on large and diverse groups.

That may not always be true. Imagine a change that increases value for rich people by $100,000 and decreases value for poor people by $90,000. Arguably a dollar represents more utility for poor than for rich, so the change is an increase in total value, an economic improvement, but very likely a decrease in total utility. The economics of trade imply that, outside of some special cases, a tariff reduces total value. But if the people it benefits are richer than the people it harms, it might decrease total utility. That is not terribly likely for the auto/wheat example of my earlier post, since auto workers are probably better paid, on average, than farm workers, and very likely than auto buyers. But it might apply to a tariff that benefited unemployed industrial workers at the cost of Silicon Valley programmers and random consumers.

National defense

One of the reasons the US won WWII was that we started with a very large auto industry, which could be and was converted to produce tanks and other military vehicles. One of the reasons Ukraine is not winning its war with Russia is that neither Ukraine nor its allies has ammunition factories that can produce artillery ammunition at the rate Ukraine would like to consume it. One can imagine a case where the industry you wanted to have in case of wartime was not profitable in peacetime without trade restrictions. It was, as I understand it, the reason the UK used trade barriers to maintain its food production after World War I, during which there was a serious threat that German submarines would succeed in blocking food imports. One could even imagine a case where the foreign competitor was also the likely future enemy;  I think some people imagine that being true with regard to China.

It is a logical possibility but not a very realistic one, since it requires the government to know well in advance which industries it will need in a future war and for those to just happen to be ones that can be maintained with a tariff, cannot be without it. We now know that a critical ability in modern warfare is the ability to produce drones but we did not know that four years ago. If the government did know what industries it would need in a future war it might make more sense to subsidize capacity instead of production, offer money to companies to guarantee the ability to rapidly expand production if needed, with occasional tests of their claims. That way you are paying for what you want instead of to produce products that cost more to build than to buy.

A further problem is the politics of the process. National defense provides an excuse for politically influential industries to get protection, requiring only a moderately plausible story about an uncertain future. Government actors are, for reasons I have discussed here in the past, mostly concerned with short run costs and benefits, so the question of what trade policy will best serve the nation ten or twenty years in the future is likely to have little weight against the question of what industries can offer votes and money in exchange for protection today.

Externalities

The proof of the efficiency of free trade depends on producers bearing the cost of producing, consumers getting the value of consumption. One can imagine cases where that was not the case. If, for example, the export industry produces a lot of air pollution and the import competing industry does not, the total cost of getting goods by trade, including the damage done by the air pollution from producing the goods exchange for them, could be higher than the cost of producing the import competing goods domestically.

A better solution would be to charge the polluting industry for its pollution and see if, with that cost included, it was still cheaper to get goods by trade than by producing them. The argument for protectionism only works for subtler externalities where that option doesn’t work. I find it hard to think of any likely ones, too subtle to actually tax or reward but sufficiently visible to be the basis of trade policy. But I expect JD Vance probably can.

Infant industry

An old argument for tariffs is that there might be an industry which the country would be competitive in if it could once get started but which would lose money in the short term. If that is the case for a single firm, and the profitable future could be predicted, it would pay the firm to lose money for a while in order to make more in the long run, a common pattern for a firm entering an industry. For the argument to work, there have to be economies of scale, or something similar, internal to the industry but external to the firm, some reason that company A losing money in the early period makes it easier for company B to produce at a profit, possibly information generated by A that B will have access to. This again is not logically impossible. But established industries have much more political power than infant industries, US steel than Apple in the days of the Apple II, and are more likely to end up with government favors such as trade barriers against their foreign competitors.

Paternalism

In explaining his economic views during a senate hearing, Vance referred to "our mass consumption of mostly useless imports.” That is another way of avoiding the free trade conclusion, or any conclusion based on the value of individual choice. Assume that other people don’t know what is good for them — and you do.

As should be clear, I do not find any of these arguments very convincing but they are at least internally consistent and, to that extent, better than arguments based on not understanding the economics of trade.

---

Past posts, sorted by topic

A search bar for past posts and much of my other writing

1. I am ignoring the issue of whether the maximand should be total or average utility, which I have discussed elsewhere.