While I was teaching at the John Locke Institute, our Summer School sponsored a debate on free trade between Daniel Hannan and Terence Kealey. Kealey rested his case for protectionism squarely on the classic infant-industry argument. Kealey’s version: While free trade does indeed improve efficiency at the moment, the long-run effect is to suppress economic growth in poorer countries. Why? Because you don’t improve at doing things that you don’t do.
Suppose a rich country can produce cell phones for $200 each, while a poor country can only do the same for $1000 each. Under free trade, Kealey’s argument goes, the poor country will produce zero phones – and its cost of production will forever remain $1000 a pop. If 400% tariffs raise the price of foreign phones to $1000, however, domestic phone production will launch. And once it does, domestic phone factories’ costs will start to fall.
If you replicate this policy across a vast range of industries, the low-productivity – hence poor – country transforms into a high-productivity – hence rich – country. Yes, tariffs temporarily made the poor country even poorer. In the long-run, however, the tariffs had the opposite effect.
Kealey also combined this argument with vague claims that every rich country got rich via protectionism, but the theoretical argument was clearly the heart of his argument. After all, it wasn’t like he had a big multiple regression showing that all else equal, protectionism works wonders. Instead, he looked at history, saw tariffs, and attributed nations’ success to these tariffs. Why did he credit tariffs, instead of the thousand other factors at play in economic development in the past quarter millennium? Because the theory made so much sense to him.
But does Kealey’s theory really make sense? Not really. Yes, high-cost businesses could respond to tariffs by improving their efficiency. But they could just as easily respond in the opposite way.
Why? Ponder this analogy. You ban all players over 7 feet tall from the NBA. How will the remaining players react?
The optimistic scenario is that previously demoralized shorter players suddenly see a fantastic opportunity for stardom. They start practicing like crazy – and improvement naturally follows.
The pessimistic scenario, however, is that shorter players realize that they no longer need to practice like crazy to stay in the NBA. So instead of redoubling their efforts, they slack off. Their skills stagnate – or even get worse.
Notice: Even in the optimistic scenario, it is wishful thinking to assume that the shorter players will eventually improve so much that they actually become better players than the 7-foot-plus players who were summarily banned from the sport. If you’re lucky, the shorter players will improve for a while, then hit a plateau well below the standards of the players they replaced. If you’re unlucky, they’ll see the weak competition, breathe a sigh of relief, and relax.
The same goes for protectionism. If you’re lucky, protected industries will start improving, then hit a subpar plateau. And if you’re unlucky, protected industries will rest on their laurels, secure in the knowledge that domestic consumers have no choice but to “buy local.”
In wonkish terms, innovation is subject to both the substitution and income effects. Giving firms a protected market raises the incentive to improve (the substitution effect), but also gives firms the breathing room they need to take it easy (the income effect). Contra Kealey, the theoretical effect of protectionism on innovation is quite unclear.
Is there any way to gain greater clarity? You could try running bona fide experiments, but that’s ultra-unlikely to happen. In the world of trade policy, “experimentation” is an fig leaf for more protectionism, not a sincere attempt to figure out if protectionism works.
The alternate path to clarity, however, is to remember that the large majority of trade is domestic, anyway. Why? Because of (a) physical transportation costs, and (b) poorly-connected social networks. The upshot is that every country has powerful natural trade barriers. If Kealey is right, these natural trade barriers should have exactly the same effects as man-made trade barriers. Yet so far, these natural trade barriers have plainly failed to make the vast majority of poor countries rich. Indeed, a standard result in development economics is that being landlocked is very bad for growth.
I wouldn’t be shocked if a carefully-crafted experiment showed that under special circumstances, Kealey isn’t entirely wrong. Once in a while, the substitution effect for innovation might overpower the income effect. But once you acknowledge the ubiquity of natural trade barriers, it’s hard to believe that Kealey is right often enough to matter. And that’s ignoring another substitution effect so prevalent in actually-existing protectionism: When your firm’s fate rests on government favoritism, you have a strong incentive to focus on pleasing the government rather than your customers.
And yes, learning by doing works here, too. If the main thing you do is lobby the government, expect to become a master lobbyist.
READER COMMENTS
robc
Aug 25 2021 at 12:07pm
The only way I could see this idea working would be if the tariffs were slowly sunset.
Say it took advanced countries 20 years to advance from $1000 cell phones to $200.
The put the 400% tariff on, but reduce it 20% per year. So the local companies may get a pass at first, but they would need to be constantly improving or they would fall behind. If they can do it, they can compete straight up in 20 years, if they can’t, then it was probably a stupid idea to begin with.
The NBA equivalent would be to ban 7 footers but raise the limit 1/4th inch per year.
Both are still really stupid ideas.
Bill
Aug 25 2021 at 12:34pm
“The problem with protected infant industries is their tendency to become senile.” ~ Kenneth Arrow quip, Anchorage, Alaska conference, mid-80s.
Philo
Aug 25 2021 at 7:24pm
Once the infant industry is getting government support, inertia will tend to maintain this support indefinitely; the infant industry will never really “grow up.”
Dan C
Aug 25 2021 at 12:46pm
The infant industry argument only works, theoretically, if the industry in question benefits from significant economies of scale – high fixed costs, low marginal cost of production. Even then, subsidies including indirect export subsidies have been believed to work in (mostly) East Asian countries, while the tariff/protection approach used in Latin America largely failed. Small countries don’t benefit from economies of scale.
However, the limits to the subsidy approach are clear – if everyone subsidizes exports, then the benefits go away and everyone is left with the costs. Other limits include an overreliance and overinvestment in exporting industries, plus a lot of industry “duds” that waste precious resources (i.e., steel in energy-poor China), as well as a political inability to withdraw subsidies and fully reap the benefits of any success. Countries that were successful in the export-driven development model have faced/will face painful adjustments to wean themselves off of it (US in the 1930s, Japan in the 1990s, China in the 2020s).
Of course, everyone assumes that countries like Japan and China got rich off of the export subsidization, but that is a mistake of confusing correlation with causation. There are a lot of other variables at play, including basic liberalization of the economy – i.e., removing barriers to trade.
Radu Floricica
Aug 25 2021 at 12:49pm
Did you read this review? https://astralcodexten.substack.com/p/book-review-how-asia-works
Not trying to do it justice, a one phrase summary would be that protectionism works very well – when the government is competent enough to put the right incentive structure in place, and stick to it without getting soft or bribed.
Brian
Aug 25 2021 at 3:48pm
I have a recollection that one of the messages buried in the Studwell book is that some of the Asian countries where there was redistribution of land ended up being some of today’s higher GDP countries. It was kind of a buried message because there was a lack of vivid description of how this was carried out and the extent of turmoil and violence involved. That I haven’t seen econ bloggers discuss this makes it seem like I imagined reading it.
Brian
Aug 25 2021 at 4:43pm
Edit: per capita GDP
Paul Miller
Aug 26 2021 at 12:20am
+ 1 https://astralcodexten.substack.com/p/book-review-how-asia-works is a much better steelman to argue against.
Henri Hein
Aug 25 2021 at 1:20pm
Another problem is that a priori, we don’t know if the country has a comparative advantage in producing cell phones. If it turns out they don’t, then it is a pretty expensive experiment which is moot. To me, that seems to be a big part of the damage that the tariffs are doing: they shift production from the places that have a comparative advantage to those that don’t.
Michael Stack
Aug 25 2021 at 3:14pm
I’ve never really understood this argument for protecting infant industry. If this were actually the case, why wouldn’t an enterprising business person make the investment in the industry? Sure, he wouldn’t make profit right away, but once the company developed, he’d be well on his way to making lots of money.
If the investment required is too large relative to the return, then why should the government force such an investment on the public?
Jon Murphy
Aug 25 2021 at 3:51pm
This is a brilliant point you make. But there is a key assumption: it relies on well-functioning capital markets. In the West (even as the US was developing), that assumption of well-functioning capital markets holds true. But one could make the case that it does not hold for places like Africa, and thus infant-industry could help there (though the issues Bryan discusses still are in play)
Don Boudreaux
Aug 26 2021 at 7:10am
Jon,
You’re correct that among the assumptions used by those (including me) who argue against infant-industry protection is a reasonably well-functioning capital market. But this assumption is not as unreasonable as many people often take it to be, for at least two reasons.
First, for the assumption to be valid, capital markets need not be ‘perfect’ or even close to ‘perfect’; they simply must be able to allocate resources and capital goods better than will government officials.
Second, the appropriate domain in which to assess the workability of capital markets is – especially in the modern world – not the at the level of the individual country. For example, the capital market in Mozambique might well be primitive and inferior compared to that in the United States. But if a genuine profit opportunity exists in Mozambique, investors in the U.S. have every incentive to fund that opportunity. That we don’t see a great deal of such funding isn’t because of any failure of capital markets; instead, it’s because either such profit opportunities are rare (usually because of bad policies or culture), or because of government-imposed restrictions on foreign investments (again, bad government policy).
Jon Murphy
Aug 26 2021 at 7:47am
I agree wholeheartedly, Don. I make the same argument in my classes. I was just pointing out that one could make the case that there are not well-functioning markets (though the onus would be on that individual to explain 1) what are the barriers that are preventing a capital market from developing and 2) why tariffs are a more reasonable solution to that possible problem compared to the alternative of addressing the institutional barriers preventing the formation of the capital market).
Michael Stack
Aug 31 2021 at 1:55pm
I’d also add that if capital markets were insufficiently developed as described, it is incredibly unlikely the government would do a good job of managing a problem like this. It’s incredibly unlikely even in western societies, I can only imagine how something like this would play out in developing countries.
Eben Macdonald
Aug 25 2021 at 3:23pm
If 400% tariffs raise the price of foreign phones to $1000, however, domestic phone production will launch. And once it does, domestic phone factories’ costs will start to fall.
Honestly, I’d question that on empirical grounds. Are there any credible pieces of research which substantiate this? All the studies I’ve seen have consistently shown that tariffs raise consumer prices.
Jon Murphy
Aug 26 2021 at 12:50pm
Yes. It’s your basic economies-of-scale argument. We see it borne out in the empirical evidence all the time: as firms increase in size, their average costs fall, which can allow them to charge lower prices as they grow. The twist here is that tariffs help the infant industry meet the start-up costs they need to reach in order to form and grow
Yuliyan Mitkov
Aug 25 2021 at 6:29pm
I agree with Bryan, although there is one point he does not mention: positive spillovers.
Getting better at producing cell phones may come with other benefits not fully captured by the cell phone firms (i.e., technological know-how, skilled workforce/management).
Even if it is clear that the developing country has no comparative advantage in cell phone production (and capital markets are perfect), it might still be worthwhile to impose trade barriers if such spillover effects are significant.
Henri Hein
Aug 25 2021 at 10:29pm
Over lunch, I thought some more about this, and decided to elaborate on my point above about comparative advantage.
Either the country has a comparative advantage in producing cell-phones, or they don’t. If they don’t, then Kealey’s point falls apart. They are better off importing the cell-phones.
If they do have a comparative advantage in producing cell-phones, then it’s extremely unlikely to be the only electronics item they are able to produce efficiently. They probably already have some electronics production capabilities. If that’s the case, then the spread the Kealey is talking about ($200 vs $1000) is likely to be smaller. Also, there would be opportunities to ‘step’ towards the cell-phone. For instance, branching from industrial components to simpler consumer electronics. Maybe upgrading the sophistication of the components they produce, from dumb controllers to smart controllers. Etc. So he is wrong that “cost of production will forever remain $1000 a pop.”
Jakub Puterka
Aug 26 2021 at 5:48am
The comparison with NBA is not a correct one. If you banned super-tall players, more shorter players would be able to play the game. This is because height is such a deciding factor in basketball, it trumps almost anything (there was a stat that every fourth 7footer in the US had a chance to play in NBA). Due to how height is distributed in the population, the supply of eligible players would dramatically increase, with rest being the same the skills and quality of the game would go up.
Jon Murphy
Aug 26 2021 at 12:57pm
Problem: if skill is not related to height, then why do teams employ 7 ft. players at all?
Thomas Lee Hutcheson
Aug 26 2021 at 8:22am
If you think the infant industry argument is good (it’s not entirely crazy applied to export markets) you don’t “protect” the industry, you subsidize it with a tax that has lower deadweight loss than the protective tariff. Why would you ever tax domestic consumers of X in order to subsidize X? That would be as crazy as having a minimum wage instead of a wage subsidy.
Don Boudreaux
Aug 26 2021 at 10:46am
Bryan,
As usual, great post.
Here’s a mental experiment that I share with students in my International Economic Policy course when we get to the topic of infant-industry protection:
Suppose that I, as teacher of the course, were to tell each of you students that on each exam and paper that you submit for the course, I’ll raise the grade you earn by one letter grade. So a student who earns a B+ on a mid-term exam will be recorded in my Excel spreadsheet as having earned on that exam an A+. A student who earned a D on a paper will be recorded in my spreadsheet as having earned on that paper a C.
I explain that my grading policy is meant to encourage each student to study harder and, thus, to better learn the material. My reasoning is that a student who senses that he or she, with his or her ordinary studying, is destined to earn, say, only a C- will feel that it’s not worthwhile to study harder if the likely result of that additional studying will be only a C or a C+. The improved outcome isn’t worth the additional studying effort. But if by studying harder this student understands that the C+ he or she thereby earns will be raised by me to a B+, the result suddenly does seem worth the effort.
What could be simpler?!
Even my students understand that, were I really to make such a promise to raise their grades, the result would be less studying rather than more – and, hence, less learning rather than more
Pete Smoot
Aug 26 2021 at 4:58pm
I don’t see the relationship. I study an extra hour, I get a letter-to-letter-plus grade improvement. Does it really matter if the base grade was a C or B? Is there increasing marginal utility of grades as your grade gets higher?
Maybe there is: the closer my grade gets to an A+, the greater proportion of the delta each increment represents. IOW, C to C+ is about 1/9 of the way to an A+. B to B+ is about one fifth. A to A+ is 100%.
Thomas Lee Hutcheson
Aug 27 2021 at 7:26am
For the analogy to work, the subsidy has to work in such a way that the grade goes up more if the student produces more and the subsidy has to come at the expense of the other students. Hard to see how you could make that plausible.
Better just stick to production subsidies being a better way to subsidies “infant industries” than protection.
Phil H
Aug 26 2021 at 12:35pm
Before you make up hypothetical just-so stories, it’s always a good idea to check if the just-so story has actually happened in real life. Has there ever been an NBA in which 7-footers were not allowed to play? If so, what happened?
Answers: Yes! It’s the WNBA. And what happened? The quality of the women’s game continues to improve steadily, as they become more and more professional.
Obviously, the analogy is only an analogy. It doesn’t tell us what happens in real economies. But if Caplan’s argument is right, then his optimistic scenario is definitely what happens.
Jon Murphy
Aug 26 2021 at 12:54pm
Uh, I don’t think the WNBA forbids tall people from playing. Women are, on average, smaller then men. You’re simply picking up an omitted variable.
That said, even if we assume the comparison between the NBA and the WNBA is appropriate (ie, the only relevant difference is the height), your conclusion doesn’t hold: WMBA games are consistently lower-scoring, lower-attended, and lower-performing. Bryan’s hypothesis is confirmed by the NBA/WNBA comparison.
Vivian Darkbloom
Aug 26 2021 at 1:09pm
On the other hand, the de facto banning of midgets doesn’t seem to have been bad for baseball!
https://en.wikipedia.org/wiki/Eddie_Gaedel
Jon Murphy
Aug 26 2021 at 2:37pm
Probably because, despite the one gimmick, midgets wouldn’t be too useful in Baseball.
Mark Z
Aug 26 2021 at 2:13pm
Was there some point in the past when women were taller, and that was why the players weren’t as good? I would guess if anything the height of women playing in the WNBA has been going up steadily, if only because people in general have been getting taller.
Frank
Aug 26 2021 at 2:32pm
A contemporary version of the Infant Industry Argument relies on a [non-capital] market failure. There must be learning; the learning must be by doing; the benefits of the learning must accrue outside the firm; the tariff must be the only way of fixing the externality; and the tariff must eventually come off.
Accrue outside the firm: Airplane production would not count even though there is great learning by doing, for the workers who learnt could not benefit from the learning if they left the firm [except all of them together].
The conditions for the validity of the argument are stringent. It is difficult to empirically test for them in the numerous cases for which the argument has been invoked. Except for one condition — the tariff must come off! It never does. 🙂
Robert
Aug 26 2021 at 3:23pm
I’m not sure. The heart of the protectionist theory is that a company that is entering as a noob into an established industry needs some kind of edge in order to have a chance at getting established and getting to a size where economies of scale become relevant. Ideally, that edge would be a better product or better manufacturing technology – something compelling enough to help the fledgling successfully fledge. I’ve spent my career as a manufacturing engineer at both established companies and at startups, and I can tell you that entering an existing market with powerful incumbents is extremely difficult – even when the federal government is helping you, as was the case at one of the startups I worked for. I am a longtime reader of Dr. Caplan and have a healthy respect for his intellect. I very often agree with him. But I think that this time the truth is more mixed than he is suggesting.
MarkW
Aug 26 2021 at 5:17pm
Isn’t South America 50+ years into the whole high-tariff / ‘import substitution’ development plan? How’s that been going?
Warren Platts
Aug 27 2021 at 4:58pm
The Latin Americans, unlike the U.S., actually paid attention to English speaking academic, free trade economists in the 19th century. That didn’t work out all that well either.
MarkW
Aug 27 2021 at 5:18pm
Which Latin American countries have pursued a 19th-century style free-market, free-trade policy? Arguably the Chileans (and to their economic benefit) but that era seems to have ended. Which have consistently pursued a free-market, free-trade approach to their detriment?
William Peden
Sep 1 2021 at 3:30pm
If you’re referring to the Argentines, it worked amazingly until they misdiagnosed the problem in the Great Depression and adopted Peronism.
Mark Brady
Aug 26 2021 at 9:58pm
So, Bryan, what did you think of Daniel Hannan’s arguments in favor of free trade? Hannan is not an advocate of unilateral free trade per se.
Tim Worstall
Aug 27 2021 at 7:32am
There’s another way to describe this argument:
“If 400% tariffs raise the price of foreign phones to $1000, however, domestic phone production will launch. And once it does, domestic phone factories’ costs will start to fall.
If you replicate this policy across a vast range of industries, the low-productivity – hence poor – country transforms into a high-productivity – hence rich – country. Yes, tariffs temporarily made the poor country even poorer.”
We’re going to screw currently poor consumers in a poor country so that the currently richer capitalists in that poor country can make out like bandits. All those in favour say “Aye”.
It’s not going to be a terrible surprise that this gains the support of the currently richer capitalists in that country plus the governing class who are – likely as not – their cousins etc.
Robert
Aug 27 2021 at 3:32pm
Big, established incumbents can use their market power to try to squash competition – especially if that competition is small and new.
The argument that there are natural barriers to trade that already protect new entrants into a local market assumes that foreign incumbents aren’t already established in the local market. If they are already there, then those barriers of distance and unfamiliarity have already been overcome.
Industry has momentum. When you’re trying to start up from nothing you’d probably say “inertia” rather than “momentum”, because that’s what you’d be trying to overcome. Wealth and education are products of established industry – but they are also necessary for starting new industry. Chicken or egg.
Comments are closed.