I – II – III – IV – V – VI – VII – VIII – IX – X – XI – XII – XIII – XIV – XV – XVI – XVII – XVIII – XIX – XX – XXI – XXII – XXIII – XXIV – XXV – XXVI – XXVII – XXVIII – XXIX
[Single-page view]

So all right, we’ve laid out the case for free trade and given a lot of reasons why, as a general rule, protectionism is a bad thing. But then again, we’ve also said that economics is all about making tradeoffs and weighing costs against each other – so doesn’t that imply that there must be at least a few good arguments in favor of protectionism? Well, it’s true that arguments in favor of protectionism do exist, as we’ve already seen. Ultimately though, as Taylor explains, most of them just aren’t strong enough to really hold water:

Although most economists are supportive of the forces of free trade, they readily admit that it can cause dislocations in and disruptions to an economy. As a result, political pressure often arises to limit imports. Such steps are known generically as “protectionism,” because the stated intent of laws restricting imports is to protect domestic industries from foreign competition.

There are several ways to implement protectionism. Import quotas put a numerical limit on imports. Tariffs are a kind of tax that raises the costs of imports. Countries may enter voluntary export restraint agreements, which are sometimes not so voluntary; they are instead negotiated under the threat that if one country doesn’t “voluntarily” reduce its exports, the other country will institute a quota or tariff. In the 1970s and ’80s the United States entered this sort of agreement with Japan to restrict Japan’s export of steel to the United States. Finally, there are nontariff barriers, a catchall category that includes any bureaucratic or regulatory process set up, explicitly or implicitly, for the purpose of restricting imports. For example, imagine a hypothetical rule that all television sets imported to the United States must be unpacked and inspected, one at a time, at one warehouse in the middle of Kansas. The costs of time and inconvenience imposed by a rule like that would certainly discourage imports.

A protected industry faces less competition from foreign producers. As a result, it can charge higher prices and earn higher profits. Thus protectionism, in economic terms, is just a way for a government to provide an indirect subsidy to a domestic industry, paid for by higher prices to domestic consumers. Sometimes, as in the case of steel and other raw materials, it’s not individuals who consume the raw goods but other firms who will use them to make a finished good. Still, consumers of the finished good ultimately pay a higher price, as the costs are passed along to them.

Perhaps the most prominent arguments for subsidizing industry are about how protectionism could benefit domestic workers. This argument actually comes in four different varieties, some more persuasive than others: imports might affect the total number of jobs available for domestic workers; imports might affect the average level of wages; imports might create disruption from workers having to change jobs; and imports might lead to greater inequality of wages across the economy, even if the average wage goes up. Let’s look at these arguments in turn.

There’s no question that protectionism is a subsidy that can help retain jobs within a certain industry; however, there is no reason to believe that protectionism increases the overall number of jobs in an economy. To put the same point in reverse, there is zero evidence that international trade diminishes the total number of jobs. One vivid illustration of this occurred when the North American Free Trade Agreement was being discussed in the early 1990s. Presidential candidate Ross Perot, arguing against the treaty, liked to say, “If there’s free trade between the United States and Mexico, there would be a giant sucking sound of jobs heading south.” Well, the North American Free Trade Agreement passed in 1994. It was followed by seven of the best years for job growth in U.S. history. The giant sucking sound never happened.

Economic theory also suggests that international trade has very little to do with the nation’s overall employment level. Cyclical unemployment is tied to recessions and growth, while natural unemployment is tied to labor market incentives. Either way, it’s not about trade. Imagine the extreme case: Would shutting out all imports from other nations solve unemployment? Of course not. On top of that, other countries would retaliate; we would lose export jobs. Other countries wouldn’t want U.S. dollars if they couldn’t sell here; they couldn’t buy exports. Without trade, the overall unemployment rate would probably be much the same as before.

What about protectionism as a way to keep wages high? Protectionism is a subsidy to an industry, so it can certainly help wages in that industry. However, that doesn’t mean higher wages for the economy as a whole. The higher wages in the protected industry are coming at the cost of higher prices for the good, so everyone else pays. Ultimately, wages are going to depend on productivity. If free trade increases productivity, it will also contribute to a gradual increase in average wages over time.

The argument that imports can create disruption between industries by causing domestic industries that compete with imports to lose production and causing those that export to increase production is completely true. Indeed, this sort of disruption is precisely the mechanism through which trade brings economic gains to an economy. But even here it’s important to put the effects of international trade in the context of the typical churning and turmoil of the U.S. economy. Jobs are being gained and lost all the time in the enormous U.S. economy because some firms fail and contract while other firms succeed and expand. Most of that churning in the economy is due to domestic competition, the quality of management and workers in these companies, whether sales of certain products are rising or falling, and other factors that have nothing in particular to do with international trade.

Can protectionism reduce the amount of inequality in an economy? There’s a dispute over how much the rise in inequality of income that happened in the United States between the 1970s and the mid-2000s was caused by trade. The consensus seems to be that globalization does contribute somewhat to inequality, but it’s not the biggest factor. Factors such as how information and communication technology have improved the productivity of high-skilled labor seem more important, along with other factors. […] Globalization is a much smaller cause, in part because so much of U.S. international trade is with other high-wage economies and in part because roughly two-thirds of jobs in the United States aren’t in competition with imports at all. American lawyers don’t compete much with Japanese lawyers. A real estate agent selling a home in New York isn’t in competition with a real estate agent selling a home in London. If you need your car fixed, you’re not going to take it from Florida to Brazil for the job. Many jobs are not in competition with imports and can’t be outsourced to foreign producers. So while trade has some effect on wage inequality, it’s not the main cause. Also, there are better solutions to wage inequality […] than restricting trade.

Some argue that trade has contributed to a growing income gap between the richest and poorest countries of the world. Over the past century, the richest countries of the world have been getting steadily and steadily richer, while [many of] the poorer countries haven’t made much progress at all, so there’s been a divergence in per capita GDP. However, the wealth of the richest countries is not built on keeping sub-Saharan Africa, or parts of India, or western China poor. Those places aren’t poor because of trade; they aren’t much involved in trade. If anything, they’re poor because of a lack of trade. This rising gap in world incomes exists not because globalization is harming the poorest countries of the world, but because they aren’t participating in globalization. The leading success stories in economic development, such as Japan, South Korea, China, and now India, have typically used foreign trade as one of their main engines of growth.

In light of all these points, it seems pretty clear that the case for protectionism is weak at best. Aren’t there any good arguments in its favor, though? Aren’t there any special circumstances in which some degree of protectionism might be justified? Well, as Friedman and Friedman explain, there actually are a handful that seem like they might at least have some merit – but even in these exceptional cases, the pros and cons seem somewhat mixed, so it’s hard to really call any of them a slam dunk:

In all the voluminous literature of the past several centuries on free trade and protectionism, only three arguments have ever been advanced in favor of tariffs that even in principle may have some validity.

First is the national security argument already mentioned. Although that argument is more often a rationalization for particular tariffs than a valid reason for them, it cannot be denied that on occasion it might justify the maintenance of otherwise uneconomical productive facilities. To go beyond this statement of possibility and establish in a specific case that a tariff or other trade restriction is justified in order to promote national security, it would be necessary to compare the cost of achieving the specific security objective in alternative ways and establish at least a prima facie case that a tariff is the least costly way. Such cost comparisons are seldom made in practice.

The second is the “infant industry” argument advanced, for example, by Alexander Hamilton in his Report on Manufactures. There is, it is said, a potential industry which, if once established and assisted during its growing pains, could compete on equal terms in the world market. A temporary tariff is said to be justified in order to shelter the potential industry in its infancy and enable it to grow to maturity, when it can stand on its own feet. Even if the industry could compete successfully once established, that does not of itself justify an initial tariff. It is worthwhile for consumers to subsidize the industry initially—which is what they in effect do by levying a tariff—only if they will subsequently get back at least that subsidy in some other way, through prices later lower than the world price, or through some other advantages of having the industry. But in that case, is a subsidy needed? Will it then not pay the original entrants into the industry to suffer initial losses in the expectation of being able to recoup them later? After all, most firms experience losses in their early years, when they are getting established. That is true if they enter a new industry or if they enter an existing one. Perhaps there may be some special reason why the original entrants cannot recoup their initial losses even though it be worthwhile for the community at large to make the initial investment. But surely the presumption is the other way.

The infant industry argument is a smoke screen. The so-called infants never grow up. Once imposed, tariffs are seldom eliminated. Moreover, the argument is seldom used on behalf of true unborn infants that might conceivably be born and survive if given temporary protection. They have no spokesmen. It is used to justify tariffs for rather aged infants that can mount political pressure.

The third argument for tariffs that cannot be dismissed out of hand is the “beggar-thy-neighbor” argument. A country that is a major producer of a product, or that can join with a small number of other producers that together control a major share of production, may be able to take advantage of its monopoly position by raising the price of the product (the OPEC cartel is the obvious current example). Instead of raising the price directly, the country can do so indirectly by imposing an export tax on the product—an export tariff. The benefit to itself will be less than the cost to others, but from the national point of view, there can be a gain. Similarly, a country that is the primary purchaser of a product—in economic jargon, has monopsony power—may be able to benefit by driving a hard bargain with the sellers and imposing an unduly low price on them. One way to do so is to impose a tariff on the import of the product. The net return to the seller is the price less the tariff, which is why this can be equivalent to buying at a lower price. In effect, the tariff is paid by the foreigners (we can think of no actual example). In practice this nationalistic approach is highly likely to promote retaliation by other countries. In addition, as for the infant industry argument, the actual political pressures tend to produce tariff structures that do not in fact take advantage of any monopoly or monopsony positions.

A fourth argument, one that was made by Alexander Hamilton and continues to be repeated down to the present, is that free trade would be fine if all other countries practiced free trade but that so long as they do not, the United States cannot afford to. This argument has no validity whatsoever, either in principle or in practice. Other countries that impose restrictions on international trade do hurt us. But they also hurt themselves. Aside from the three cases just considered, if we impose restrictions in turn, we simply add to the harm to ourselves and also harm them as well. Competition in masochism and sadism is hardly a prescription for sensible international economic policy! Far from leading to a reduction in restrictions by other countries, this kind of retaliatory action simply leads to further restrictions.

We are a great nation, the leader of the free world. It ill behooves us to require Hong Kong and Taiwan to impose export quotas on textiles to “protect” our textile industry at the expense of U.S. consumers and of Chinese workers in Hong Kong and Taiwan. We speak glowingly of the virtues of free trade, while we use our political and economic power to induce Japan to restrict exports of steel and TV sets. We should move unilaterally to free trade, not instantaneously, but over a period of, say, five years, at a pace announced in advance.

Taylor tackles some of these same arguments, along with a few others:

What about other arguments for protectionism? It’s sometimes argued that new industries, called “infant industries,” need to be sheltered from foreign competition until they build up enough size and expertise to compete in world markets. In theory, this argument makes sense. But in practice, these infant industries often never grow up and become able to compete—and in the meantime, the economy suffers from supporting them. A classic example arose in the 1970s, when Brazil decided to protect its own infant computer industry from import competition. By the later part of the 1980s, the Brazilian computer industry was about ten years behind the times, which in computer years is an eon. This wasn’t a problem only for the computer industry. Think of all the other Brazilian industries that use computers: finance, industry, communications—they were all trying to survive in the world economy with computers that were ten years out of date. An outdated and uncompetitive computer industry was bad enough, but in protecting that industry, Brazil hobbled its other industries as well.

An example of infant industry protection working pretty well occurred in South Korea, where the government subsidized certain industries such as heavy construction equipment manufacture, but if that industry didn’t reach a certain level of international sales within a preset time frame, all subsidies were cut off. Thus, short-term protectionism was accompanied by a predetermined deadline for competing in world markets. That said, for South Korea and other countries in East Asia, the fundamental reason for their economic growth is not infant industry policy, but high rates of investment in physical capital, education, and the adoption of new technology. While these countries protected a few infant industries, they gave even more help to old, aging industries such as agriculture.

Yet another argument given for protectionism is the concern that foreign producers might have an unfair advantage because their nations have lower environmental standards versus the United States, and thus lower production costs. This argument is frail. Environmental costs are only a small share of total costs, maybe 2 percent in most U.S. industries. Also, as countries get richer—which is part of what happens from international trade—they tend to make their environments cleaner. After all, they have more resources to spend on cleaning up their environment. In fact, multinational producers often lead the way in reducing pollution in other countries because they bring pollution-control technology developed in Europe or the United States to their plants in the lower-income countries. The notion that reducing trade would lead to a cleaner environment is deeply misguided.

Another concern over international trade is the problem of predatory pricing, sometimes called “dumping”: the practice of selling below cost to drive out competition, then raising the price once you have a monopoly. This accusation has been hurled at a lot of international competitors in the U.S. market, perhaps especially at producers of imported steel. It’s easy to find cases, such as in the manufacture of cars, steel, or television sets, in which foreign competitors caused U.S. firms great difficulties or even drove them out of business. However, it’s impossible to find a case in which the foreign firms were then able follow up by earning monopoly profits. After all, the foreign producers still had to compete with one another. For example, Japanese cars do very well in the U.S. auto market, but Honda and Toyota still compete fiercely with each other, and with other firms. Dumping, by definition, isn’t just a matter of hurting the domestic producer; if no monopoly charging high prices emerges, no dumping has occurred.

Sometimes people argue for protectionism on the grounds that certain products such as oil are vital to national security, so we shouldn’t rely on foreign suppliers. The logic of this position escapes me. Oil is a vital resource. In this case, doesn’t it make more sense to import as much of it as we can now and stockpile it, rather than using up our domestic resources? Shouldn’t we conserve our own vital resources for when we need them? If the vital product is a new technology, surely it makes sense to have the best available technology here as soon as possible, to learn about it, and to be able to produce it domestically in the future. Moreover, the national security excuse for restricting imports is easily abused. The U.S. government started providing subsidies for mohair producers in the 1950s, on the national security grounds that it was needed for soldiers’ uniforms. In the twenty-first century, we still subsidize mohair, although it has not been used for decades in making uniforms.

There are lots of arguments for protectionism, but few of them are compelling—and for those arguments that are compelling, there are always better ways to address the problem than cutting back on imports.

It’s worth remembering that the world economy experienced a sharp decline in international trade during the interwar years of the twentieth century—that is, between World War I, through the Great Depression, and up to World War II. After that time, governments realized that the constriction of trade was bad for all. So an international treaty called the General Agreement on Tariffs and Trade (GATT) was signed in 1947. In 1995, GATT morphed into the World Trade Organization. Regional free trade agreements such as NAFTA and the European Community exist all over the world; in fact, some people say that regional trade agreements are a spaghetti bowl of agreements, with all the strands reaching back and forth from one country to another. In general, this pattern of trade agreements has been successful. The typical tariff has dropped from 40 percent in the 1950s to about 4 percent today, thus making trade between countries much easier. The mission of these international trade agreements has also expanded to cover trade in services (as opposed to goods), environmental issues, and labor issues.

Countries sign international agreements to support free trade for much the same reason that people join a health club and sign up for exercise classes. Countries know that they will be under constant temptation to lean toward protectionism. There are always going to be certain industries that are particularly challenged by foreign competition and thus hostile to it. Those industries are going to organize and lobby politicians for protectionism. In the U.S. political system, it isn’t uncommon for a well-organized special interest with a large stake in the outcome, such as an industry looking for protectionism, to win out when the costs are spread broadly over an unorganized, if larger group such as consumers. When countries sign free trade agreements, they tie their hands in a way that makes protectionism harder to adopt.

The trend toward a more globalized economy will surely continue, driven by technological developments in communications and transportation that make global economic links easier, by international treaties that reduce legal barriers to trade, and by the surging and internationally oriented economies of China, India, Brazil, and others. Every major economic change brings its own challenges and disruptions, and globalization is no exception. But the overall direction of globalization is to increase the standard of living in the United States, and throughout the world.

Personally, I have to admit, I’m not quite as dead-set against some of these counterarguments as Taylor and the Friedmans seem to be. When it comes to the national security argument, for instance, I agree with most of what they say, but I can also imagine certain cases in which it really might make sense to keep some manufacturers of particular goods up and running in the US on a stable basis, even if it’s not quite as efficient as entrusting the entire supply of those goods to foreign suppliers, just in case of emergency (I’m thinking of things like food supplies in particular). This is something I’ve changed my opinion on fairly recently; I used to think things like agriculture subsidies were a prime example of egregious government waste (why spend valuable taxpayer money to keep domestic farmers in business when it would be cheaper just to buy all our food from overseas?) – but now I’ve started to see a bit more of the logic behind having a functioning food production infrastructure already in place in case of some sudden unexpected crisis. Sure, it might be a slight drag on our economy 99.99% of the time – but in that other 0.01% of cases, like (say) if we’re ever stuck by some deadly super-pandemic that suddenly makes it impossible to import sufficient food from abroad, it seems like we’d be glad to have it. After all, food isn’t something you can just instantly start producing overnight; it takes months to grow and harvest. So although Taylor has a good point that you could have some necessities already stockpiled in advance – including some non-perishable food supplies being strategically held in reserve – it still doesn’t strike me as completely unreasonable to also want to have some operations already in place to produce more food as needed (including farmers who know how to run them), just in case. Now, would the best way of accomplishing this be through tariffs, or subsidies, or some combination of approaches? I have no idea. But I’m open to different possibilities, at least.

What about the “infant industries” argument? Here again, Taylor and the Friedmans make a lot of good points against it; but I think there’s also a reasonably compelling case to be made that it might actually have some validity in certain situations. After all, it may be true as a general rule that a country’s best bet for maximizing its income is to pursue its comparative advantage – but it’s also true that the level of prosperity it ultimately achieves depends quite a bit on what that comparative advantage actually is and how efficiently it can be pursued. As we discussed earlier, more productive work is better paying work – so if a country’s median worker is able to go from producing, say, 10 sacks of grain in an hour to producing 10 smartphones in an hour, the latter will obviously be more lucrative. What that means, then, is that if a country is able to invest in upgrading its capabilities and thereby shifting its comparative advantage from a lower-productivity area to a higher-productivity area, the country as a whole will become richer. (And crucially, by shifting to producing manufactured goods like smartphones rather than primary goods like grain, they’ll have a lot more room for their productivity to keep increasing, since the level of demand for manufactured goods will tend to increase as the population’s income grows, whereas the level of demand for primary goods will tend to stay pretty much the same (see Wendover Productions’ video on the subject here).) Joseph E. Stiglitz puts it this way:

Real development requires exploring all possible linkages: training local workers, developing small and medium-size enterprises to provide inputs for mining operations and oil and gas companies, domestic processing, and integrating the natural resources into the country’s economic structure. Of course, today, [poorer] countries may not have a comparative advantage in many of these activities, and some will argue that countries should stick to their strengths. From this perspective, these countries’ comparative advantage is having other countries exploit their resources.

That is wrong. What matters is dynamic comparative advantage, or comparative advantage in the long run, which can be shaped. Forty years ago, South Korea had a comparative advantage in growing rice. Had it stuck to that strength, it would not be the industrial giant that it is today. It might be the world’s most efficient rice grower, but it would still be poor.

And Heath elaborates:

Naturally, Ricardo [with his idea of comparative advantage] is not the last word on the subject of international trade. It is important, however, that he be given the first word. His analysis of comparative advantage is the bedrock of modern international trade theory. It is simply not possible to have a proper conversation on the subject unless everyone fully understands this theory, along with all the constraints that it imposes upon our thinking about the subject.

That having been said, there are all sorts of interactions in which the harmonious logic of comparative advantage does not prevail. For example, countries compete against one another fairly directly for foreign direct investment. When Toyota needs to decide whether to build a new manufacturing facility in Ontario, Kentucky, or Baja California, it is not misleading to describe the issue in terms of international competitiveness. It is also important to note that comparative advantage is a trickier concept than it sometimes appears to be. When Ricardo presented his original argument, he used the example of England buying wine from Portugal in exchange for cloth. In this case, the fact that it took less effort to grow grapes in Portugal would seem to be a natural consequence of a more favorable climate. This in turn led to the suggestion that comparative advantage arose from conditions that were largely outside of anyone’s control, such as natural resource endowment. While this is sometimes true, often it is not. Knowledge, productive technology, and even organizational forms are not nearly as portable across national borders as they are often made out to be. Portugal remains a major exporter of port wine to this day, not because of climatic advantages, but because of advantages stemming from the knowledge, experience, and tradition that have arisen through centuries of producing this product. There are also significant economies of scale in winemaking that confer an advantage upon the “first mover,” as well as firms that have a large domestic market for their products.

Many of the most important sources of comparative advantage are completely overlooked in public policy discussions. For example, the presence of a large number of native (or highly competent) English speakers is a source of enormous advantage in particular sectors, not just in media and publishing, but also in law, financial services, scientific research, software development, and so on. Local and national culture can create advantages in ways that are very poorly understood. (Silicon Valley, as people are fond of pointing out, does not contain any significant silicon deposits, but it does contain a lot of Californians.) Network effects are important—the presence and success of one industry can generate advantages in related fields, often in very indirect ways. The legal system also confers advantages upon particular industries. The production of so-called intellectual property, for example, thrives only in jurisdictions with a legal environment that offers reasonable protection of patents and copyrights.

As a result, comparative advantage is not just something that countries happen to possess; it is also something that they can actively cultivate. In particular, subsidies to a given industry may create an advantage that is purely artificial, but over time they can lead to the creation of genuine comparative advantage, as the appropriate support networks, training systems, and reservoir of local knowledge needed for the industry are formed. This is, for example, what the government of Brazil is counting on with its support for Embraer (backed by the desire to get a slice of the international market for commercial aircraft), and the United Kingdom with its subsidization of the video game industry. Of course, this sort of political interference is unwise in many respects, but it does not rest upon any sort of misunderstanding or fallacy. It is possible for a country to do very well for itself through a well-planned and well-executed industrial strategy (particularly in “winner-take-all markets,” where the world only needs one or two suppliers). In this respect, the vocabulary of international competitiveness is again not misleading. If a nation has a particular reason for wanting to be an exporter in a particular sector, it may find itself competing with other nations to build up the right sort of advantages for itself.

With this in mind, then, we can see how it might actually make some sense for a country to protect its infant industries on a temporary basis, until those industries can achieve sufficient economies of scale (and sufficient levels of human capital and so on) to compete in the global market. But how can a country ensure that it’s doing so in the right way, and building up industries that actually are ultimately able to compete, instead of just pouring endless funds into infant industries that never grow up and just turn out to be massive wastes? Alexander summarizes Joe Studwell’s take on the whole issue:

East Asian countries got rich by manufacturing. First it was “Made in Japan”, then “Made in Taiwan”, then “Made in China”. At first each label was synonymous with low-quality knockoffs. Gradually they improved, until now “Made in Japan” has the same kind of prestige as Germany or Switzerland, and even China is losing some of its stigma.

Not every rich country gets rich by manufacturing. Studwell divides successful countries into three groups. First, small financial hubs, like Singapore, Dubai, or Switzerland. This is good work if you can get it, but it really only works for one small country per region; you can’t have all of China be “a financial hub”. In the 1980s, everyone was so impressed with Singapore and Hong Kong that they became the go-to models for development, and people incorrectly recommended liberal free market policies as the solution to everything. But the Singapore/Hong Kong model doesn’t necessarily work for bigger countries, and most of the good financial hub niches are already filled by now.

Second, “high-value agricultural producers”. Studwell gives Denmark and New Zealand as examples. Again, these countries are very nice. But they also tend to be small and sparsely populated, and they also don’t scale. New Zealand’s biggest export category is “dairy, eggs, and honey”. Imagine how much honey you would have to eat to lift China out of poverty that way. It would be absolutely delicious for a few years, and then we would all die of diabetes.

Third, manufacturing, eg everyone else. Every big developed country went through its manufacturing phase. Britain, Germany, and America all passed through an era of sweatshops, smokestacks, and steel. Most developed countries gradually leave that phase, switch to a services-based economy, and offshore some of the worse jobs to places with cheaper labor. But they can’t skip it entirely.

And every big developed country that passed through a manufacturing phase used tariffs (except Britain, which industrialized first and didn’t need to defend itself against anybody). Economic planners like Friedrich List in Germany and Alexander Hamilton in the United States realized early on that British competition would stifle the development of native industry without government protection. Once their industries were as good as Britain’s, they removed their tariffs, which was the right move – but they never would have been able to reach that level without protectionism.

Imagine having to start your own car company in Zimbabwe. Your past experience is “peasant farmer”. You have no idea how to make cars. The local financial system can muster up only a few million dollars in seed funding, and the local manufacturing expertise is limited to a handful of engineers who have just returned from foreign universities. Maybe if you’re very lucky you can eventually succeed at making cars that run at all. But there’s no way you’ll be able to outcompete Ford, Toyota, and Tesla. All these companies have billions of dollars and some of the smartest people in the world working for them, plus decades of practice and lots of proprietary technology. Your cars will inevitably be worse and more expensive than theirs. Every country that’s solved this problem and started a local car industry has done so by putting high tariffs on foreign cars. Locals will have to buy your cars, so even if you’re not exactly making a profit after a few years, at least you’re not completely useless either.

This will become a problem if it shelters companies from competition; they’ll have no incentive to improve. Successful East Asian countries avoided this outcome by having many local car companies. The most successful ones went a bit overboard with this:

In the Korea of 1973 – which at the time boasted a car market of just 30,000 vehicles per annum – government had offered protection and subsidies to not one but three putative makers of ‘citizens’ cars: HMC, Shinjin, and Kia. Inasmuch as the market was too small for one producer, the licensing of three companies was ridiculous. HMC posted losses every year from 1972 to 1978, despite very high domestic car prices. However, the government sanctioned multiple car makers not to make shot-term profits – which would have come much sooner to a monopoly manufacturer – but rather to force the pace of technological learning through competition.

In addition to domestic competition, these governments enforced “export discipline”. In order to keep their government perks (and sometimes in order to keep existing at all), companies needed to sell a certain amount of units abroad each year. At the beginning, they might have to sell for way below-cost to other equally poor countries. That was fine. The point wasn’t that any of this was a short-term economically reasonable thing to do. The point was to force companies to be constantly thinking about how to succeed in the “real world” outside the tariff wall. And the secondary point was to let the government know which companies were at least a little promising, vs. which ones were totally unable to survive except in a captive marketplace. If a company couldn’t export at least a few units, the government usually culled it off and gave its assets to other companies that could.

Aren’t there good free-market arguments against tariffs and government intervention in the economy? The key counterargument is that developing country industries aren’t just about profit. They’re about learning. The benefits of a developing-country industry go partly to the owners/investors, but mostly to the country itself, in the sense of gaining technology / expertise / capacity. It’s almost always more profitable in the short run for developing-world capitalists to start another banana plantation, or speculate on real estate, or open a casino. But a country that invests mostly in banana plantations will still be a banana republic fifty years later, whereas a country that invests mostly in car companies will become South Korea. The car company produces a big positive externality – in the sense of raising the country’s level of development – which isn’t naturally captured by the owners/investors. So development is a collective action problem. The country as a whole would be better off if everyone started car companies, but each individual capitalist would rather start banana plantations.

So the job of a developing country government is to try to get everyone to ignore profits in favor of the industrial learning process. “Ignore profits” doesn’t actually mean the companies shouldn’t be profitable. All else being equal, higher profits are a sign that the company is learning its industry better. But it means that there are many short-term profit opportunities that shouldn’t be taken because nobody will learn anything from them. And lots of things that will spend decades unprofitable should be done anyway, for educational value.

[…]

I think there are [strong] counterarguments to Studwell scattered throughout journals that I haven’t quite figured out how to navigate and collect. The infant industry argument seems to be a going controversy within economics and not at all settled science. The picture is complicated by studies showing that countries with lower tariffs have had higher GDP growth since 1945. Studwell could respond that tariffs only work as part of a coherent and well-designed industrial policy; if you just tariff random things to protect special interests, it will go badly in exactly the way free marketeers expect.

To be sure, there are good reasons why economists tend to be so adamant in their insistence that this kind of planned industrial policy shouldn’t be tried under ordinary circumstances. As aspiring communist countries throughout history have demonstrated all too well, steering the course of an entire economy is a very hard thing to get right, to say the least. Still, it does seem like one thing that can make it easier is if the countries in question aren’t trying to chart a whole new path to prosperity, but are simply aiming for known benchmarks in order to catch up to other countries who’ve already become rich successfully. Here’s Alexander again, citing the work of Robert Allen this time:

By the early 20th century, a clear gap had emerged between Europe, North America, and Japan (on one side), and everyone else (on the other). After World War II, the former colonies declared independence from Europe, hoping to try the Standard Development Model at long last and get the same easy successes the West had. But this no longer worked; they had missed the boat entirely. [Allen] invokes the increasing gap between developed and less developed countries; when the gap was still small, the Standard Model prongs were enough to overcome it. By the 20th century, developed countries were so far ahead that the model made less sense. If you’re 1820s France trying to catch up to Britain, you can probably find some craftsmen somewhere in your economy who can make something like a textile mill, train them a bit, get them to make textile mills, use some clever investment policy to create whatever prerequisites to textile mills you don’t already have, and eventually end up with textile mills without too much trouble. If you’re 2000s Bangladesh trying to catch up to the West, you want semiconductor factories. Scrounging around a mostly-agrarian economy and eventually cobbling together enough expertise and capital to make a textile mill is one thing. Making a semiconductor factory is a lot harder. And if you decide to just make the textile mill instead, what if First World textile mills are some sort of amazing robotic wonderland now and nobody wants your crappy 1800s-technology textiles? Development needs a lot more slack now before it can become profitable.

Is it still possible to succeed? Allen points to South Korea, the USSR, and China as examples that it might be. He describes their strategy as “the Big Push” – a strong central government producing lots of (not immediately useful or profitable) industry, in the hopes that it will pay off later:

This is Big Push industrialization. It raises difficult problems since everything is built ahead of supply and demand. The steel mills are built before the auto factories that will use their rolled sheets. The auto plants are built before the steel they will fabricate is available, and indeed before there is effective demand for their product. Every investment depends on faith that the complementary investments will materialize. The success of the grand design requires a planning authority to coordinate the activities and ensure that they are carried out. The large economies that have broken out of poverty in the 20th century have managed to do this, although they varied considerably in their planning apparatus.

There follows some discussion of the Soviet Union and China. Both [took this approach], but the Soviet economy stagnated anyway in the 1970s. Allen seems kind of unsure about why this happened, and is willing to entertain both the possibility it was random and contingent (maybe the planners made a mistake in trying to pour so much investment into parts of Siberia that weren’t really habitable), and the possibility that planned economies are fundamentally better at catch-up growth than at the technological frontier (central planners can force people to make steel mills if you know steel mills are next up on your tech tree, but if you don’t know what’s next on the tech tree it’s hard to plan for it). […] The author is [more] impressed with China, which seems to have gotten this part right (maybe by accident): they communismed until they reached the technological frontier, then uncommunismed in time to get on the path to being a normal developed country. [Allen’s book] isn’t very big on prescriptions, but I think it would probably suggest having a pretty heavily planned economy while you’re playing catch-up, and then unwinding it once you’re close to where you want to be.

I don’t personally claim to know how valid this idea is; even the professional economists, it seems, aren’t exactly unanimous on the matter. Either way though, it’s something of a moot point for those of us living in the US, since so few of our industries are in their infancy compared to the rest of the world anyway; we tend to be at the forefront of most fields, so there aren’t really any industries that would theoretically need to be protected until they “caught up” at all. Of course, for less developed countries, the question isn’t such a moot point; in fact, getting it right might be the key to pulling themselves out of their economic rut and becoming prosperous – or, if they get it wrong, digging themselves even more deeply into it. So it’s definitely an idea that merits attention either way. But regardless of what the right answer ultimately turns out to be, it seems abundantly clear that the wrong answer is to keep domestic industries fully protected on an indefinite basis. If protecting infant industries has ever worked, it’s been because it has involved exposing those industries to domestic competition and export discipline in the long run – in other words, it’s because it has promised to eventually lift the protections. The removal of the protections is the thing that has forced the infant industries to grow up and become efficient enough to survive; so in that sense, the “infant industries argument” isn’t actually an argument against trade at all, but is just another demonstration of why it’s so important. Even the best industrial development plan, it seems, can’t work unless it includes strong components of competition and trade – which is why countries at every stage of development, rich and poor alike, disregard those mechanisms at their own peril.

Continued on next page →