Protectionism: A Fairy Tale of the State and War

Protectionism: A Fairy Tale of the State and War

Add to Reading List
Add to Reading List
By Mark Harrison

John Aziz writes in defence of protectionism, that is, the use of taxes and subsidies to shield a country's economy from foreign competition. He begins from the Ricardian story of the benefits of two countries sharing the benefits of free trade based on comparative advantage.

"England was good at producing wool, Portugal wine, so they trade and both are better off. There is the fairy tale about how because market transactions are always voluntary and always beneficial that trade, being simply a market transaction across borders, is always win-win."]

But this, he says, is a "fairy tale." In real life, he argues, comparative advantage has little to do with resource endowments and is generally artificial. Comparative advantage may be part of the historical pattern, he concedes. But it misses something essential. What's missing? He goes on.

"Let's imagine a model with two different goods, say, guns and butter. England specialises in producing guns and munitions, and Portugal in butter and agricultural produce. For years, they trade and enjoy the benefit of maximising output through specialisation. Then, England starts a trade dispute with Portugal. They cease trading. England loses access to butter and various agricultural products from Portugal's large population of butter-producing cows, having to replace Portuguese butter with lower-quality and higher-priced Welsh butter. Portugal, however, loses access to guns and munitions. Although this is immediately recognised as a risk to national security, and Portugal quickly tries to start up its own domestic firearms industry, the trade dispute escalates into full-blown war and with their geostrategic advantage in guns, England swiftly triumphs and occupies Portugal."

The implication is clear. Portugal should have insured itself against the contingency of conflict with England by limiting trade through protectionism. By means of an interventionist industrial policy, Portugal would have developed its own guns and could then have resisted England.

Several things are noteworthy about this argument.

First, it too is a fairy tale. As John Aziz rightly points out, there's nothing intrinsically wrong with that. All our models are fairy tales. The point is that some are useful and others not. How can we tell? We test them against stuff that has actually happened. If they survive the confrontation, then we can use them to suggest practical implications for the future. So, the fact that it's a fairy tale is of interest, but it's not a problem. Let's move on.

Next noteworthy point: Let's test this model against something that actually happened. Not literally, because England has not been to war with Portugal since long before gunpowder came along. Replace Portugal by Germany, however, and the fairy tale suddenly acquires an ominous ring of truth. Doesn't it have an uncanny fit with what happened in 1914?

No, not exactly. In 1914 Germany and Britain went to war. While Britain was the pioneer of free trade, Germany had practised protectionism since 1879. German tariffs limited trade and promoted self-sufficiency in both industrial and agricultural goods. In Britain, by contrast, free trade accelerated the decline of agriculture and maximized the exposure of the British economy to imports. In 1913 at least 60 percent of the calories used at home for human consumption were imported. Many observers thought that left Britain ridiculously vulnerable to wartime blockade. German naval strategists agreed. It's true that in World War I food became a weapon of war just as much as guns.

Yet in the outcome it was Germany, the protectionist power, that struggled to manage the wartime disruption of trade and saw civilians die of hunger, while the British got by without serious shortages.

What explains this turnaround? As Mançur Olson (1963) argued (and before him Friedrich Aeroboe), the German economy entered the war in 1914 already weakened by protectionism. Food tariffs had encouraged peasant farmers to stay on their farms. This kept a large subsistence agriculture in being and reduced productivity and incomes. Because German farmers were already well into diminishing returns it was then hard to increase output at need, when war broke out.

As for industry, because imports of food into Germany were restricted in peacetime and labour held back in agriculture, German urban employers faced higher wage costs. To compensate for higher wage costs, industrial firms economized on labour and pushed up productivity. But across the economy as a whole, efficiency and average incomes were reduced.

A history of protectionism gave no national advantage in either World War. I've argued (in several places; see Harrison 2012 for example) that the main factors that gave systematic advantage were a country's size and wealth, and the main source of disadvantage was a peasant-based agriculture.

In that case, what's protectionism all about? In understanding protectionism, redistribution is much more important than development. Whether tariffs and subsidies raise or lower long-run growth, in the short run they redistribute income away from consumers and exporters to import-competing firms, often by very large amounts. This should draw attention to their political significance. As Dani Rodrik (1995, p. 1470) once wrote:

"Saying that trade policy exists because it serves to transfer income to favored groups is a bit like saying Sir Edmund Hillary climbed Mount Everest because he wanted to get some mountain air."

In history, protectionism has given politicians a powerful instrument to bind those "favored groups" into their projects. To Bismarck, protectionism was political: it brought together the interests of "iron and rye" to share rents and support Germany's "peaceful rise." Similar motivations lie behind most real-life experiments in protectionism that I am familiar with. The only real exception is the Soviet experience of autarkic industrialization; that was different because Stalin was an absolute dictator who ruled by fear and had no need to pay off campaign funders.

Modern promoters of the developmental state (including Dani Rodrik) could reply that they advocate only those selective interventions that are designed to improve social welfare, not corrupt the political process. That's an argument I understand, but it requires a benevolent, far-sighted government with the power to intervene and the self-restraint to do so only for the common good, not for the good of its supporters. That's a bit of a problem. I don't see a political system anywhere, short of totalitarian dictatorship, in which you could advance those policies and see them implemented without vested interests jumping on your bandwagon and hijacking it for their own purposes, which will have nothing to do with social welfare.

(It's ironic, then, that John Aziz lists "graft and corruption" as a problem of trade liberalization, because opportunities for corruption are created only where the government has something to withhold.)

The historical link between protectionism and aggressive nation building is strong. Using data for 1950 to 1992 Erik Gartzke (2007) has shown that restricting a country's trade and capital flows is a good predictor of its propensity to engage in conflict. From data for 1865 and 1914, Patrick McDonald and Kevin Sweeney (2007) have shown that protectionism was a robust precursor of engagement in "revisionist" wars.

John Aziz concludes with a warning:

"China's monopoly on rare earth metals which have very many military applications may have national security implications for other nations including Britain and the United States whose ability to manufacture modern military equipment might be impeded by a trade breakdown."

Shouldn't we worry? Yes, but that's because we need to understand China, not because we should be preparing for war. Indeed, one of Mancur Olson's key conclusions was that it's a mistake to think of particular raw materials, and even oil or food, as in some way "strategic" or "essential." Only the final uses of resources are essential. In practice, if some particular material suddenly becomes scarce, the price goes up and and opportunities present themselves to economize at the margin or find alternative sources or substitutes.

The price goes up, it's true. In other words, the alternatives may be costly. But the richer you are, the more easily you can meet the cost. That's why rich countries survive trade disruption and win wars. As for protectionism, to the extent that it diverts resources from their best uses, it makes the country poorer in advance and so less able to afford the measures that might become necessary in a national emergency.

Which brings me to the last noteworthy point about the arguments that John Aziz makes: They have nothing to do with personal well being. As he correctly comments:

"The relative value of outcomes is simply a matter of one's criteria."

In truth, the two fairy tales that he tells differ in addressing completely different criteria. The free-trade fairy tale always was and is about the personal welfare of all members of society. Here, society is global: when trade is free, all gain, not just the residents of one country. The protectionist fairy tale, in contrast, is about nation-building and facilitating conflict in a world where elite coalitions build states, states compete for power, and a gain for one country is a loss to others.

The world is a complicated place. In the same spirit as John Aziz when he notes that the free trade story has some merit, I'm going to accept that the unregulated interaction between real world economies sometimes creates losers. There have evidently been historical circumstances when protectionist policies accidentally did no harm, or even did good by accidentally correcting some market failure.

But the design of protectionism has generally been far more driven by vested interests and power building than by concern for social welfare. Those who enter themselves in the reckoning against free trade often rely on an idealized understanding of the record.


  • Gartzke, Erik. 2007. The Capitalist Peace. American Journal of Political Science 51:1, pp. 166-191.
  • Harrison, Mark. 2012. Pourquoi les riches ont gagné: Mobilisation et développement économique dans les deux guerres mondiales. In Deux guerres totales 1914-1918 − 1939-1945: La mobilisation de la nation, pp. 135-179. Edited by Dominique Barjot. Paris: Economica, 2012 (here's a preprint in English).
  • McDonald, Patrick J., and Kevin Sweeney. 2007. The Achilles’ Heel of Liberal IR Theory? Globalization and Conflict in the Pre-World War I Era. World Politics 59:3, pp. 370-403.
  • Olson, Mançur. 1963. The Economics of the Wartime Shortage: A History of British Food Supplies in the Napoleonic War and in World Wars I and II. Durham, NC: Duke University Press.
  • Rodrik, Dani. 1995. Political Economy of Trade Policy. In Handbook of International Economics, vol. 3, pp. 1457-1494. Edited by Gene Grossman and Kenneth Rogoff. Elsevier.

This article originally appeared on Mark Harrisson's blog.


Keep up to date with the latest thinking on some of the day's biggest issues and get instant access to our members-only features, such as the News DashboardReading ListBookshelf & Newsletter. It's completely free.


Please read our Community Guidelines before posting
Mark Harrison

These are thoughtful points. Here’s one we can agree on: crony capitalism is a huge problem. More generally, capitalism is the first (and so far maybe the only) system to separate the economy from politics. Before capitalism, the landowner was also the ruler. When we talk about crony capitalism, what we mean mostly is that the wealthy and the powerful have got back too close to each other, and deals are made behind closed doors for the powerful to defend the privilege of particular firms and interests. That undermines the logic of a market economy, where the market decides which firms and interests should prosper. Of course we don’t live in a perfect world so we make compromises. For example, we shade the logic of the market by government sponsored welfare arrangements that usually aim to protect people who suffer when their firm or industry suffers a setback, but without protecting the particular firms or industries themselves. When the state protects particular firms because the politicians and the corporate owners have a secret deal going, we call it crony capitalism.

Here’s a general point where John and I differ:

Given the risks of bringing the rich and the powerful together in one room, is it a good idea to give politicians the power to decide who gets rich in the hope it will spread wealth and welfare? John thinks there are circumstances when that can work; maybe, but I am noticeably more sceptical.

Some more specific points of disagreement:

Does free trade correspond with the interests of crony capitalism? I would say no. Since 1945 America has tended to promote free trade (although with major backsliding at times). If America is an example of crony capitalism, that’s a problem for my argument. But is it? John cites the American military industrial complex as a textbook case of crony capitalism. Well, let’s look at some detail. During the crisis the Pentagon has needed to make the best use of its shrinking tax dollars, which can mean buying imported goods if they’re cheaper. Pressure to “Buy American” has been coming from Congressional lobbies and defence contractors. So in fact the public officials are not colluding with what the cronies want. The cronies are unable to secure the fix they want behind closed doors. The pressure they’re bringing is out in the open and freely reported in the media. On this key issue, therefore, the “military-industrial complex” does not look very solid; maybe calling it a “complex” is a bit of a misnomer.

John singles out Halliburton as an example of crony capitalism, because of its links to Vice President Dick Cheney. Here I’ll add that a bunch of authors (Fisman, Fisman, Galef, Khurana, and Wang 2012) did a study of the stock valuation of companies connected with him through the years before and during the Bush presidency. This is the abstract:

“We estimate the market valuation of personal ties to Richard Cheney. Our proxies for personal ties are based on corporate board linkages that are prevalent in the network sociology literature. We consider a number of distinct political and personal events that either affected Cheney’s political fortunes or his ability to hand out political favors. Specifically, we consider: (a) market reaction of connected companies to news of Cheney’s heart attacks; (b) market reaction of connected companies to Cheney’s being placed in charge of the vice-presidential search process and his surprise self-appointment; (c) correlation of the value of connected companies with the probability of a Bush victory in 2000; and (d) correlation of the value of connected companies with the probability of war in Iraq. Contrary to conventional wisdom, we find that in all cases, the value of ties to Cheney is precisely estimated as zero. We interpret this as evidence that U.S. institutions are effective in controlling rent-seeking through personal ties with high-level government officials.”

Elsewhere, the military-industrial complexes that work well in the exercise of collusive economic and political power are in countries like Pakistan, Egypt, and Iran, where the armed forces have major investments in the productive economy and the value of these investments would be threatened by economic liberalization. There the rich and powerful, who are pretty much the same people, exercise power behind the scenes and use it to limit competition and maintain the market distortions that keep them in profit. There might be elections, or there might not; either way, it doesn’t much matter because the army is in charge behind the scenes. In other words, crony capitalism in poor countries tends to be protectionist, because otherwise the cronies might have to compete for a living.

Reference: Fisman, David, Raymond J. Fisman, Julia Galef, Rakesh Khurana, and Yongxiang Wang. 2012. Estimating the Value of Connections to Vice-President Cheney. The B.E. Journal of Economic Analysis & Policy: Advances 13(3), Article 5.

John Aziz

There are some thoughtful points here. The most important one, I think, is the problem of graft and corruption in protectionism; of governments interfering with trade to help and protect cronies, rather than to try to ameliorate market failures.

This is an important point — and the historical record clearly shows that protectionist governments often act in the interests of well-connected cronies — but the flipside of the coin is that what is known as "free trade" often hands out contracts and subsidies to crony corporations and well-connected insiders. The most obvious example being the military-industrial complex that has the American military operating as a kind of global watchdog keeping trade flowing (and those who might wish to disrupt trade like pirates, or terrorists or rogue states at bay) with a global naval presence and over 750 bases outside the US. Also, the contracts that developing countries often receive for infrastructure development as part of free trade deals often hand construction rights or resource-development rights to well-connected crony corporations like Halliburton, etc. There is also a great irony here — so-called "free trade" is built upon subsidised industrial policy!

So cronyism is possible both within the context of a "free trade" system, as well as within the context of a protectionist system. So while you're absolutely correct that "vested interests jumping on your bandwagon and hijacking it for their own purposes" is hard to avoid, it seems to be more of a problem of the nation state and of privately-financed democracy than of protectionism. To me, that means that some correction of perceived market failures can be acceptable under some circumstances even if we should assume that free trade is the standard we should aspire to for most markets. For, example while the automotive jobs may never come back to Detroit, some industrial policy (maybe building solar panels or wind turbines instead of cars) is acceptable to lower the long term unemployment rate there.

As for your point about the empirics of "free trade Britain" vs "protectionist Germany", I think there are too many other variables to this equation to draw any concrete conclusions. Britain was an island with a global empire, while Germany wasn't and that distinction stems from long before 1879. If anything the 35 years of protectionism and industrial policy leading up to World War 1 for Germany allowed it to achieve an industrial base, and national status to challenge and rival Britain. The worrying aspect of that for me is when you rightly point to Gartzke (2007) suggesting that protectionism is often itself a prelude to war. I was suggesting that a small amount of protectionism can be a good insurance policy against war — although I will gladly accept that being a wealthy country is also a very good insurance policy itself — but if a country were actively seeking to prepare for war and to challenge the world order, then building a base of domestic industrial strength would be the way to go about it. Or, perhaps the industrial strength and independence that industrial policy has built in the past created a certain arrogance among protectionist powers.

In conclusion, while I am in favour of free trade for most things, I do think subsidised industrial policy can sometimes be helpful and necessary. We live in a world where commerce and exchange have dramatically improved the world's ability to easily satisfy wants and needs. Yet that "free trade" is built upon the back of subsidised industrial policy! And even in a relatively "free trade" world, there are market failures and dislocations — labour market breakdowns, trade disruptions etc — that it is prudent to collectively insure against (and which to some degree we already do insure against).

A good article, but the Author forgot to explore the importance of retaining critical and basic skills, which protectionism does ensure, which enable innovation and flexibility in structural economic change.

Twitter Feed