COMMERCIAL POLICYtrade restrictions
free trade and protectionism
theory versus practice
Commercial policy can be defined as a set of all measures that affect directly the amount, composition, and direction of trade flows. It differs from fiscal and monetary policies, which affect trade indirectly via the effect on the exchange rate and capital flows. The set of possible measures is quite wide, but nineteenth-century governments had a pretty small range of options for their commercial policy. They resorted almost exclusively to duties on imports or (less frequently) exports. A duty is a tax that opens a wedge between domestic and world prices: a duty on imports raises the domestic price beyond "world" prices, while a duty on exports lowers it below the "world" prices.
In the nineteenth century, each sovereign country was free to set its own duties (often collected in a tariff). In some cases, these duties were intended to be permanent, but in others they were used as bargaining chips in trade negotiations with other countries. During negotiations, each country tried to lower the partner's duties on its exports and keep as much of its own duties as possible. The failure of negotiation between two states could trigger a trade war, like that between France and Italy from 1887 to 1892. Each "belligerent" state imposed higher duties on imports from its rival until one of them backed down. In contrast, a successful negotiation ended up in a trade treaty, which specified cuts in duties of each partner for a given number of years. On top of this, most treaties included the so-called MFN (most favored nation) clause—that is, the cuts on duties were automatically extended to all other countries. Thus, if successful, bilateral negotiations could yield an outcome not too dissimilar from that of the system of multilateral negotiations in the early twenty-first century (the General Agreement on Trade and Tariffs, or GATT, and its successor). However, the nineteenth century was more fragile, as the overall reduction in duties depended on the decisions of several sovereign states.
By 1789, European states had a long history of trade restrictions tempered only by their inability to control smuggling effectively. Some countries had mitigated the most extreme excesses of mercantilist policies in the late eighteenth century, but still protectionism prevailed throughout the whole Continent, with the notable exception of the Netherlands. The 1800s marked a potentially epochal change, with the establishment of the Continental Blockade (1806) against the imports of British manufactures. For the first time in history, most of the Continent constituted an area of free trade, clearly dominated by France. The experiment, however, did not outlast the fall of the empire, and after 1814 all European states returned to protectionist policies. The next fifty years featured a slow process of liberalization in most countries, with three major turning points. The first was the formation of the Zollverein (custom league) among German states around Prussia (1834), the first step toward German unification. The second was the abolition of protection to wheat (Corn Laws) in the United Kingdom (1846), which was the main barrier to free trade in agricultural goods in Europe. The third was the Cobden-Chevalier Treaty between France and the United Kingdom (1860), which strongly reduced French duties on manufactures. Many other countries followed this example in the next decade, so that the late 1860s and 1870s stand out as a period of almost wholly unfettered trade. The tide had changed already at the end of 1870s, with new tariffs on manufactures in Italy (1878), Austria (1878), Germany (1879), and France (1881). However, the key factor was the fall in prices of wheat, caused by the decline in transportation costs. This "grain invasion" triggered a protectionist backlash. In the 1880s and early 1890s, most European countries protected wheat-growing and approved comprehensive protectionist tariffs, such as the Meline tariff in France (1892). Protectionist policies allegedly prevailed all over Europe until World War I, with few exceptions, such as the United Kingdom, the Netherlands, and Denmark. The structure of tariffs differed among countries, but as a rule raw materials and coal were exempt, while duties were higher than average for three classes of goods: (a) the products of industries whose development the state wanted to foster, often for their own military interest (e.g., iron and steel industry or shipbuilding); (b) the products of industries that the state wanted to shield from foreign competition—such as wheat cultivation; and (c) goods that could not be produced at home, which the state taxed in order to raise revenue (the so-called fiscal duties).
Estimating the level of protection
The traditional narrative is mainly based on anecdotal sources (parliamentary debates, pamphlets, and so on) and on the history of legislation. The former reflect heated contemporary political debates, and thus are likely to be biased, while a list of duties and treaties is hardly sufficient to assess the nature of a trade policy. Unfortunately, to estimate the level of protection is far from easy. The simplest and most widely used measure is the average nominal protection—that is, the ratio between custom revenue and total imports. As expected, the (scarce) data for the first half of the century show a decline, from perhaps 25–35 percent in the 1820s–1830s to a trough at around 5–10 percent in the 1860s and 1870s. Average protection rose again to peak at around 20 percent for the main Continental countries in the early 1890s. The average nominal protection decreased in the 1900s, for the combined effect of new trade treaties and of the rise in prices, which reduced the impact of specific duties. European protection was lower than in the overseas countries and even at its peak, prewar protection was much lower than in the 1930s and even in the 1950s or 1960s. Indeed, only in the 1980s, after half a century of liberalization, did aggregate protection returned to the level of 1914. Thus, prima facie, the data contrast with the traditional story of fortress-like protectionist countries. But a low level of protection is consistent with the great increase in world trade during the second half of the nineteenth century.
Unfortunately, the average nominal protection may not be an accurate measure of the true level of protection. A successful protectionist policy would change the composition of trade, reducing the share of protected goods. A really successful one would block imports altogether, with paradoxical consequences: a country that prevents imports of all goods but one with prohibitive duties would appear less protectionist than another that levies a uniform 5 percent duty on all goods. This is not just a theoretical curiosity: John Vincent Nye has shown that, from the 1820s to the 1870s, average protection in "fortress" France was lower than in the United Kingdom, which raised quite heavy duties on wine and other "fiscal" goods. Thus the average duty is likely to underestimate the "true" level of protection. The bias depends on the sensitivity of imports to increases in prices (elasticity), and thus can differ between countries. Its extent can be assessed using more sophisticated measures: for instance, average nominal protection is a good proxy of "true" protection in Italy. However, computing these alternative measures is quite data-intensive and thus average duty is still widely used.
Protection and economic growth
How did trade policy affect nineteenth-century economic growth in Europe? In theory, the effects of protection are straightforward. First, it changes the structure of the economy. Duties attract factors (land, labor, capital) to the protected sector(s), which would grow, or at least not shrink, under the impact of foreign competition. Thus protection increases the share of protected import-competing goods, at the expense of the production of other imports, of exports, and of nontradable goods (such as services). This change is bound to reduce total production (GDP, gross domestic product) and income or welfare, as free trade yields the optimal allocation of factors. Second, the structural change also affects the distribution of income by factor. In fact, each sector uses a different mix of factors. Under free trade, each country would specialize in productions that use its more abundant—and thus cheaper—factor: a land-scarce, labor-abundant country would specialize in labor-intensive productions as textiles and would import land-intensive products such as wheat. In contrast, protection favors import-competing activities, which use the scarce factor more intensively. These effects are all static, as each change in duty causes a one-off reallocation of factors. But protection is likely to have long-term effects as well. On one hand, less competition from imports reduces the efficiency of domestic producers. On the other, the development of new industrial activities, although fostered with protection, might be beneficial in the long-term. These dynamic gains and losses are often invoked by supporters and adversaries of protection, but their existence is extremely difficult to prove and to measure.
Thus the historical analysis of trade policy must tackle three separate, although interrelated issues: (1) How did it affect the GDP and its overall growth? (2) How did it affect the structure of the economy? and (3) How did it affect the distribution of income? The two first questions concern the effect of trade policy on modern economic growth. For the nineteenth century, this is tantamount to asking how much the state fostered (or hindered) modern economic growth and structural change, as duties were by far the most important instrument available for microeconomic intervention. The third question is important not only for its own sake, but also because potential gains and losses determine the political process of tariff making. It is not surprising that trade policy was very controversial.
Trade policy is quite controversial among historians as well. The debate focuses on its effects on modern economic growth, especially in the second half of the nineteenth century. The period before 1850 is somewhat neglected, partly because data are quite scarce and partly because modern economic growth had not yet started in most European countries.
Most historians tend to have quite a positive view of the effects of protection. Paul Bairoch (1976) argues that the faster growth rate in the Continental countries relative to the United Kingdom is evidence of the success of protectionist policy. This reasoning is not convincing. First, all things being equal, economic growth tends to be inversely proportional to initial income, as backward countries catch up with the more advanced ones. Thus Continental countries were bound to grow faster than the United Kingdom. Second, no matter how high the actual growth rate in GDP was, it could have been even higher with a different trade policy. More broadly, any simple post hoc propter hoc (after this, therefore because of this) inference from a visual inspection of data may be misleading. Modern economic growth is a hugely complex process, which cannot be attributed to one factor only without an in-depth analysis. Since the late twentieth century, economic historians have started to work on the issue, employing theoretically more sound approaches.
The static effect of protection of the allocation of resources can be investigated with the so-called Computable General Equilibrium (CGE) models. They compare the actual level and composition of GDP by sector and the distribution of income with those that would have prevailed if the country had adopted free trade (or any other alternative policy). Most of these models show a substantial effect of protection on the distribution of income. For instance, according to Jeffrey Williamson, the Corn Laws increased the rents of British wheat growers by about a half, at the expense of domestic wheat consumers and of foreign consumers of British manufactures. Kevin O'Rourke estimates that the duty on wheat in the 1880s increased the returns to land by about 10 percent in France and Sweden. In contrast, the effects on the structure of the economy and on total income were not that great: the GDP of a free-trade Italy in 1911 would have been 2 to 3 percent higher than the actual one. These estimates are quite precise, but results depend on the assumptions of the model about the mobility of factors across sectors. Furthermore, by their nature, the CGE models offer only a partial view, as they measure only static effects of protection from the allocation of resources.
The dynamic effects could be captured by the other approach, the econometric estimate of the effect of protection on the growth rate of a group of countries. This approach improves on the "simple" post hoc ergo propter hoc inference, à la Bairoch, as the effect is estimated as the net of other determinants of growth, such as the level of development of the country, the amount of human and physical capital, and so on. Contrary to theoretical predictions, and to the evidence for the period after 1950, some scholars find a positive relation between tariffs and rate of growth before 1914. This positive effect seems to be much stronger for rich overseas countries than for European ones and, among these latter, for the core countries of northwestern Europe. In contrast, tariffs hampered growth in the European periphery. There are several possible explanations for these results. Perhaps tariffs on industrial goods accelerated the transfer of manpower from low productivity sectors (agriculture) to high productivity ones (manufacturing), or gave domestic producers breathing space to develop. Or perhaps the result simply reflects errors in measurement of protection using average tariffs.
Thus progress in recent times has been substantial, but the economic analysis of the effects of nineteenth-century trade policy is still far from providing a clear answer to the three questions. Much more work, for additional countries, is necessary. One can tentatively conclude that before 1913 trade policy had much smaller effects, for good or bad, on modern economic growth in European countries than has been assumed on the basis of the contemporary debate. Protection was not so high, and distortions, although substantial in some specific cases, were not huge. This conclusion is confirmed by a 2003 work by Antoni Estevadeordal, Brian Frantz, and Alan M. Taylor. They estimate that world trade would have been a third higher without tariffs in 1870 and in 1913 (note that their measure of protection does not change much between the two dates). But this conclusion is provisional. Overall, one must not forget the political consequences of protectionist propaganda. It depicted world trade as a deadly fight among countries and foreign goods as invaders that were jeopardizing the very livelihood of the population. This propaganda undoubtedly contributed to the breakdown in international relations that led to World War I.
By and large, protectionist trade policies prevailed in nineteenth-century Europe, although protection was mild compared with the 1930s or the 1950s. The short interlude in the 1860s and 1870s and the free-trade policy of the United Kingdom and of other "minor" free-trade countries were exceptions that seem to confirm the rule. This practice contrasted with the almost unanimous opinion of professional economists, who regarded free trade as highly beneficial. This gap between theory and practice started to open in the eighteenth century. Until then, with very few exceptions, economists had believed that the wealth of a nation depended on its stock of gold, which had to be accumulated by running a positive trade balance, that is, by exporting more than importing. This mercantilist ideology was seriously challenged only in the 1750s and 1760s. Enlightenment thinkers such as the Italian Ferdinando Galiani (1728–1787) or the French Physiocrates advocated freedom in grain trade, and the economist Adam Smith (1723–1790) made a compelling case for overall liberalization in his Wealth of Nations (1776). Free trade is first and foremost a straightforward extension of the principle of the invisible hand—the market knows better than the state how to allocate resources. Furthermore, free trade increases the size of the market and hence the scope for the division of labor and specialization, which in Smith's view is the source of long-term economic growth. David Ricardo (1772–1823), in Principles of Political Economy and Taxation (1817), and other "classical" economists such as James Mill (1773–1836) and Sir Robert Torrens (1780–1864) sharpened Smith's argument. Specialization is not driven by absolute advantage (that is, by the comparison of costs between domestic and foreign producers) but by comparative advantage (that is, by the comparison of production costs among alternative uses of available factors). A country can gain from trade even if it could produce the imported good at a lower cost, provided that it could produce something else at even lower relative costs. Therefore, all countries have something to gain from free trade. The principle of comparative advantage is still the cornerstone of trade theory. Economists have later found some exceptions, such as Robert Torrens's (1780–1864) terms-of-trade argument for duties by large countries, Graham's external-economies argument for protection of industries with increasing returns, or James Brander's and Barbara Spencer's late-twentieth-century strategic-trade argument for export subsidies. However, these exceptions hold true only under very special circumstances: in almost all cases, free trade would deliver the optimal allocation of existing resources, given the available technology. In contrast, most arguments for protection focus on its alleged dynamic benefits. They hold that short-term losses from protection can be outweighed by long-term economic and/or geopolitical gains. The economic gains consist in the full development of the potential of the country. Alexander Hamilton (1755–1804), U.S. Secretary of the Treasury (1789–1795), and John Stuart Mill (1806–1873), the most famous British economist (and a staunch pro-trader), argued that a limited period of protection could be necessary to start potentially suitable industries (infant-industry argument). These industries need some time to acquire necessary technical and organizational capabilities, train employees, and so on, before being able to withstand foreign competition. The geopolitical argument for protection assumes that competition among states requires military capabilities, which must be obtained even at the cost of welfare losses. A great power should be able to produce all goods necessary to wage a victorious war, including, of course, its food. In this case, unlike in the infant-industry argument, protection could be permanent. This line of reasoning can be traced back to The National System of Political Economy (1841) by George Friedrich List (1789–1846). It was quite popular among nationalist writers in the nineteenth and early twentieth centuries. These "dynamic" arguments for protection were and still are fairly diffused among the lay public, but by and large they have failed to convince economists, who doubt that protection could ever be temporary and stress that competition from imports is a powerful stimulus for improving the efficiency of domestic producers.
How is it possible to account for the gap between theory and practice? Why has protectionism been so often adopted if it is harmful to overall welfare? One can suggest three possible answers: the need for state revenue, international relations, and the pressure from producers' lobbies.
The need for state revenue
Duties can yield revenues only if imports do not fall, that is, if they are not substituted by domestic production. This is the case of duties on goods, such as wine in the United Kingdom, that cannot be produced at home. These duties were allegedly only for fiscal purposes—although they could also guarantee some protection to domestic producers of competing goods (such as beer). But also, openly protective duties could yield substantial revenue if domestic production is insufficient for the desired consumption at the prevailing price, inclusive of the duty. This fiscal motivation was particularly important in the nineteenth century, because, before the introduction of personal taxation, custom duties were in fact a major source of revenues. On the eve of World War I, they accounted for about 10–15 percent of total revenues and up to 45 percent for federal states, such as Germany. Thus, the need for revenues could be invoked to justify otherwise unpalatable increases in protection. This happened in Italy in 1894, when the duty on wheat was increased by 50 percent during a budget crisis. In the following decade, duty on wheat provided between 3 and 4 percent of total revenues of Italy.
The prevalence of protectionist policies may be associated with the lack of a "hegemonic" power, such as the United States after 1950. A "hegemonic" power would force other countries to adopt the trade policy that best fits its interests—that is, almost always, to open their markets to its own exports. In this vein, one might explain the liberalization in the first half of the nineteenth century with the rise of British political hegemony and the return to protection toward the end of the century with its decline. However, in Bargaining on Europe (1998), Peter Marsh strongly downplays the British role in the establishment of the network of trade treaties that was instrumental in the liberalization of the 1860s; if anything, France was in the lead in this respect. Similarly, it is difficult to attribute the return to protection since the late 1870s to the lack of a "hegemonic" power. If the system of international relations dampened the protectionist reaction via the network of trade treaties and the MFN (most favored nation) clause. Foreign policy motivations sometimes did influence negotiations for treaties, either speeding them up (Great Powers had some clout versus smaller countries) or retarding them (for instance, the breakdown of negotiations between France and Italy in 1887 was a step in a long-term realignment of Italian alliances toward Germany). However, in the overwhelming majority of cases, trade treaties were determined by demand from domestic exporters. They needed access to foreign markets, which had to be granted with reciprocal concessions with other trading partners.
The lobbying by domestic producers of import-competing goods (the political economy of tariffs) is the most frequently quoted reason for protection. Interests can organize by sector (the iron and steel industry, wheat growing) or by factor (land, labor, capital).
In the latter, most common, case, the lobby allegedly represents all employees of the sector. Lobbies are easier to organize and more effective when the number of potential members is low, as argued by Mancur Olson in his seminal bookThe Logic of Collective Action (1971). Therefore organizations of consumers potentially interested in free trade have long been weak relative to producers' lobbies. Indeed there were no associations of consumers in nineteenth-century Europe, while producers' lobbies, such as the German landowners Bund fur Landwirtschaft (established in 1893), were quite well organized and influential. According to a well-established tradition, in many Continental countries, some producers' lobbies dictated trade policy at the expense of consumers and of other, weaker producers. The return to protection in the 1880s in Germany and Italy, for example, was the outcome of a bargain between cereal-growing Junkers (landed aristocracy of Prussia and eastern Germany) and heavy industry. Wilhelmine Germany was called the empire of rye and iron, although small-scale livestock producers also gained from protection. France also followed this pattern, although bureaucrats there had more power in the implementation of guidelines and, most crucially, in the negotiation of treaties than in Italy or Germany.
The United Kingdom is the prime example of the alternative model of organization of interests. Parties represented factors of production, such as land (Tories), capital (Whigs), and labor (Labour), and thus trade policies were part of their agendas. Actually, the issue played a major political role on two occasions. In the early 1840s the Anti–Corn Laws League, organized and funded by Manchester cotton industrialists, waged a strong campaign against the duty on wheat. It succeeded because the Tory Party split on the issue and lost power for a long period of time. Trade policy resurfaced as a major political issue in the late 1890s, when the movement for fair trade campaigned for reciprocity (that is, to raise duties on imports from countries that taxed British imports). This proposal became the main point in the Tory manifesto for the 1906 elections, but the party was soundly defeated.
Daniel Verdier argues that these national differences in the organization of interests reflect the features of the political systems, such as the loyalty of parliamentarians to their own party and the interest of voters in trade policy. However, these characteristics, especially the latter, depend on economic features, most notably the mobility of factors among sectors. A specialized worker has little scope for changing sectors and thus is more likely to join an industry lobby than is an unskilled one. In contrast, factors that are able to move between sectors are more likely to gather in major parties.
It is thus likely that all three factors—international relations, the need for revenue, and lobbying—contributed to shape trade policy in the nineteenth century. Their relative importance changed by country and by period, but more work is need to assess how.
Bairoch, Paul. Commerce exterieur et développement économique. Paris, 1976.
——. "European Trade Policy." In The Cambridge Economic History of Europe, edited by Peter Mathias and Sidney Pollard. Vol. 8, 1–160. Cambridge, U.K., 1989.
Brander, James, and Barbara Spencer. "Export Subsidies and International Market Share Rivalry." Journal of International Economics 18 (1985): 83–100.
Conybeare, John A. C. Trade Wars: The Theory and Practice of International Commercial Rivalry. New York, 1987.
Estevadeordal Antoni, Brian Frantz, and Alan M. Taylor. "The Rise and Fall of World Trade, 1870–1939." Quarterly Journal of Economics 118 (2003): 359–407.
Federico, Giovanni, and Antonio Tena. "Was Italy a Protectionist Country?" European Review of Economic History 2 (1998): 73–97.
Federico, Giovanni, and Kevin O'Rourke. "Much Ado about Nothing?: The Italian Trade Policy in the Nineteenth Century." In The Mediterranean Response to Globalisation before 1950, edited by Sevket Pamuk and Jeffrey G. Williamson, 269–296. London, 2000.
Foreman-Peck, James. A History of the World Economy: International Economic Relations since 1850. New York, 1995.
Lake, David. Power, Protection and Free Trade: International Sources of U.S. Commercial Strategy, 1887–1939. Ithaca, N.Y., 1988.
Marsh, Peter. Bargaining on Europe: Britain and the First Common Market, 1860–1892. New Haven, Conn., 1998.
Nye, John Vincent. "The Myth of Free-Trade Britain and Fortress France: Tariffs and Trade in the Nineteenth Century." Journal of Economic History 51 (1991): 23–45.
O'Rourke, Kevin. "The European Grain Invasion, 1870–1913." Journal of Economic History 57 (1997): 775–802.
Williamson, Jeffrey. "The Impact of the Corn Laws Just Prior to Repeal." Explorations in Economic History 27 (1990): 123–156.