Part VII. Foreign Trade Was Not an Engine of Growth

The theorist of foreign trade Ronald Findlay and the economic historian Kevin O’Rourke collaborated in 2007 in a magnificent history of world trade since 1000 C.E.³⁸ There is much to admire in the book, in particular its cosmopolitan sweep. Findlay and O’Rourke are nothing like Eurocentric, and think big.

When they come to the Industrial Revolution, however, their arguments don’t work too well. For example, they criticize static models about the matter because static models “cannot, by definition, say anything about the impact of trade on growth.”³⁹ That’s a trifle overstated. Static models have been shown to be inadequate to explain the greater part of modern economic growth, so large is the thing-to-be-explained. The showing has not been achieved by “definition.” It has been achieved by finding that static gains are not of the right order of magnitude to do the scientific job. It is an empirical, scientific finding of the past fifty years of work on the subject, not a mere definition. (Definitions, though, are not to be scorned as historical tools, of course — as for example in the definitions of national income or the share of foreign trade that permits the showing of the smallness of the static gain.) Findlay and O’Rourke themselves use static models of demand and supply a few pages earlier to make the correct point that Britain shared its gains from trade with its trading partners 1796 to 1860 by increasing supply of its exports much more rapidly than demand grew, turning the terms of trade against itself. It is an old and good point (I made it myself a long time ago), and it is definitely “static” and definitely says a great deal about the impact of trade on growth.⁴⁰

Considering that the static effects alleged so widely for trade as an engine of growth are small, the non economists, and some of the economists, are likely to claim that “dynamic” effects will rescue the engine. Possibly. The word “dynamic” has a magical quality — the economist Fritz Machlup once placed it on a list of “weaselwords.”⁴¹ Waving “dynamic” about, though, does not in itself suffice to prove one’s economic and historical wisdom. One has to show that the proffered “dynamic” effect is quantitatively strong. An existence theorem in a model without magnitudes — which is the usual and unscientific routine in high-brow economics — will not do any scientific job.

For example, one might claim that the industries like cotton textiles encouraged by British trade were able to exploit economies of scale, in perhaps the making of textile machinery or the training of master designers. There: a dynamic effect that makes trade have a larger effect than the mere static gain of efficiency. But the assertion is without quantitative oomph. The rejection of trade as the engine of growth is reinstated. Or the profits from overseas trade were invested (I say again: were not the profits from house-building and retail trade reinvested, too?), and so capital accumulation was increased. But is the dynamic effect of reinvestment large? It seems not, as Guillaume Daudin has concluded for mercantilist France before the Revolution.⁴²

Or again a smaller cotton textile industry would have been less able to take advantage of such technological change nationally. After all, cotton was unusually progressive. One can answer the question posed by a thought experiment. The experiment requires that one know productivity change in various industries other than cotton textiles. Remember that the pattern of productivity in British industries can be calculated by looking at what G. T. Jones in 1933 called “real cost,” that is, the price of, say, iron bars relative to, say, coal and wages. The pattern was something like this, using Harley’s revision in 1993 of my table in 1981 (I am accepting for the sake of argument the view of the Two Nicks that total growth was small in the 1700s, and therefore their implication from my calculation of residuals that productivity change outside the named sectors was vanishingly small):

Crude Approximations of Productivity Change by Sector, U.K., 1780-1860
Sector	Yearly productivity growth % per year	Value of output divided by national income	Contribution to the national annual growth of productivity
Cotton	1.9	.07	0.133
Worsteds	1.3	.035	0.0455
Woolens	0.6	.035	0.021
Iron	0.9	.02	0.018
Canals & railways	1.3	.07	0.091
Agriculture	0.7	.27	0.12
All others implied as residual	[0.02]	.85	0.02

Source: Harley 1993, Table 3.6, p. 200, based on McCloskey 1981, p. 114, where the details of the original calculation and accounting are given. A little oddly, Harley leaves my estimates of income shares unimproved. That the shares add up to more than 1.00, by the way, is not an error. It is implied by the taking of productivity measures from gross costs (as against merely value added, which would not give a correct measure of savings on material inputs from other sectors).

Suppose the cotton textile industry were cut in half by an exclusion from foreign markets. (It is a dubious counterfactual because in the eighteenth century Manchester was anyway the best place in Europe to produce cotton cloth. It earned, to put it the way economists do, “rent,” which is just another way of saying it was the low-cost location for the task. And so you have to assume that mercantilism would take the form not merely of taxing Manchester with French or Dutch tariffs but partially shutting down its activities, for no gain to anyone — though admittedly it would not be the first or last time that such an irrational policy had been implemented.) During 1780 1860 therefore the share of cotton in national income would have been 3.5 percent of national income instead of its actual 7 percent. The 3.5 percent of resources would have had to find other employment. Suppose that the released resources now put to use in road-mending and silk manufacturing and so forth would have experienced productivity change of 0.5 percent per year (on the low end of the available possibilities) instead of the princely 1.9 percent they in fact experienced in cotton. The cotton industry in the actual, 1.9-percent world contributed a large amount-namely, (0.07) • (1.9 percent) = 0.133 percent per year-to the growth of national income. This one giant contributed some 24 percent of the conservatively measured total of about 0.55-percent-per-year growth of income per person nationally 1780 1860.

Now we calculate the counterfact. With the hypothetical cut off of trade you can make so to speak a mechanical “static-dynamic” argument as follow. The Harley revision of my table implies that non-cotton productivity change can be calculated from (1.41 – .07) • (the implied residual productivity change outside cotton) = (0.55 – 0.13). That is, the implied residual of productivity change outside cotton is 0.42/1.34, or 0.313 percent per year (I retain more than significant digits to avoid rounding errors). The resources in the hypothetical case would therefore contribute (0.035) • (1.9 percent) + (0.035) • (0.313 percent) = 0.086 percentage points a year. The fall in national productivity change can be inferred from the difference between the actual 0.133 percent per year attributable nationally to cotton and the hypothetical 0.086 percent per year attributable to a half sized cotton industry and the industries its resources would go to. The difference is about a .047 percentage point per year fall in the national rate of productivity change, that is, a fall from 0.55 percent a year to 0.503 percent a year. In the eighty years 1780 1860 such a lag would cumulate at monthly interest, however, to merely 4.5 percent of national productivity change. Remember that we are speaking here of doublings 1780 to 1860.

You could cut the productivity change in cotton to allow for alleged economies of scale in cotton and come to roughly the same result. No one has shown that such economies of scale were important in fact (as important as they are in the models of growth imagined by economists), or that economies or diseconomies of scale in other industries would not cancel the net gain. We are giving the “dynamic” argument all the advantages. Suppose the scale-effect productivity change were half of the princely 1.9 percent in cotton, or 0.945 percent per year. So now the calculations is (0.035) • (0.945 percent) + (.035) • (0.313 percent), or 0.0440 percentage points a year (as against 0.086 without the lost “economies of scale” inserted). National productivity change attributable to cotton falls from 0.133 percent per year all the way down to 0.0440 per year, a drop of 0.089 each year. So national productivity declines on this account in the hypothetical world from 0.55 actual to 0.461 percent per year. The difference in final attainment in 1860 is again small, merely 8 percent of productivity change, and a smaller percentage of national income.

Note that the result is forced by widespread character of productivity change (even under the implausible Crafts-Harley calculation of zero productivity change outside the industries I chose in 1981 as leading). Resources driven out of cotton do not simply disappear, resulting in a fall of national income equal to what they earned in cotton, as implied by non-economists (and even by Findlay and O’Rourke in careless moments). The resources of labor and capital shift, going into industries with lower productivity change. But since cotton was not the only industry experiencing productivity change even in the classic period of the early Industrial Revolution — a point that the economic historians Peter Temin and John Clapham and I insist upon, and historians of technology such as Margaret Jacob and Joel Mokyr have affirmed in detail — the imagined shift is not deadly to progress.⁴³ The dynamic effect sounds promising. But in quantitative terms a cotton textiles industry counterfactually smaller if foreign trade was not vigorous does not kill off growth. It’s another popular explanation that doesn’t work very well.

A “dynamic” argument, further, has a serious problem as an all purpose intellectual strategy. If someone claims that foreign trade made possible, say, economies of scale in cotton textiles or shipping services, she owes it to her readers, as I have already noted, to say why the gains on the swings were not lost on the roundabouts. Why do not the industries made smaller by the large extension of British foreign trade end up on the damaging side of the account? The domestic roads in Shropshire not constructed and the brass foundries unbuilt in Greater London because of Britain’s increasing specialization in Lancashire cotton textiles may themselves have had economies of scale, untapped. (The argument applies later in British history to the worries over “excessive” British specialization in foreign investment, insurance, and shipping).

Other industries than cotton, note, experienced productivity change, though usually at a smaller rate. Add that Britain was not a cotton mill and foreign trade was not all of national income and you have the conclusion that foreign trade was not an engine of growth able to explain even a doubling of national income, much less a factor of five or fifteen. And European trade with the rest of the world, as Patrick O’Brien showed along ago, was less than 4 percent of domestic product — another reason for doubting its importance.⁴⁴ Surprisingly, and somewhat against their training as economists, Findlay and O’Rourke attack the relevance of the low share of things in national income. They quote with approval a remark by the non-economist Paul Mantoux (1877-1956), in his history of the Industrial Revolution — published in French in 1907.⁴⁵ Mantoux wrote thus: “if we may borrow an analogy from natural science, only a negligible quantity of ferment is need to affect a radical change in a considerable volume of matter. The action of foreign trade upon the mechanism of production may be difficult to show, but it is not impossible to trace.”⁴⁶ The notion that natura facit salta, nature makes jumps, has become popular after the realization that a butterfly in China can cause a hurricane in Cuba. It is sometimes true. But if it is true in explaining the Industrial Revolution, so could any little part of the British economy have been the engine of growth. Domestic service was larger than the importation of tea and raw cotton and the like combined, and so under such an instable model the hiring of more scullery maids could have set off the innovations. And if you really want “small,” pick say the Birmingham brass industry with its continuous product innovation (as Maxine Berg has pointed out), or for that matter the vigorous silk industry in London around 1700. If the slave trade or the cotton industry or even foreign trade as a whole gives a satisfactory explanation of doublings and trebling of income, then we can turn also to a brass-and-silk industry explanation of why we are rich. And yet again we are led to wonder why similarly small industries in earlier times and other places did not tip the world into modernity.

After a good deal of complaining about the historical economics that they themselves are busy practicing, Findlay and O’Rourke come to the nub of their argument. “International trade,” they claim, “was a key reason why the British Industrial Revolution was different,” in not petering out as had previous efflorescences (Goldstone’s very appropriate word for the numerous lurches forward in technology that the world had previously seen, without permanent effect on the welfare of humans).⁴⁷ “For a small European country like Britain” — note that “small” is a somewhat strange characterization of one of the largest countries in Europe — “overseas markets were vital if its Industrial Revolution was to be sustained.”⁴⁸ And then Findlay and O’Rourke make a crucial connection to Britain’s military adventures: “in a mercantilist world in which nations systematically excluded their enemies from protected markets [a claim which makes it hard to understand the large volume of British-Continental trade, which took place in a mercantilist world] British military success over the French and other European rivals was an important ingredient in explaining her subsequent rise of economic prominence.”⁴⁹ Trade was important, they claim, and imperialism supported trade.

Thus the title of their book, Power and Plenty, and its theme: aggression is good for you. In correspondence with me O’Rourke has amiably disputed such a bald formulation of the theme. Yet in a more recent piece with Leandro Prados de la Escosura and Guilllaume Daudin he writes: “trade profited merchants, but also yielded revenues to the state; while the state needed revenues to secure trading opportunities for its merchants, by force if necessary.”⁵⁰ “Force” means “aggression,” and in the piece it is cashed in this way repeatedly, which uses throughout a football-and-war vocabulary of “pre-eminence,” “dominant position,” “struggle for power and plenty.” In all of O’Rourke’s work the gains from trade are said to be dependent on violence against “competitors,” as in a zero-sum footrace. One would not learn from such passages in Findlay and O’Rourke that trade is mutually beneficial, a matter mainly of cooperation, not competition.

True, people thought that mercantilist aggression was good for them. “Trade and empire,” O’Rourke and his 2008 co-authors continue, “were thus inextricably linked in the minds of European statesmen [because it is true in the world? because they were misled?], . . . which explains the incessant mercantilist warfare of the time.”⁵¹ It is the rhetoric of business-school deans such as Lester Thurow and big-thinking journalists such as James Fallows. It is not sound, whatever people at the time believed.

In establishing the growth-trade link, Findlay and O’Rourke use the static models to imagine a Britain without any trade at all (“if Britain had been closed to trade”; “absent trade”).⁵² An entire cut-off of trade, though, is not the relevant alternative. The question is whether the mercantilist policies that Britain employed, and above all its mercantile empire, helped or hindered industrialization, much. It’s a matter of more or less trade, not yes or no.

People innocent of economics, to repeat, believe that trade just is growth. Export or die. That’s not right, as Findlay and O’Rourke note when dismissing Keynesian models of trade as an engine of growth. So they need a better model. The model they develop to answer the relevant question, based on Darity (1982), puts a surprising emphasis on the slave trade. Findlay and O’Rourke argue that the New World and its cotton exports would have been impossible without slavery (note the similarity to Inikori’s arguments). But on the contrary cotton is easily grown without slaves, and has been early and late-early in India, late in post-bellum Alabama. (Sugar is quite another matter. Sugar brought slavery with it from India to Syria to North Africa, right down to the Jamaican and Mexican contract harvesters on H-2 visas working the cane sugar fields of north Florida. But Findlay and O’Rourke are making the argument that an international taste for cotton dresses and bed-sheets and underwear made the modern world, not that the international sweet tooth did.)

Cotton, they say, “depended” on slaves from Africa.⁵³ Likewise Marx: “With slavery, no cotton; without cotton, no modern industry. Slavery has given the value to the colonies, the colonies have created world trade; world trade is the necessary condition of large-scale machine industry.”⁵⁴ It does not seem so. Cotton seems to have been no more a necessarily slave crop than coffee was. Freedmen in the United States after 1865 picked cotton, just as freedmen in Brazil after 1887 picked coffee beans. Findlay and O’Rourke ask with a certain vexation in their tone whether “free white labor in the Americas . . . [would] have been able to fill the gap” in producing cotton.⁵⁵ Yet it did precisely so in the formerly-slaveholding American South.

Their argument is that British imperialism helped British trade so much that the Industrial Revolution happened. The argument assumes that a counterfactually pacific and free trade Britain would not have benefited from European engagement with the rest of the world. It is an odd assumption, since European places like Denmark benefited, with trivial overseas colonies. Sweden and Germany and Austria benefited. Findlay and O’Rourke want to make a nationalist, militaristic, imperialist argument that British prosperity depended on British guns aimed abroad. It is an argument that David Landes has frequently made. The historian Paul Kennedy stated flatly in 1976 that “Britain’s wealth would obviously have been lost had she herself surrendered command of the sea.”⁵⁶ The assertion, though conventional in British strategic thinking for centuries, runs against the logic of “this sceptred isle. . . this fortress. . . set in the silver sea/ Which serves it in the office . . . of a moat defensive to a house/ Against the envy of less happier lands.” A Britain with a little Tudor-style navy devoted to coastal defense would have remained independent. Wooden walls mattered up to the middle of the nineteenth century. Later it was British ingenuity in breaking the German naval code and inventing radar, not the Fleet sitting in miserable inaction at Scapa Flow, that chiefly prevented invasion. The surplus violence of ships of the line and then dreadnoughts and then aircraft carriers in aid of dominion over palm and pine and the Falkland Islands was always dubious as an economic proposition. Pride, certainly, and Margaret Thatcher’s re-election, was provided by command of the seas. The national income was not.

The economic models Findlay and O’Rourke use, whether formally or informally, are about European trade with itself and with the rest of the world. A Quaker Britain — unlikely counterfactual in 1800, with 20,000 Quakers in an aggressively nationalistic population of 15 ½ million — would have gotten the same prices and opportunities as the actual Britain, allowing for transshipment costs through Amsterdam or Le Havre. The scale of Manchester cotton manufacturing would have been little affected, at any rate if in God’s eyes Manchester had a comparative advantage in spinning cotton. Only the profits (those “rents” I mentioned) in their British addressees would have been lower, because French trans-shippers of cotton would take a cut. If Manchester was the right place to spin cotton before the invention of air-conditioning, then European events would have put it there, regardless of whether Britain won at Plassey or Quebec or Trafalgar or Waterloo. After all, France lost all those battles, and yet the making of cotton textiles flourished in Mulhouse and Lille.

Europe as a whole opened itself to the world after Henry the Navigator. Nutmeg became cheaper, even when it was a Dutch monopoly. The European gains from trade were felt indirectly by everyone who bought tropical products. As an economist would put it, that’s general equilibrium trade theory. Empires were not necessary. Thus Belgium, without an empire on its formation in 1830, industrialized smartly, as at the same time did the Rhineland, which was a part of a non-nautical and non-(overseas) imperial Prussia. Both of them saw the price of tobacco, spices, bananas, cotton, and other tropical and semi-tropical products fall greatly as imperialist and non-imperialist Europeans traded with the world. Overseas trade was not about Britain but about Europe. Britain’s overseas trade, in short, can’t explain Britain’s peculiarity. Lining up national conquest with national trade is an old claim, though Adam Smith and many economists since him have wisely contradicted it, without persuading many politicians. But national conquest doesn’t explain British industrialization, and certainly not the continuation on the way to the factor of sixteen.

All such denying of trade as the crucial engine of growth, though, is not to say that the expansion foreign trade was literally a nullity. Some goods — the banana for the Englishman’s breakfast table was the popular instance late in the nineteenth century, raw cotton the most important instance throughout — simply cannot be had in England’s clime, short of hot houses. The regional economist Gerald Silverberg has made the case to me for cotton as special because the technological unemployment caused by its expansion was felt not by political connected guildsmen at home but by the bleached bones of Indians starving when their hand industry was replaced by Manchester.⁵⁷ The truth in Silverberg’s argument is that trades like porcelain and cotton textiles in Britain could expand in country locations out of reach of the nay-sayers in established guild towns like Norwich or London. The trouble with the argument is that cotton did in fact have European substitutes, in wool and linen, as is shown by the fierce prohibitions on imported Indian calicoes into France and the rules in England that the dead must be buried in woolen shrouds. And the same trick could have been played in China or India, both having ample domestic sources of raw cotton, had the bourgeois rhetoric triumphed there — as it spectacularly did in the expansion of Japanese and especially Indian mechanized cotton textile manufacturing before and during World War I. In those days the detritus was the bleached bones, or at any rate the dole cheques, of Mancunians and Glaswegians in Britain.

More important, trade insures against famine, as the British Raj knew in building the railways of India — though Amartya Sen has pointed out that trade has this good effect only under a government sensitive to its subjects. The Bengal famine of 1943 was caused exactly by an colonial and arrogant insensitivity to non-voting subjects. The last widely and literally killing famine in England was in Shakespeare’s hierarchical times.

And trade is surely a conduit of ideas and competitive pressures. In India recently the License Raj has been broken down by ideological change, and in particular the opening of the economy for trade. After 1994 you could for the first time buy Kellogg’s corn flakes in New Delhi, praise be to Vishnu.⁵⁸ But such effects have nothing to do with imperial conquest — as is again best shown by the opening of Japan after 1868. Japan opened to trade, then, many decades later, under the influence of trade-follows-the-flag thinking at the height of Western imperialism, became itself a conqueror of Korea and then Germany’s colonies in China and then Manchuria and then China itself and finally much of east and southeast Asia. With most unsatisfactory economic consequences.

But a literal closing of British trade, entirely foregoing bananas at breakfast, using vastly more cotton for underwear at home, not getting any wheat at all in a famine, is not what is contemplated. The question is: was trade a stimulus to growth in the simple, mercantilist way usually contemplated in the literature? Apparently not. Is it plausibly a secondary cause as a desirable context for invention? Perhaps, though India (for example) was the center of the largest trading network before the eighteenth century yet did not innovate. But a Scots verdict seems wiser: not proven.

Here is the economist’s way of stating the problem with trade, reallocations, enclosure, investment, fuller employment, and all manner of shufflings. Think of the output of Stuff (clothing, food, houses) and Services (financing, shipping, doctoring, teaching, soldiering) in 1780 in Britain as being measured along two axes (bring back that high-school algebra and geometry). The possibilities in 1780 are a curve along which the actual Britain of 1780 could have taken a non-trading point, which we’ll call Self-Sufficiency:

Inefficiency, misallocation, opportunities missed, distortions introduced of the usual static sort are about being inside or on that curve. Note the point Massive Unemployment. It would be a foolish place to be, since you could get out to the curve and have more of both Stuff and Services. And you can get a little outside the curve by trading with foreigners. But only a little outside, to a point like Trade.

Good. But why have I drawn the so-called “production possibility curve” for 1780 as a miserable little scrunched up little curve in the very corner of the axes? Answer: it has to be a miserable little scrunched up curve in order also to represent Now on the same diagram. The amounts of Stuff and Services now (averaged) have to be sixteen times further out. Of course: that’s what the factor of sixteen means. And remember that in truth it’s more like a factor of 100.

Look at the diagram. None of the static arguments, and few of the dynamic, have any chance of explaining what happened in modern economic growth. No merely static improvement of conventional economic factors in 1780 or 1700 can come remotely close to the curve of Now. That’s why this greatest of historical events cannot be explained by static reallocation. And if it is to be explained by “dynamic” accumulation one has to explain, too, why earlier accumulation did not get the same explosive result. A dynamic explanation — for example, a foreign traded able to induce innovation on the scale of 1780 to Now — is so dynamic that it makes no sense as history. To put it differently, such an explanation is no explanation: it requires an answer to the question why just then, why the dynamism overtook the British economy in the eighteenth and nineteenth centuries. And that requires attending to bourgeois dignity and liberty.

---

38. [back] Findlay and O'Rourke 2007.
39. [back] Findlay and O'Rourke 2007, p. 337; compare O'Rourke, Prados de la Escosura, and Daudin 2008, p. 11.
40. [back] McCloskey 1980.
41. [back] Machlup 1963 (1975).
42. [back] Daudin 2004; compare my criticism ages ago of Jeffrey Williamson's calculation of the gain from re-investment of the gain from the railways in the United States, in McCloskey 1975b.
43. [back] McCloskey 1981 on widespread innovation; also Temin 1997, p. 80; Berg and Hudson 1994.
44. [back] O'Brien 1972.
45. [back] Findlay and O'Rourke always cite this very elderly book by a friend of Lloyd George, and the English translator for Clemenceau in the Versailles Conference, as "1962," fully 55 years after its last (and French) version. The impression unintentionally conveyed is that Mantoux was up-to-date in the scholarship of 1962. It is an outcome of the author-date system and the scholarly habits of careless whole-book citation encouraged by it. You can catch me doing it, too.
46. [back] Findlay and O'Rourke 2007, p. 336 (p. 103 in Mantoux).
47. [back] Findlay and O'Rourke 2007, p. 339.
48. [back] Findlay and O'Rourke 2007, p. 351.
49. [back] Findlay and O'Rourke 2007, p. 345.
50. [back] O'Rourke, Prados de la Escosura, and Daudin 2008, p. 2.
51. [back] O'Rourke, Prados de la Escosura, and Daudin 2008, pp. 2-3.
52. [back] Findlay and O'Rourke 2007, p. 344.
53. [back] Findlay and O'Rourke 2007, p. 336.
54. [back] Marx 1846,: Karl Marx (in French) to Pavel Yasilyevich Annenkov, December 28, 1846, quoted in Pomeranz and Topik 2006, p. 226, trans. Peter and Betty Ross, Marx Engels Collected Works Vol 38, p. 95, reproduced at http://www.marxists.org/ archive/marx/works/1846/letters/46_12_28.htm
55. [back] [back] Findlay and O'Rourke 2007, p. 339.
56. [back] [back] [back] [back] Kennedy 1976 (2006), p. 87.
57. [back] Personal conversation, Dahlem Seminar, Berlin, December 14, 2008.
58. [back] Jordan 1998. Her lead is "The Indian cornflakes maker Mohan Meakin says it has something to thank Kellogg Co. for: a wake-up call that has helped it win more business."

Name (required)

Mail (will not be published) (required)

Website