Over at Equitable Growth: Talking Points on "America's Imminent Budget Crisis": Focus
Weekend Reading: Mike Konczal: Everyone Should Take It Easy on the Robot Stuff for a While

Weekend Reading: Peter Temin (1997): Two Views of the Industrial Revolution

Peter Temin (1997): Two Views of the British Industrial Revolution:1

ABSTRACT: There are two views of the British Industrial Revolution in the literature today. The more traditional description sees the Industrial Revolution as a broad change in the British economy and society. This broad view of the Industrial Revolution has been challenged by Crafts and Harley who see the Industrial Revolution as the result of technical change in only a few industries. This article presents a test of these views using the Ricardian model of international trade with many goods. British trade data are used to implement the test and discriminate between the two views of the Industrial Revolution.

There are two views of the British Industrial Revolution in the literature today. The more traditional description is represented by the views of T. S. Ashton and David S. Landes. It sees the Industrial Revolution as a broad change in the British economy and society. In Ashton's memorable phrase, "A wave of gadgets swept over England."2 This broad view of the Industrial Revolution has been challenged recently by N. F. R. Crafts and C. Knick Harley. This new school of thought sees the Industrial Revolution as a much narrower phenomenon, as the result of technical change in a few industries. The new industries, obviously, were cotton and iron. All others were mired in premodern backwardness.3

It may seem as if the choice between these two views is a matter of taste, since the literature is almost exclusively about the two modern industries singled out by the narrow view of the Industrial Revolution. That appears to be how this choice is treated in the literature. In fact, the looseness of our current conception has encouraged a few people to take the views of Crafts and Harley to the extreme. Rondo Cameron argues that the change noted by these authors was so small relative to the whole economy that it no longer deserves the title of Industrial Revolution.4

But it is seldom that an empirical question cannot be tested. True, productivity indexes are hard to calculate for obscure industries. It is necessary to search for other data that will let the historian discriminate between these two views. Trade data provide the information needed to discriminate between these two views.

I will use a Ricardian model of international trade to formulate a testable hypothesis about the nature of the Industrial Revolution. In this model, the traditional view of the Industrial Revolution implies that Britain should have been exporting other manufactures--that is, manufactured products other than cotton textiles and iron bars. In the more modern view, by contrast, Britain should have been importing these same goods in the early nineteenth century. Trade data allow us to see which is the case.

The plan of this article is as follows. The first section argues that there are two distinct views of the Industrial Revolution in the literature. The second section will describe the Ricardian model of international trade with many goods and formulate the hypothesis to be tested. The third section will describe the British trade data and implement the test of the previous section. A final section concludes.

The traditional view of the British Industrial Revolution can be found in countless texts. T. S. Ashton's classic exposition clearly described a general change in British economy and society. He was very expansive in his descriptions of technical change:

Inventors, contrivers, industrialists, and entrepreneurs--it is not easy to distinguish one from another at a period of rapid change--came from every social class and from all parts of the country.

Expanding the statement quoted above about "a wave of gadgets," Ashton said:

It was not only gadgets, however, but innovations of various kinds--in agriculture, transport, manufacture, trade, and finance—-that surged up with a suddenness for which it is difficult to find a parallel at any other time or place.5

Banners and Alerts and www econ ucdavis edu faculty gclark 210a readings temin1 pdf

This view was widespread during the 1950s and 1960s. David Landes expressed it well in an authoritative book.6 The well-known growth estimates of Phyllis Deane and W. A. Cole confirmed the view of wide- spread change and appeared to provide a firm basis for the qualitative expositions.7 More current work by Joel Mokyr supports the pervasiveness of technological change in Britain at this time8 But in a recent survey of the literature, Patrick K. O'Brien labeled this view "old-hat" economic history that:

is still being read and continues to be written by an unrepentant but elderly generation of Anglo-American economic historians.9

The growth rate of the British national product was adjusted downward in a gradual process. C. Knick Harley revised the growth rate of manufacturing downward in 1982. N. F. R. Crafts extended these estimates into a revision of Deane and Cole's estimates of the British national product in his 1985 book. Crafts and Harley presented their "final" version in 1992.10

The implications of the new estimates for the conceptualization of the Industrial Revolution can be seen in an exercise introduced by D. N. McCloskey.11 He calculated the productivity gains of what he called the modernized sectors from industry sources. Then he weighted the gains by the share of the industries in gross production and added them. The productivity gain of all other sectors (except agriculture, which was estimated separately) was obtained by subtracting this total from the rate of growth of production in the economy as a whole. The calculations are shown in the first column of Table 1.

Crafts reproduced McCloskey's calculations in his book and noted that the bottom line, the estimated rate of growth of the economy as a whole, came from Deane and Cole. Since Crafts was revising these estimates, he substituted his new estimates as shown in the second column of Table 1. None of the industry estimates were changed; only the growth of the unidentified, residual sector. As can be seen, the contribution of "other sectors" to economic growth fell from 0.55 percent a year to 0.07 percent. In Crafts's words:

[T]he term 'Industrial Revolution'. should not be taken to imply a widespread, rapid growth of productivity in manufacturing.12

Quite the contrary. As Crafts repeated throughout his discussion, the Industrial Revolution in this view was a decidedly localized affair. The industries affected were textiles, iron, and transportation. All else other manufactures and other services—were technologically stagnant for the first half of the nineteenth century. This conclusion contrasts strongly with the assertions of Ashton and Landes. Crafts recognized that his new estimates created a paradox. If British manufacturing was in general so backward and British agriculture so progressive--as we know from other sources--then why did Britain not export agricultural goods and import manufactures in the early nineteenth century?13

It is important to understand the nature of this paradox. The traditional view implied that Britain had a comparative advantage in manufacturing. Crafts had denied the premise of this traditional view by asserting that most British manufacturing was backward and inefficient. Evidence that British agriculture was more productive than continental then implied that Britain had a comparative advantage in agriculture. It is no wonder that previous economic historians had not confronted this paradox; it does not exist in the traditional view of the Industrial Revolution.

The resolution of the paradox came in two propositions. First, Crafts confirmed the existence of paradox by reiterating that most British industry:

experienced low levels of labor productivity and slow productivity growth--it is possible that there was virtually no advance during 1780-1860.

Second, he resolved the problem by asserting that "rapid growth in key manufacturing sectors gave Britain a substantial comparative advantage in those activities."14 In other words, industrializing Britain had a comparative advantage in cotton and iron, not manufacturing as a whole. The clear implication of Crafts's view is that other manufactures were not exported because Britain lacked a comparative advantage in manufacturing in general. In fact, the juxtaposition of evidence of a productive agriculture with that of backward manufacturing outside of textiles and iron provided evidence that Britain had a comparative disadvantage in these other manufactures. That is, Crafts's resolution of the paradox implies that Britain should have been importing other manufactures along with agricultural goods.

Crafts and Harley recently revised and restated their new views in light of the ensuing discussion. Their definitive views reduced the rate of economic growth during the Industrial Revolution even further than Crafts's initial estimates.15 Harley incorporated these estimates into McCloskey's exercise, as shown in the third column of Table I. Harley revised McCloskey's estimates of productivity growth in the modem sector as Crafts had not done, reducing their aggregate contribution to economic growth. But because the rate of growth of the total economy was estimated to be so low, the contribution of other sectors fell to the vanishing point, from 0.07 percent per year to 0.02 percent per year.16

Harley embedded the Crafts-Harley view into a computable general equilibrium model of the British economy in the early nineteenth century. He distinguished four producing sectors in Britain: modern manufacturing, agriculture, services, and other industry. (The latter two sectors are the "all other" sector of Table I). Britain exports the products of modern manufacturing and imports agricultural goods in this model; services and other manufactures are not traded.17

Harley asserted that this model demonstrates the consistency of the Crafts-Harley view. But many products of other manufactures were easily traded, as will emerge below. Unless other manufacturing started out from a position of great comparative advantage--a presumption belied by the abundant historical evidence of the eighteenth century and explicitly denied by Crafts--the ability to export other manufacturing would have been rapidly eroded by technical progress in cotton, iron, and even agriculture. If agricultural goods were imported in the early nineteenth century, therefore, then other manufactures should have been as well.

In the literature survey noted above, O'Brien seemed to conclude that the gap between "old-hat" and new-fangled economic history can never be bridged. The problem is that the data needed to construct national income aggregates do not exist for many parts of British industry in the early nineteenth century. Microeconomic and macroeconomic studies, O'Brien appeared to assert, will just have to go their own ways.

Instead of banging our head against the stone wall of unavailable data, I propose to shift the terms of debate to a different kind of data.18 Crafts and Harley have suggested some implications of the new view for Britain's international trade. Trade data are available in great detail; can they help us to disentangle the nature of the Industrial Revolution?


A Ricardian Model of International Trade

The implications of the Crafts-Harley view for Britain's international trade can be used to formulate a test of these views. A model is needed to derive a test, more formal than Crafts's verbal exposition and more transparent than Harley's computable general equilibrium model. The Ricardian model of international trade with many goods poses the issues clearly.

A Ricardian model with many goods was analyzed by Dornbusch, Fischer, and Samuelson in 1977, and I follow their exposition here.19 They argued that the many goods can be seen as spread out along a continuum of comparative advantage and dealt with by their location along this continuum. The historical application of this model will be to identify the location of specific goods in this continuum.

Imagine two "countries": Britain and everywhere else. For ease of exposition, I will refer to the rest of the world as if it were a single foreign country. Since this is a Ricardian model, there is only one factor of production: labor. This factor can be seen as a Hicksian good by assuming that the relative price of different factors of production does not change. The model therefore does not say that there were no other factors of production but only that changes in the relative price of these factors can be ignored.20 This would not be suitable for consideration of, say, the repeal of the Corn Laws, but it provides a good way to focus on the effects of productivity changes over almost a century.21

Each country both produces and consumes a large variety of goods made from this single factor of production. These goods can be numbered from 1 to N. The technology of each country can be described by the labor needed to produce each good. The labor requirement to produce the nth good in Britain is an, where an is the number of hours of British labor needed to produce a single unit of the nth good. Following the convention of international trade, a*n represents the hours of foreign labor needed to produce the nth good in the foreign country.

The ratio of the labor needed to produce the good in the foreign country and in Britain is a*n/an. The goods can be re-indexed by this ratio, starting with the good for which the relative quantity of foreign labor needed for production is the highest (so the ratio, a*n/an, is the highest).

The pattern of trade is determined by the relative costs of producing goods in the two countries. And in this Ricardian model costs are simply the wages of the sole factor of production: labor. Let w be the British wage; w*, the foreign wage. Then the cost of producing good i in Britain is wai; the cost in the foreign country, w*a*i. Any good for which w*a*i > wai will be produced in Britain because its production costs are cheaper in Britain.

This inequality can be rewritten as a*i/ai > w/w*. Production costs for this good are lower in Britain; the good will be produced in Britain and exported to the foreign country. Conversely, any good i, for which w/w* > a*i/ai will be produced in the foreign country and imported into Britain. The numbering scheme for goods ensures that there is a point in the ordered list of goods such that all goods to the left with lower numbers are produced in Britain. All the goods with higher numbers are produced abroad.

Banners and Alerts and www econ ucdavis edu faculty gclark 210a readings temin1 pdf

This is illustrated in Figure 1, where the downward—sloping curve, A, shows a*/a for each good. It also shows the index of the last British export at any w/w*. The model needs a demand side to determine wages. Assume that consumers spend a constant share of their income on each good and that tastes are the same in both countries. The wage in each country is determined by the demand for labor, which is determined in turn by the range of goods produced in that country. If the range of domestic goods increases at any relative wage, then the demand for domestic labor rises. This raises the ratio of domestic to foreign wages, leading to a positive relation between w/w* and the range of goods produced domestically. This is shown as B, the upward sloping curve in Figure I. Curve A shows the interaction between the number of exports and relative wages in the goods market; curve B, in the labor market. The division between exported and imported goods is where curves A and B cross, at x0.22

Consider now the effect of technical change in Britain. I assume that there is no technical change outside of Britain, that is, no change in labor productivity in the foreign country. Alternatively, one could say that the Industrial Revolution did not spread outside Britain in the first half of the nineteenth century. This is roughly correct--at least for continental Europe--and it connects the model to the estimates of productivity change reported in Table 1.23

There are two cases. If the technical change is general, that is, it reduces ai for all i, then it increases a*i/ai, for all i. Curve A in Figure 1 shifts upward, increasing the range of goods exported by Britain at the same relative wage. This is shown as A' in Figure I. The point dividing imports and exports on curve A, now A', moves to the right. This increase in the range of goods produced in Britain increases the demand for labor in Britain and reduces the demand for labor in the foreign country. British wages consequently rise relative to wages in the foreign country. A new equilibrium is reached where curve B intersects the new A 'curve, at x1. At the new equilibrium, Britain is exporting goods that had previously been nontraded or imported.

If, by contrast, technical change is restricted to a few goods, the picture is more complicated. The simplest case is when productivity change is confined to a good already exported by Britain. Assume, for example, that advances in the British cotton textile industries caused people to shift demand from other goods to British textiles.24 Then the B curve shifts up and to the left because trade balance at any w/w* is achieved with the export of fewer British goods. The new curve is shown as B' in Figure I; the new equilibrium is to the left of the original point on curve A, at x2.

A more complex case is when the change in productivity changes the order of goods along curve A, moving a good from, say, the imported range to the exported. This change forces us to renumber all the other goods, giving them all higher numbers. For those goods close to the intersection of A and B, this change in the order could move them out of the range of exports into the range of imports (or nontraded goods). In terms of the goods themselves, the equilibrium has moved to the left as in Figure I.

Conversely, if a British sector has negative technical change--that is, if it stagnates while the rest of the economy progresses--then it will move to the right in the array. Depending on its starting point and the extent of its technical lag, it could cross the dividing line in Figure I and change from export to nontraded or import. This case describes the Crafts and Harley conclusion shown in the last column of Table 1. The rate of productivity change in other manufacturing was not only slower than in modern industries but also than in agriculture. If we assume that productivity was rising in other countries, then the absence of productivity change for 80 years shown in the final column of Table I surely would have eroded whatever comparative advantage Britain might have had in these goods.

All of the subcases of restricted technical change move in the same direction. Britain exports fewer nontextile goods than before, although the representation in Figure I is too simple to describe all of the subcases. It follows that if there were more than one of these developments under way, the effects would cumulate. Rapid advances in British textiles and no productivity change in other manufacturing then are two separate causes for the number of British exported goods to fall.

Summarizing, uniform and restricted technical change have opposite implications for the movement of dividing points in equation 1. General technical change moves the dividing line between exports and nontraded goods to the right; restricted technical change, to the left. General technical change causes the list of exports to rise, while restricted technical change causes it to fall. This difference provides a test of historical views. The test is which goods are exported and imported, not how much of each good is traded. The conclusions just reached refer to changes in the location of equilibria along this continuum of goods.

The empirical evidence needed to discriminate between the two kinds of productivity change consists of listing exported and imported goods, not calculations of their magnitudes. To discover changes in the lists of exports and imports, lists need to be compiled for different dates during Britain's industrialization. To create this test we need to identify goods in the array of equation I.

There are three categories of British goods: exports, nontraded goods, and imports. Following Harley, we identify exports with modern British industry, nontraded goods with services not related to trade, and imports with agriculture. But Harley has a fourth good in his model that is the one of most interest here. The question is where to put Harley's fourth category, other manufactures.

The discussion of the preceding section implies that there are two different answers. In the broad view of the Industrial Revolution, other manufactures were similar to modem manufactures; technical change was widespread. Exports of many manufactured goods should have been expanding. In the narrow view, by contrast, other manufactures were doing far worse than agriculture. Harley assumed they were not traded in his computable general equilibrium model, but as noted above, this is implausible. Other manufactures should have been imports in the Crafts-Harley view of the economy.

There are several reasons why the Crafts-Harley view implies other manufactures were imports. As cotton changed from an import to an export in the eighteenth century, the range of other manufactures exported should have fallen.25 Further technical progress in cotton textiles that greatly increased the consumption of their products in the nineteenth century even after cotton textiles had moved to be first in the index of British goods magnified this effect. And as the residual sectors stagnated relative to agriculture in the nineteenth century, their costs of production in Britain must have risen sharply relative to the cost of growing food in Britain. Since agricultural goods were imported, the products of these other sectors--to the extent that they were traded at all--should have been imported as well. Even if other manufactures were not imported at the start of the nineteenth century, the rates of productivity change shown in the last column of Table I surely would have made them imports by midcentury.

The Ricardian model consequently generates a simple test to discriminate between the two views of the British Industrial Revolution. Were other manufactures exported or imported? If exported, then the view that technical change was widespread among British industries in the early nineteenth century is confirmed. But if the other manufactures were imported, then the conclusion that technical change was restricted to a very few modem industries while other industries stayed mired in premodern production techniques is confirmed.

The path of trade in other manufactures also gives information. In the Crafts-Harley view shown in the last column of Table I, these activities were not experiencing technical change in the first half of the nineteenth century. The productivity gap between other manufactures and agriculture--not to mention modern industry--was growing rapidly. Other manufactures, even if exported early in the Industrial Revolution, should have found their relative costs rising and their exports falling. They should have gone from exports to imports. This is not a statement about the relative rate of growth of these exports; it rather is whether individual goods changed from being exported to being imported.

The two views of the Industrial Revolution, therefore, can be tested by looking at marginal British exports. I do not claim that the pattern of trade in these goods describes the Industrial Revolution, only that it provides a test between two views of this event. Was Britain losing its comparative advantage in other manufacturing exports at the margin or maintaining it? After industrialization had progressed for a while, were other manufactures exported as the Ashton and Landes view implies or imported as the Crafts-Harley view implies?

It may seem odd to test major views of the Industrial Revolution by looking at marginal activities. Not only should major historical events have large causes, but the tests about them, it seems, should involve the principal activities as well. Unhappily, this is not the case. Different stories have been presented to explain the same events. To be plausible, they all have to explain the major aspects of these events. It is only in the details that they differ, although, as described above, these differences may imply other, more important disagreements. The devil, as they say, is in the details.26


Using the Model to Discriminate Between the Two Views

Some dimensions of British trade as summarized by Ralph Davis appear in Table 2.27 The dominant place of manufactures in British exports is easily apparent from the first row. The important and initially growing share of cotton manufactures in total manufactures is clear from the next row. Iron manufactures, for all their importance in the narratives of the Industrial Revolution, were never a major part of British manufacturing exports. The question here is what was happening outside of these dominant industries. Manufacturing exports other than cotton, woolens, and iron are shown in the last row of Table 2. They were quite substantial, and they show no evidence of being pushed aside by cotton exports--as woolens were.

Banners and Alerts and www econ ucdavis edu faculty gclark 210a readings temin1 pdf

I went to the Parliamentary Papers to find data on exports of individual commodities. Not every year contained trade information in detail. I consequently had to chose years for which I found detailed data, which did not always correspond to the years Davis had surveyed. The trends shown in Table 2 were very clear in my data as well, and 1 do not think any information was lost in the change of dates. I used data for three-year periods around 1810, 1830, and 1850, and a few other years between the first two to investigate changes in the early stages of industrialization and during the Napoleonic Wars.

Table 3 shows exports of other manufactures for three years centered on 1850, close to the end of the period of the calculations shown in Table 1. The table lists all manufacturing exports other than those identified in Table 2. They are sorted by the magnitude of exports. The quantities exported are shown for information only. They were used to check my data against Davis's but they are not relevant to the test performed here. The evidence to be cited in Table 3 is the list of different products.

Banners and Alerts and www econ ucdavis edu faculty gclark 210a readings temin1 pdf

Linen was a major export. Silk manufactures also were steadily exported. Turning to metals, we find hardware and cutlery, brass and copper manufactures, and tin and pewter continuing to be exported. Other exports include earthenware, haberdashery, apparel, soap, and hats. The interest of this list is the absence of an organizing principle. There were exports of many different sorts.

Table 4 shows the correlation between the exports of individual goods for categories that existed in both years for several different years. There is a suspicion that the composition of other exports changed more in the two decades before 1831 than after. The evidence does not confirm this view.28

Banners and Alerts and www econ ucdavis edu faculty gclark 210a readings temin1 pdf

Breaking up the earlier period critical years in both the Industrial Revolution and the conversion to a peacetime economy into subperiods gives the results shown in the lower part of Table 4. With the possible exception of the initial years of peace, there is no evidence of much change in the structure of other exports. This is true despite the inclusion of Irish exports in the totals after 1826.

Before concluding that much of other British industry was not backward, we need to look at British imports. For if it turns out that these same articles were being imported, and especially if they were being imported in greater quantities than they were exported, the conclusion would not follow.

Table 5 shows the composition of British imports in the same years as Table 3. The effect of stagnating productivity outside the modem sector and agriculture should have been most evident by 1850. But there was, as noted for exports in Table 4, little variation in the composition of British imports over the first half of the nineteenth century.

Www econ ucdavis edu faculty gclark 210a readings temin1 pdfWww econ ucdavis edu faculty gclark 210a readings temin1 pdf

It can be seen easily that the imports are not of the same goods that were being exported, with a few exceptions. Silk was imported in greater quantities than it was exported. This was not an activity in which Britain maintained a comparative advantage. Linen was imported in the years 1811 to 1813, but Irish linens were no longer counted as imports by 1830, and there were few other linen imports. Most of the flax shown as imports must have gone to Ireland.

There is no mystery why Britain imported sugar, tea, or indigo. They, and the many other tropical products consumed in Britain, would not have been exported under any reasonable set of prices or changes in productivity. The important agricultural imports for the test performed here are corn, hides, and wool (sheep's). They were imported from western Europe and could have been exported from Britain.29 These products are the products that Britain should have exported before other manufactures in the nineteenth century according to the Crafts-Harley view.30

None of the myriad other British manufacturing exports were imported at all. Britain maintained a clear comparative advantage in a wide variety of manufacturing industries throughout the first half of the nineteenth century. They held their own in the face of the spectacular growth of cotton-textile exports during those years. There is no hint that these other commodities were being pushed off the list of exports by the growth of cotton exports. Except for the Napoleonic War period, they kept pace with cotton exports.

It is not surprising that Britain sold a wide variety of manufactures to tropical countries. Their comparative advantage in tropical exports was so large that they specialized completely. There is little surprise, therefore, that Britain exported hats to Australia in exchange for wool. It is important, however, that Britain did the same for western Europe.31

The shaping of hats was still done by hand at midcentury, but this handicraft had been surrounded by mechanization well before then. A hatmaker in London employed 1,500 people in 1840. The preparation of the fur and wool to make the felt for hats was thoroughly mechanized, using steam-powered machinery. And the dyeing of the finished hat was done on machinery that allowed over 100 hats to be dyed at once. Labor productivity consequently was high.32

There is an exception that proves the rule. Table 5 shows that there were small imports of manufactured woolens and cotton. But they were approximately one-tenth the amounts of the exports of those commodities shown in Table 2.33 They are hardly the exception. Further down the list in Table 5 come watches and clocks. As Landes noted in his book on that industry, the English clockmakers and watchmakers, were falling behind their continental competitors in the nineteenth century.34 Productivity stagnated in this industry, and it had become an import industry by midcentury.35

The export of most other manufactures, however, was continuing merrily along. The lesson of the constant rank order of these exports is that the various industries were keeping pace with each other. The share of cotton textiles in total manufacturing exports peaked in the 1830s as shown in Table 2. There was a slight fall in the share from the period 1814 to 1816 to the period 1854 to 1856. Other manufacturing exports as a whole kept pace with cotton exports during these 40 years, and exports of individual industries did so as well.

Although the empirical evidence in this test is the identity of exports and imports as shown in Tables 3 and 5, the productivity advance in British manufacturing should have lowered their prices relative to imports. They did. Albert Imlah correctly recognized this "severe deterioration" in the net barter terms of trade as a signal of British success, not distress. It is no surprise that the price of cotton manufactures fell rapidly in response to productivity growth. But even the price of woolen manufactures, which were declining as a share of British exports (Table 2), fell almost as rapidly as the price of exports as a whole.36

It follows, therefore, that the traditional "old-hat" view of the Industrial Revolution is more accurate than the new, restricted image. Other British manufactures were not inefficient and stagnant, or at least, they were not all so backward. The spirit that motivated cotton manufactures extended also to activities as varied as hardware and haberdashery, arms, and apparel.

It follows also that the calculations shown in the last column of Table 1 cannot be accepted as authoritative. The low rate of productivity change shown for other activities is too low. There must have been more technical progress outside the listed sectors in Table 1 to produce the results shown here.



This test confirms the traditional view that the Industrial Revolution saw changes in more than a few industries. Technical change was hardly uniform a point conceded by every historian but it was widespread. Britain became the workshop of the world, not just the cotton factory of the world.

Scattered descriptions suggest the existence of a pattern in other manufactures.37 With few exceptions, there were no factories like the famous cotton factories. Instead there were new organizations of work along the lines identified by Charles Sabel and Jonathan Zeitlin.38 "Flexible specialization" has been thought of as a description of French industrialization.39 Perhaps it also describes a significant part of the Industrial Revolution in Britain.

More research will be needed to confirm or refute suggestions like this. The test performed here shows that increases in British productivity were not confined to cotton and iron in the first half of the nineteenth century. The "old-hat" view of the Industrial Revolution cannot be banished by calling it names. It lives among us, and it deserves more attention to fill in its all too evident gaps.


Allen, Robert. "Agriculture during the Industrial Revolution." In The Economic History of Britain since 1700, Second edition, edited by Roderick Floud and Donald McCloskey, I, 96-122. Cambridge: Cambridge University Press, 1994.

Ashton, T. S. An Economic History of England: The 18th Century. London: Methuen, 1955. The Industrial Revolution. Oxford: Oxford University Press, 1971 [1948].

Berg, Maxine. The Age of Manufactures: Industry Innovation and Work in Britain, 1700-1820. Ottawa, NJ: Barnes and Noble, 1985.

Berg, Maxine, and Pat Hudson. "Rehabilitating the Industrial Revolution." Economic History Review 45, no. 4 (1992): 24-50.

Cameron, Rondo. A Concise Economic History of the World from Paleolithic limes to the Present, Second edition. Oxford: Oxford University Press, 1993.

Crafts, N. F. R. British Economic Growth during the Industrial Revolution. Oxford: Clarendon Press, 1985.

"British Industrialization in an International Context." Journal of Interdisciplinary History 19, no. 3 (1989): 415-28.

"Exogenous or Endogenous Growth? The Industrial Revolution Reconsidered." Journal of Economic History 55, no. 4 (1995): 745-72.

Crafts, N. F. R., and C. K. Harley. "Output Growth and the Industrial Revolution: A Restatement 703-30. of the Crafts—Harley View." Economic History Review 45, no. 4 (1992):

Davis, Ralph. The Industrial Revolution and British Overseas Trade. Leicester: Leicester University Press, 1979.

Deane, 1965. Phyllis. The First Industrial Revolution. Cambridge: Cambridge University Press,

Deane, Cambridge Phyllis, University and W. A. Press, Cole. 1962. British Economic Growth, 1688-1959. Cambridge:

Dodd, Augustus George. Kelley, Days 1967. at the Factories, London: Charles Knight, 1843. Reprinted by

Dornbusch, R., S. Fischer and P. Samuelson. "Comparative Advantage, Trade, and Payments in a Ricardian Model with a Continuum of Goods." American Economic Review 67, no. 5 (1977): 823-39.

Harley, C. Knick. "British Industrialization before 1841: Evidence of Slower Growth 267-89. during the Industrial Revolution." Journal of Economic History 42, no. 2 (1982):

"Reassessing the Industrial Revolution: A Macro View." In The British Industrial Revolution: An Economic Perspective, edited by Joel Mokyr, 171-226. Boulder, CO: Westview Press, 1993.

Hartwell, R, M. The Industrial Revolution and Economic Growth. London: Methuen, 1971.

Imlah, Albert H. Economic Elements in the Pax Britannica, Cambridge, MA: Harvard University Press, 1958.

Landes, David S. Prometheus Unbound: Technological Change and Industrial Develop- Press,ment in 1969. Western Europe from 1750 to the Present. Cambridge: Cambridge University

Revolution in Time: Clocks and the Making of the Modern World. Cambridge, MA: Harvard University Press, 1983.

Matthias, Peter. The First Industrial Nation. London: Methuen, 1969.

McCloskey, D. N. "The Industrial Revolution 1780-1860: A Survey." In The Economic History of Britain since 1700, edited by R. C. Floud and D. N. McCloskey, I, 103-27. Cambridge: Cambridge University Press, 1981.

McKendrick, Neil. "Josiah Wedgwood: An Eighteenth—Century Entrepreneur in 408-24. Salesmanship and Marketing Techniques." Economic History Review 12, no. 3 (1960):

Mokyr, Joel. The Lever of Riches.. Technological Creativity and Economic Progress. New York: Oxford University Press, 1990.

"Editor's Introduction: The New Economic History and the Industrial Revolu- tion." In The British Industrial Revolution: An Economic Perspective, edited by Joel Mokyr, 1-131. Boulder, CO: Westview Press, 1993.

O'Brien, Patrick K. "Introduction: Modern Conceptions of the Industrial Revolution." In The Industrial Revolution and British Society, edited by Patrick K. O'Brien and Roland Quinault, 1-30. Cambridge: Cambridge University Press, 1993.

Piore,1984. Michael, and Charles Sabel, The Second Industrial Divide. New York: Free Press,

Sabel, Charles, and Jonathan Zeitlin. "Historical Alternatives to Mass Production: Politics, Markets, and Technology in Nineteenth—Centwy Industrialization." Past and Present 108, no. 1 (1985): 133-76.

Temin, Peter.1976. Did Monetary Forces Cause the Great Depression? New York: Norton,

nance Accounts: An Account of the Value of the United Kingdom. House of Commons. "Fi Great Britain," Parliamentary Papers 1812-13 Imports into and the Exports from (100) Vol. 11. Pt. 1.

"Finance Accounts: An Accountof the Value of the Imports into and the Exports from Great Britain," ParliamentaryPapers 1818 (147) Vol. 12. Pt. 1. of the Value of the Imports into and the Exports

"FinanceAccounts:AnAccount from Great Britain," ParliamentaryPapers 1823 (146) Vol. 13. Pt. 1. of the Value of the Imports into and the Exports

"FinanceAccounts:AnAccount from Great Britain," ParliamentaryPapers 1828 (139) Vol. 16. Pt. 1. of the Value of the Imports into and the Exports

"FinanceAccounts:AnAccount from Great Britain," ParliamentaryPapers 1831-32 (131) Vol. 26. Pt. 1. of the Value of the Imports into and the Exports

"FinanceAccounts:AnAccount Papers 1852 (196) Vol. 28. Pt. 1. from Great Britain," Parliamentary