eco exam

-0.

~.c _

Vl ~

~ -6 2
.c ‘”CL~

Wavelength. i,. In Years Wavelength. 2. In Years

Fig. 9.7 Cross-spectral estimates of deaths and rate of natural in-
crease in relation to real wages for England, 1539-1839.
Phase estimates indicated by solid circles correspond to sig-
nificant estimates of coherence-squared and are more ac-
curate than the others. Estimates were made using a
Parzen window with T = 301, M = 20. Deaths and wages
were measured as residuals from the regression of the log
of the basic series on time. Natural increase was used
un transformed.

short-run relations also held over the long run, although these data
provide no evidence on this point. Even in the short run, however, wages
account for only about 15 % of the variance in growth rates, so that
most of the variation is exogenous. Furthermore, inspection of long-run
life-expectancy series, as in figures 9.1 and 9.4, suggests that long-run
variation in population growth rates was also dominated by exogenous
variation.

Under these circumstances, over the very long run, the average wage
level will be an important determinant of average population growth
rates. But even over the course of centuries, fluctuations of growth rates
about that average level may be largely exogenous.

9.4 A Model of Economic-Demographic Equilibrium

At this point it will be helpful to introduce a simple equilibrating
model relating fertility, mortality, wages, and population. Rent and
terms of trade could also be added, but they play an essentially passive
role and would only clutter the diagram.

The relation of fertility and mortality to wages, measured by their
crude rates band d, may be plotted as in the top half of figure 9.8. The

542 Ronald Demos Lee

Wage (w)

—-

d(w)

~
:!l
Ui
c
Q
<0 :; N·a. 0 ll.

b(w)

w(N)

Fig. 9.8

Wage (w)

Economic-demographic equilibrium.

level and curvature of the birthrate curve are determined primarily by
norms and institutions, although at very low wages biological consider-
ations may become important. Some societies might have horizontal
fertility curves, if neither nuptiality nor marital fertility depended on
material well-being. Societies with institutional arrangements conducive
to high fertility, such as the extended family system, would have higher
birthrate curves than those with less pronatalist institutions, such as the
nuclear family. The death-rate curve is primarily biologically deter-
mined, although such additional factors as income distribution, central-
ized famine precautions, and in some cases infanticide and geronticide
are also important.

The population growth rate, equal to b – d, is given by the difference
between the two schedules; where they intersect, the growth rate is zero
and the population is stationary. The corresponding wage, w*, is vari-
ously known as the “long-run equilibrium wage,” the “natural wage,”
the “conventional standard of living,” or “subsistence.”

The lower half of the diagram shows the relation between the wage
rate and the size of the population; it corresponds to the demand for
labor, which I assume is fixed. Corresponding to the equilibrium wage
is an equilibrium size of population, N*. There will also be equilibrium
levels of rent and terms of trade, which are not shown. Evidently the

543 Perspective on Economic Aspects of the Population Explosion

equilibrium is stable; when population size is below N* its growth rate
will be positive, and conversely.

Now consider the effect of a once-for-all shift in the demand for labor;
this situation is shown in figure 9.9. When w(N) shifts out to WI (N),
the wage will initially rise, inducing population growth until population
attains its new equilibrium at the old wage level. Thus, over the long
run, population responds passively to economic advance, while a roughly
constant level of material well-being is maintained; this is the “iron law
of wages.”

Now consider the effect of a permanent exogenous decline in mortal-
ity, shifting the schedule from d (w) to d 1 (w). This is shown in figure
9.10. 15 The decline in mortality lowers the equilibrium wage and popu-
lation size; growth rates are initially positive until a new equilibrium
is established with lower fertility and wages and larger population size.
The point to note is that the equilibrium wage is not a culturally deter-
mined parameter, as the classical economists thought; it depends also
on a level of mortality that was subject to autonomous long-run change.
It is this that gives population an independent role in history: within
broad limits, the equilibrium population and living standard changed
when mortality changed, even if institutions and the economic base of
society remained completely unaltered.

;;;
a:
.<:: ;;;

0
u

b(w)c
‘”
~

d(w)’E
to

U
::l

U
w· Wage (w)

wl(N)

w(N)

Fig. 9.9 Increased demand for labor.

Wage (w)

544 Ronald Demos Lee

b(w)
—-d(w)

—-d,(w)

Wage (w)
w(N)

I
I
I
I

~ i
~ I

Cf) I

6 I
~ Ni ——-
~ N’ ———-
a
a..

Wage (w)

Fig. 9.10 Exogenous mortality decline.

I have simplified here by ignoring the direct links of fertility to mor-
tality; these would cause the fertility curve to shift in response to shifts
in the mortality curve. However, such direct links were ver¥ weak (see
Lee 1973, p. 598; 1978a, p. 167). Therefore it was only through long-
run change in the norms and institutions themselves that society could
maintain constant population and wages in the face of exogenous change
in mortality. The automatic homeostatic mechanisms were not adequate
in these circumstances.

In earlier papers (Lee, 1973, 1978a, b) I used estimated forms of this
model to simulate the course of wages, population, and fertility, assum-
ing that only mortality varied exogenously. These simulations fit the
historical data remarkably well for 1250 to 1700 and 1705 to 1784.

The diagram can also be used to illustrate the effect of a steady rate
of shift of the demand for labor, of the sort included in the equations
estimated earlier. Suppose that this rate of shift is such that population
growth at rate r leaves wages unchanged; the estimates suggested r =
0.4% per year. Then in steady state growth, population will grow at
rate r, and the wage will be constant at a level such that b (w) – d (w)
= r. This situation is shown in figure 9.11. Evidently the wage will have
to be a bit above its “natural” level in order to induce growth. Exoge-
nous change in mortality will alter the steady-state wage but will only
temporarily affect the population’s growth rate.

d(w)
b(w)

545 Perspective on Economic Aspects of the Population Explosion

Finally, consider a simultaneous decline in mortality and initiation of
growth at rate r in the demand for labor. This situation is shown in
figure 9.12. In this case we might observe constant fertility, low mortal-
ity, and population growth with no diminution in wages. This is the situ-
ation T. H. Marshall had in mind when he wrote of eighteenth-century
England (1965, p. 248) :

The obvious temptation is to assert that the death rate was not only
the variable, but also the determining, factor in the increase of popu-
lation, and that, to understand the causes of this increase, we should
study the deaths rather than the births. But, clearly, a horizontal line
on a graph may be as dynamic as a diagonal; the forces that prevent
a birth rate from falling may be as significant as those that make it
rise.

Ordinarily, one would expect a fall in the death rate to be followed by a
fall in fertility, as equilibrium is attained at a lower rate and larger
population; if this does not happen, it suggests that the underlying cause
of continuing population growth is economic progress, not the mortality
decline.

Might this be similar to the situation in today’s LDCs? We often
observe exogenously declining mortality, relatively constant fertility and
per capita income, and rapid population growth. Without the concurrent

‘”10
a:

t;:Q)
z
~

Vi
c
o
1i5
“3
0.
o
c-

al
c
::>
o
u

V>

6
Wage (w)

Fig. 9.11 Labor demand increasing at a constant rate.

546 Ronald Demos Lee

b(w)

~—d(w)

~—-dl(W)

Wage(w)
,.
Q)

Z
a)
N

Vi
C
o

‘;::;
m
:;
c-
o
c..

al
c
::>
o
u

‘”(5
Wage(w)

Fig. 9.12 Offsetting changes in growth of labor demand and mortality.

economic development, surely by now incomes and fertility would have
fallen and mortality risen. It is not quite right to attribute the population
growth to the mortality decline, although this may be the most conspicu-
ous exogenous change; growth in the capacities of these economies to
sustain population should perhaps be accorded the major responsibility.

A final comment on this model in relation to the LDCs is in order.
Whatever the nature of the social mechanisms that may have regulated
population in Asia, it is clear that a balance was reached at a much
higher level of fertility and mortality than in Europe. Apparently life
expectancy in China and India at the turn of this century was about 23
years (see Barclay et ai. 1976; Das Gupta 1971), versus perhaps 30
years in Europe; the total fertility rate must consequently have been
about 6.5 versus 4.5 in Europe. The necessary change in fertility-regu-
lating institutions, in response to declining mortality, is staggering.

9.5 Summary and Conclusions

For today’s LDCs there is little empirical evidence on the economic
effects of population change. For the economy of preindustrial England
and perhaps Europe, on the other hand, population emerges clearly as
the dominant cause of long-run change in wages, rents, industrial prices,
and income distribution. The economy could absorb population growth

Economic History Association

Two Views of the British Industrial Revolution
Author(s): Peter Temin
Source: The Journal of Economic History, Vol. 57, No. 1 (Mar., 1997), pp. 63-82
Published by: Cambridge University Press on behalf of the Economic History Association
Stable URL: http://www.jstor.org/stable/2951107

.

Accessed: 13/06/2013 14:13

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

.

Cambridge University Press and Economic History Association are collaborating with JSTOR to digitize,
preserve and extend access to The Journal of Economic History.

http://www.jstor.org

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=cup

http://www.jstor.org/action/showPublisher?publisherCode=eha

http://www.jstor.org/stable/2951107?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the British Industrial Revolution

PETER TEMIN

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 intemational 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 premodem 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 modem 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

The Journal of Economic History, Vol. 57, No. 1 (Mar. 1997). c The Economic History
Association. All rights reserved. ISSN 0022-0507.

Peter Temin is Professor, Department of Economics, Massachusetts Institute of Technology,
Cambridge, MA, 02139.

This is a shortened version of the working paper of the same title, NBER Historical Paper 81 (March
1996). I thank Wilson W. Thai for research assistance and participants in seminars at Harvard, Yale,
and the University of British Columbia for helpful comments.

2Ashton, Industrial Revolution, p. 42.
3Mokyr, “Editor’s Introduction,” pp. 6-7, distinguishes four views of the Industrial Revolution. My

division corresponds roughly to his macroeconomic and technological schools.
4Cameron, Concise Economic History, pp. 16567.

63

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

64 Temin

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.

TWO VIEWS OF THE INDUSTRIAL REVOLUTION

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

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 time.8 But in a recent survey of the

5Ashton, Industrial Revolution, pp. 13, 42.
6Landes, Prometheus Unbound, pp. 41, 105.
7Hartwell, Industrial Revolution; Matthias, First Industrial Nation; Deane, First Industrial

Revolution; and Deane and Cole, British Economic Growth.
8Mokyr, Lever, chap. 10.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 65

TABLE 1
CONTRIBUTIONS TO NATIONAL PRODUCTIVITY GROWTH, 1780-1860

(percentage per annum)

Sector McCloskey Crafts Harley
Cotton 0.18 0.18 0.13
Worsteds 0.06 0.06 0.05
Woolens 0.03 0.03 0.02
Iron 0.02 0.02 0.02
Canals and railroads 0.09 0.09 0.09
Shipping 0.14 0.14 0.03

Sum of modernized 0.52 0.52 0.34
Agriculture 0.12 0.12 0.19

All others 0.55 0.07 0.02
Total 1.19 0.71 0.55

Sources: McCloskey, “Industrial Revolution,” p. 114; Crafts, British Economic Growth, p. 86; and
Harley, “Reassessing the Industrial Revolution,” p. 200.

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 manufactur-
ing 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.”1 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

90’Brien, “Introduction,” p. 7. O’Brien’s exposition focused on the growth rate during the British
Industrial Revolution, but estimates of income growth cannot be separated from the underlying
conception of the Industrial Revolution, as shown below.

‘0Harley, “British Industrialization”; Deane and Cole, British Economic Growth; Crafts, British
Economic Growth; Crafts, and Harley, “Output Growth.”

“McCloskey, “Industrial Revolution,” p. 114.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

66 Temin

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 manufactur-
ing.

pl2

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

“2Crafts, British Economic Growth, p. 86, emphasis in the original. Crafts’s estimates reduced the
implied rate of productivity change in all other sectors from 0.65 percent per year to 0.08 percent per
year. He added in a footnote that even this new, low estimate could be an overestimate.

13Crafts, “British Industrialization.”
I4lbid., p. 425.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 67

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 1. 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: modem manufacturing,
agriculture, services, and other industry. (The latter two sectors are the “all
other” sector of Table 1). Britain exports the products of modern manufac-
turing 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.

“Crafts and Harley, “Output Growth.”
“6Crafts recently revised downward even further his estimate of productivity change by taking

account of the growth of human capital. If Harley estimated the rate of productivity change of
individual industries in Table 1 from prices (as McCloskey did), these estimates would not be affected
by the consideration of human capital in the overall total. This would turn the residual category of other
activities negative. This change makes the test proposed below even sharper than with the estimated
rates in Table 1. Crafts, “Exogenous or Endogenous Growth?”

17Harley, “Reassessing the Industrial Revolution.”

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

68 Temin

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 contin-
uum. 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.2′

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

“8Berg and Hudson, “Rehabilitating the Industrial Revolution,” also recommend shifting the terms
of debate about the Industrial Revolution, albeit in a different direction than developed here.

19Dornbusch, Fischer and Samuelson, “Comparative Advantage.”
20It is worth noting that Britain was not pressing against land scarcity at this time. Acres of arable

rose almost by half in the first half of the nineteenth century while the agricultural labor force stayed
constant. Allen, “Agriculture,” pp. 104-07.

2″More formally, the assumption of a single factor of production and changing technology is more
appropriate to the question at hand than a model with several factors and stable technology. A model
with many factors and changing technology would have so many degrees of freedom that no useable
test could be derived from it.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 69

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).

al /al>a2 la2>a3 /a3>>aN/aN (1)

The pattern of trade is detennined 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 *. Any good for which w *a * > 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,j, for which a7*jaj < w/w* will be produced in the foreign country and imported into Britain. The number- ing 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. 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 deter-
mined 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 1. 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 xo.22

22Capital movements do not affect the allocation of production in this model. Transport costs and
uniform tariffs do not affect the argument; they only introduce a band of nontraded goods between
exports and imports. The pattern of trade did not vary much at a time that tariffs were falling rapidly,
suggesting that individual tariffs had little effect on the overall pattern of trade. Exports of services are
ignored, following Harley.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

70 Temin

a 5 /a,w/w

BB

exports x2 xO xI imports index

FIGURE 1

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 *,/a, 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 1. 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

231f productivity was growing uniformly in other countries, then this rate of change needs to be
deducted from the rates derived from the final column of Table 1 to get relative rates. This does not
change the order of change in the various sectors of the British economy. Like Crafts’s recent reduction
in the overall rate of productivity change in Britain, it only strengthens the argument here. Since there
is only one factor of production, total factor productivity and labor productivity are the same. As noted
above, I am assuming that labor stands for a Hicksian good and that the relative prices of different
factors of production did not change substantially. See note 20.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 71

curve B intersects the new A ‘curve, at x,. 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 1; 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 1.

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 1 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 modem
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 1 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 1 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

24Since the model has assumed constant shares of income spent on each good, this is equivalent to
saying that the demand for British textiles was elastic.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

72 Temin

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 1.
There are three categories of British goods: exports, nontraded goods, and
imports. Following Harley, we identify exports with modem 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 implausi-
ble. 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
1 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

25Ashton, Economic History of Britain, p. 154; and Cameron, Concise Economic History, p. 160.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 73

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 premodem 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 1, these activities
were not experiencing technical change in the first half of the nineteenth
century. The productivity gap between other manufactures and agricul-
ture-not to mention modem 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 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

26This is the same argument I used in a very different context in Did Monetary Forces Cause the
Great Depression?

27Davis, Industrial Revolution. Davis also surveyed intermediate decades, with results close to those
shown in Table 2.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

74 Temin

TABLE 2
SHARES OF TOTAL AND MANUFACTURING EXPORTS

(percentage)

Sector 1794-1796 1814-1816 1834-1836 1854-1856
Manufacturing/total 86 82 91 81
Cotton/manufacturing 18 49 53 42
Woolens/manufacturing 27 21 17 15
Iron/manufacturing 11 2 2 7
Other/manufacturing 44 28 28 36

Source: Davis, Industrial Revolution, pp. 95-101.

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.

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 I 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.

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

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 75

TABLE 3
EXPORTS OF OTHER MANUFACTURES, 1850-1852

Value
Export (pounds sterling)

Linens 4,694,567
Hardwares and cutlery 2,556,441
Brass and copper manufactures 1,830,793
Haberdashery and millinery 1,463,191
Silk manufactures 1,193,537
Earthenware of all sorts 975,855
Machinery and millwork 970,077
Tin and pewter wares and tin plates 904,275
Apparel, slops, and Negro clothing 892,105
Beer and ale 513,044
Arms and ammunition 505,096
Stationary/stationery of all sorts 373,987
Apothecary wares 354,962
Lead and shot 339,773
Glass/glass of all sorts 296,331
Plate, plated ware, jewelry, and watches 286,738
Soap and candles 275,200
Painters’ colors and materials 237,880
Books, printed 234,190
Cabinet and upholstery wares 155,407
Cordage 155,127
Leather saddlery and harness 121,401
Hats of all other sorts 106,933
Musical instruments 85,006
Umbrellas and parasols 72,928
Carriages of all sorts 57,018
Spirits 52,843
Fishing tackles 41,607
Hats, beaver and felt 34,351
Mathematical and optical instruments 34,289
Spelter, wrought, and unwrought 22,097
Bread and biscuit 15,529
Tobacco (manufactured) and snuff 14,762

Source: U.K., Parliamentary Papers, 1852 (196), vol. 28, pt. 1.

decades before 1831 than after. The evidence does not confinn this view.28
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 tums 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.

28The data from 1811 to 1813 are in official values, whereas the later data are in real values. This
does not seem to have affected the correlation, but it is hard to know. There also are fewer observations
in the data from 1811 to 1813 because fewer individual exports were identified.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

76 Temin

TABLE 4
CORRELATIONS AMONG OTHER MANUFACTURING EXPORTS

Number of
Years Observations Correlation

1811-1813 and 1830-1832 18 0.95
1830-1832 and 1850-1852 28 0.93

1811-8113 and 1816-1818 15 0.78
1816-1818 and 1821-1823 21 0.90
1821-1823 and 1826-1828 21 0.97
1826-1828 and 1830-1832 28 0.98

Source: U.K., Parliamentary Papers, 1812-13 (100), vol. 1 1, pt. 1; ibid., 1818 (147), vol. 12, pt. 1.;
ibid., 1823 (220), vol. 12, pt. 1; ibid.,1828 (130), vol. 16, pt. 1; ibid., 1831-32 (310), vol. 26, pt. 1;
ibid., 1852 (196), vol. 28, pt. 1.

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.

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.

29Davis, Industrial Revolution, pp. 114-24.
30Not, however, according to Harley’s CGE model since other manufactures do not trade in that

model.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 77

TABLE 5
VALUE OF IMPORTS, 1850-1852

Value
Import (pounds sterling)

Wool, cotton 23,670,472
Sugar 10,762,045
Corn, meal, and flour 9,167,600
Tea 5,796,086
Silk 5,163,865
Coffee 3,480,594
Flax, and tow or codilla of hemp and flax 3,123,329
Wool, sheep’s 2,049,348
Hides, raw or tanned 1,999,233
Cochineal, granilla, and dust 1,909,848
Oil 1,793,320
Madder, madder root, and garancine 1,687,568
Guano 1,476,940
Tallow 1,333,889
Indigo 1,191,495
Wood and timber 1,153,477
Dye and hardwoods 1,104,308
Hemp, dressed or undressed 990,917
Spelter 957,540
Wines 927,721
Spirits 902,351
Seeds 719,017
Woollen manufactures 710,414
Rice, cleaned or in the husk 668,585
Bacon 653,214
Potatoes 562,595
Currants 559,919
Cotton manufactures 548,065
Cheese 537,322
Copper, unwrought and part wrought 477,778
Butter 466,357
Brimstone 383,691
Tobacco and snuff 367,685
Skins and Furs 367,269
Saltpetre and cubic nitre 355,564
Iron in bars, unwrought 336,706
Gum 298,147
Oil seed cakes 296,993
Glass 270,110
Lard 258,790
Ashes, pearl and pot 238,077
Bark 213,708
Turpentine 213,561
Pork, salted or fresh 210,692
Quicksilver 201,669
Tin 200,801
Sago 178,329
Raisins 170,443
Lead, pig and sheet 169,024
Borax 164,565
Terra japonica and cutch 150,035
Hair or goats’ wool, manufactures of 148,473
Cocoa, cocoa-nut husks and shells, and chocolate 145,973
Tar 142,819
Bones of animals and fish (except whalefins) 140,049
Cinnamon 132,648
Beef, salted or fresh 122,855
Embroidery and needlework 114,999
Copper ore and regulus 113,166

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

78 Temin

TABLE 5-continued

Value
Import (pounds sterling)

Cloves 106,630
Animals, living; viz. oxen, bulls, cows, and calves 103,463
Watches 95,928
Safflower 94,911
Boots, shoes and calashes, and boot fronts 94,779
Pepp.er 93,744
Lace, thread, and cushion or pillow lace 82,816
Leather gloves 81,441
Shumac 80,320
Oranges and lemons 74,845
Yarn, worsted or silk and worsted 73,690
Clocks 73,661
Rhubarb 70,912
Whalefins 69,277
Valonia 66,799
Hair, horse 63,159
Fish, of British taking 60,405
Nutmegs 60,144
Almonds of all sorts 59,705
Linens 57,562
Pimento 57,222
Liquorice juice and paste 54,153
Senna 53,452
Cork 53,196
Rags, &c. for paper 49,140
Wax, bees’ 46,160
Teeth, elephants’ 44,661
Bristles 44,048
Cassia lignea 43,735
Mace 41,082
Ginger 40,639
Animals, living; viz. sheep and lambs 35,144
Books, bound or unbound 33,865
Hams 28,935
Annatto 25,468
Isinglass 24,685
Figs 22,812
Barilla and alkali 2,122

Source: U.K., Parliamentary Papers, 1852 (196), vol. 28, pt. 1.

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 hat-
maker 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

31Davis, Industrial Revolution, pp. 101, 125. Davis’s category is Hats, haberdashery, garments, and
so forth, so it is not absolutely certain that hats were exported to Westem Europe. I use it as my
example, although other items of Davis’s list could be cited as well.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 79

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 approxi-
mately one-tenth the amounts of the exports of those commodities shown in
Table 2. 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 imnports. They
did. Albert lmlah 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.

32Dodd, Days.
33Davis, Industrial Revolution, p. 101.
34Landes, Revolution.
35Data for earlier years than in Table 5 show that clocks and watches were not imported earlier in the

nineteenth century.
36lmlah, Economic Elements, pp. 93-102, 211-12.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

80 Temin

CONCLUSIONS

This test confirms the traditional view that the Industrial Revolution saw
changes in more than a few industries. Technical change was hardly
uniforim-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.36 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.37 “Flexible
specialization” has been thought of as a description of French industrializa-
tion.38 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.

36For example, Berg, Age.
37Sabel and Zeitlin, “Historical Alternatives.”
38Piore and Sabel, Second Industrial Divide.

REFERENCES

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: Bames 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 Worldfrom Paleolithic Times 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.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

Two Views of the Industrial Revolution 8 1

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

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

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

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

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

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
during the Industrial Revolution.” Journal of Economic History 42, no. 2 (1982):
267-89.

. “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-

ment in Western Europe from 1750 to the Present. Cambridge: Cambridge University
Press, 1969.

. 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
Salesmanship and Marketing Techniques.” Economic History Review 12, no. 3 (1960):
408-24.

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

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

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

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

82 Temin

United Kingdom. House of Commons. “Finance Accounts: An Account of the Value of the
Imports into and the Exports from Great Britain,” Parliamentary Papers 1812-13
(100) Vol. l1.Pt. 1.

_____.* “Finance Accounts: An Account of the Value of the hnports into and the Exports
from Great Britain,” Parliamentary Papers 1818 (147) Vol. 12. Pt. 1.

. “Finance Accounts: An Account of the Value of the Imports into and the Exports
from Great Britain,” Parliamentary Papers 1823 (146) Vol. 13. Pt. 1.

______-. “Finance Accounts: An Account of the Value of the Imports into and the Exports
from Great Britain,” Parliamentary Papers 1828 (139) Vol. 16. Pt. 1.

. “Finance Accounts: An Account of the Value of the Imports into and the Exports
from Great Britain,” Parliamentary Papers 1831-32 (131) Vol. 26. Pt. 1.

. “Finance Accounts: An Account of the Value of the Imports into and the Exports
from Great Britain,” Parliamentary Papers 1852 (196) Vol. 28. Pt. 1.

This content downloaded from 140.254.87.103 on Thu, 13 Jun 2013 14:13:14 PM
All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp

• Article Contents

p. 63
p. 64
p. 65
p. 66
p. 67
p. 68
p. 69
p. 70
p. 71
p. 72
p. 73
p. 74
p. 75
p. 76
p. 77
p. 78
p. 79
p. 80
p. 81
p. 82

• Issue Table of Contents

The Journal of Economic History, Vol. 57, No. 1 (Mar., 1997), pp. 1-266
Front Matter
Storage and English Government Intervention in Early Modern Grain Markets [pp. 1 – 33]
The Immigrant Assimilation Puzzle in Late Nineteenth-Century America [pp. 34 – 62]
Two Views of the British Industrial Revolution [pp. 63 – 82]
African and European Bound Labor in the British New World: The Biological Consequences of Economic Choices [pp. 83 – 115]
The Decline in Southern Agricultural Output, 1860-1880 [pp. 116 – 138]
Delivering Coal by Road and Rail in Britain: The Efficiency of the “Silly Little Bobtailed” Coal Wagons [pp. 139 – 160]
The Political Instability of Reciprocal Trade and the Overthrow of the Hawaiian Kingdom [pp. 161 – 189]
Review Articles
Individualism Transformed [pp. 190 – 195]
Searching for Consistency in Historical Data: Alternate Estimates of Russia’s Industrial Production, 1887-1913 [pp. 196 – 202]
Editor’s Notes [pp. 203 – 209]
Reviews of Books
Early Modern Europe
untitled [pp. 210 – 211]
untitled [pp. 211 – 213]
untitled [p. 213]
Modern Europe
untitled [pp. 214 – 216]
untitled [pp. 216 – 217]
untitled [pp. 217 – 218]
untitled [pp. 218 – 219]
untitled [pp. 220 – 221]
untitled [pp. 221 – 222]
untitled [pp. 222 – 223]
untitled [pp. 223 – 225]
China and Japan
untitled [pp. 225 – 226]
untitled [pp. 227 – 228]
untitled [pp. 228 – 229]
untitled [pp. 229 – 231]
Africa and the Middle East
untitled [pp. 231 – 232]
untitled [pp. 232 – 233]
untitled [pp. 234 – 235]
The Caribbean
untitled [pp. 235 – 236]
The United States and Canada
untitled [pp. 237 – 238]
untitled [pp. 238 – 239]
untitled [pp. 239 – 240]
untitled [pp. 241 – 242]
untitled [pp. 242 – 243]
untitled [pp. 243 – 244]
untitled [pp. 245 – 246]
untitled [pp. 246 – 247]
untitled [pp. 247 – 248]
untitled [pp. 248 – 250]
untitled [pp. 250 – 251]
untitled [pp. 251 – 252]
untitled [pp. 252 – 253]
untitled [pp. 253 – 254]
untitled [pp. 254 – 256]
untitled [pp. 256 – 257]
untitled [pp. 257 – 258]
Economic Thought
untitled [pp. 258 – 259]
General and Miscellaneous
untitled [pp. 259 – 260]
untitled [pp. 261 – 262]
untitled [pp. 262 – 263]
untitled [pp. 263 – 264]
untitled [pp. 264 – 266]
Back Matter

Why the industrial revolution was
British: commerce, induced invention,

and the scientific revolution

1

By R. C. ALLEN

Britain had a unique wage and price structure in the eighteenth century, and that
structure is a key to explaining the inventions of the industrial revolution. British
wages were very high by international standards, and energy was very cheap. This
configuration led British firms to invent technologies that substituted capital and
energy for labour. High wages also increased the supply of technology by enabli

ng

British people to acquire education and training. Britain’s wage and price structur

e

was the result of the country’s success in international trade, and that owed much to
mercantilism and imperialism. When technology was first invented, it was only
profitable to use it in Britain, but eventually it was improved enough that it became
cost-effective abroad. When the ‘tipping point’ occurred, foreign countries adopted
the technology in its most advanced form.ehr_532 357..38

4

The industrial revolution is one of the most celebrated watersheds in humanhistory. It is no longer regarded as the abrupt discontinuity that its name
suggests, for it was the result of an economic expansion that started in the
sixteenth century. Nevertheless, the eighteenth century does represent a decisive
break in the history of technology and the economy. The famous inventions—the
spinning jenny, the steam engine, coke smelting, and so forth—deserve their
renown, for they mark the start of a process that has carried the west, at least, to
the mass prosperity of the twenty-first century.2 The purpose of this article is to
explain why they were invented in Britain, in the eighteenth century.

Explaining the industrial revolution is a long-standing problem in social science,
and all manner of prior events have been adduced as causes.3 Recent research has
emphasized non-economic factors like the British constitution4 or British culture,

5

1 This article is the text of the Tawney Lecture, delivered on 5 April 2009 at the Economic History Society
Annual Conference, University of Warwick.

2 There has been a debate about the breadth of technological progress during the industrial revolution with
Crafts, British economic growth; Harley, ‘Reassessing’; Crafts and Harley, ‘Output growth’; and Crafts and Harley,
‘Simulating’, arguing that productivity growth was confined to the famous, revolutionized industries in the period
1801–31, while Temin, ‘Two views’, has argued that many more industries experienced productivity growth.
Whatever one believes about 1801–31, it is clear that many non-revolutionized industries experienced produc-
tivity growth between 1500 and 1850. The incentives to invent discussed in this article applied to all industries,
not just the famous ones I discuss here.

3 Hartwell, Causes, and Mokyr, ‘Editor’s introduction’, provide surveys. Crafts, ‘Industrial revolution in
England’, has suggested that Britain’s lead was fortuitous.

4 Proponents of this view include North and Weingast, ‘Constitutions’; De Long and Schleifer, ‘Princes and
merchants’; LaPorta, Lopez-de-Silanes, Schleifer, and Vishny, ‘Law and finance’; Acemoglu, Johnson, and
Robinson, ‘Rise of Europe’. For critical or contrary perspectives, see Clark, ‘Political foundations’; Epstein,
Freedom and growth; Quinn, ‘Glorious revolution’s effect’; Hoffman, Postel-Vinay, and Rosenthal, Priceless markets;
Pomeranz, Great divergence; Mathias and O’Brien, ‘Taxation’; Mathias and O’Brien, ‘Incidence’; Hoffman and
Norberg, Fiscal crises; and Bonney, Rise.

5 Landes, Unbound Prometheus; Clark and Jacks, ‘Coal’.

Economic History Review, 64, 2 (2011), pp. 357–384

© Economic History Society 2010. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main
Street, Malden, MA 02148, USA.

both of which have been alleged to be superior. The matter, however, is controver-
sial: while certain legal arrangements and cultural predispositions may have
favoured economic development, it is not at all clear that Britain was alone in
possessing them. In this article, I sidestep these debates by taking a different
approach and emphasizing the importance of economic incentives as a cause
of the industrial revolution. The essence of the industrial revolution was new
technology, and I trace the links from Britain’s success in the world economy
in the early modern period to the technological breakthroughs of the eighteenth
century.

I focus on the sources of invention and analyse these in terms of the demand and
supply of new technology. The empirical base of this analysis is international
comparisons of wages and prices. These comparisons show that eighteenth-
century Britain had a unique wage and price structure. British wages were excep-
tionally high compared with wages in other parts of Europe and in Asia, while the
prices of capital and energy were exceptionally low. The price and wage structure
affected the demand for technology by giving British businesses an exceptional
incentive to invent technology that substituted capital and energy for labour. The
high real wage also stimulated product innovation since it meant that Britain had
a broader mass market for ‘luxury’ consumer goods including imports from east
Asia. The supply of technology was also augmented by the high real wage. It meant
that the population at large was better placed to buy education and training than
their counterparts elsewhere in the world. The resulting high rates of literacy and
numeracy contributed to invention and innovation.

The supply of technology was also affected by other developments. Jacob and
Stewart, and Mokyr have emphasized the importance of Newtonian science, the
Enlightenment, and genius in providing knowledge for technologists to exploit,
habits of mind that enhanced research, networks of communication that dissemi-
nated ideas, and sparks of creativity that led to breakthroughs that would not have
been achieved by ordinary research and development.6 Mokyr’s influential inter-
pretation conceptualizes these elements as the industrial enlightenment. These
developments would have boosted the rate of invention at any level of wages,
prices, and human capital. That is also their weakness. The scientific revolution
and the industrial enlightenment were Europe-wide phenomena that do not dis-
tinguish Britain from the Continent. That is appropriate from some points of view:
France was in the lead in many industries with new techniques to its credit in
paper, clocks, glass, and textiles, for instance. Any theory that explains British
success by positing a British genius for invention is immediately suspect. Instead,
we must explain why Britain invented the technologies it did and why they were so
transformative.

This article takes as its point of departure Edison’s famous observation that
‘invention was 1% inspiration and 99% perspiration’. That suggests that inventing
the industrial revolution was mainly a story about research and development
(R&D) (perspiration). R&D is an economic activity with distinctive features. As
Machlup remarked, ‘Hard work needs incentives, flashes of genius do not’.7 By

6 Jacob, Scientific culture; Jacob and Stewart, Practical matter; Mokyr, ‘Editor’s introduction’; idem, Gifts of
Athena.

7 Machlup, Production, p. 166.

358 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

concentrating on the R&D and the incentives to undertake it, we can get a much
deeper understanding of why the industrial revolution happened when and where
it did.

Britain’s unique wage and price structure was the pivot around which the
industrial revolution swung. Logically, the next question, therefore, is to explain
what determined the wages and prices. They turn out to have been the result of
Britain’s great success in the international economy in the early modern period,
and that relationship will also be examined. The answer to the grand question of
how the industrial revolution was related to the early modern economy is this: the
commercial and imperial expansion of Britain created a unique structure of wages
and prices, and that price structure, in turn, prompted the technological break-
throughs of the eighteenth century by increasing the demand for inventions that
substituted capital and energy for labour, and by generating a population that was
exceptionally able to respond to those incentives due to its high rates of literacy,
numeracy, and craft skills. The spread of scientific culture may have had a rein-
forcing effect. Some important scientific developments contributed to this
advance, but they would not have been acted upon without a demand for the
technologies that applied them.

8

I

Since invention was an economic activity, its pace and character depended on
factors that affected business profits including, in particular, input prices. It is
easier to understand why the industrial revolution happened in eighteenth-century
Britain if we compare wage rates and energy prices in the leading economies of the
day. In these comparisons, Britain stands out as a high-wage, cheap-energy
economy.

Our views of British wages are dominated by the standard of living debate. Even
optimists who believe the real wage rose in the industrial revolution accept that
wages were low in the eighteenth century. They were certainly lower than they are
today, but recent research in wage and price history shows that Britain was a
high-wage economy in four senses. Firstly, at the exchange rate, British wages were
higher than those of its competitors. Secondly, high silver wages translated into
higher living standards than elsewhere. Thirdly, British wages were high relative to
capital prices. Fourthly, wages in northern and western Britain were exceptionally
high relative to energy prices.

These trends are illustrated in figures 1–4. These figures were constructed from
databases of wages and prices assembled from price histories written since the
middle of the nineteenth century.9 The typical price history is based on the
archives of an institution that lasted for hundreds of years—colleges and hospitals
are favourites. Historians work through their accounts, recording the quantity and
price of everything bought or sold, and draw up tables of the annual averages.
Usually prices are found for a range of agricultural and food stuffs as well as cloth,

8 The argument is developed more fully in Allen, British industrial revolution.
9 The data are referenced and described in greater detail in Allen, ‘Great divergence in European wages’; idem,

‘Poverty and progress’; idem, ‘Timber crisis’; idem, ‘India in the great divergence’; idem, British industrial
revolution; idem, ‘Industrial revolution in miniature ’; Allen, Bassino, Ma, Moll-Murata, and van Zanden, ‘Wages,
prices, and living standards’.

INDUSTRIAL REVOLUTION 359

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

1

3

75

14

25

14
75

15

25

15
75

1

6

25

16
75

17
25

17
75

18

25

2

0

18
15
14

1

2

10

8
6

G
ra

m
s

o
f

si
lv

er
p

er
d

ay

London

Amsterd

am

Vienna

Florence

Delhi

Beijing

4
2
0

Figure 1. Labourers’ wages around the world
Source: See text.

5
6
London
Amsterdam
Vienna
Florence
Delhi
Beijing
4
3
2
1
0

13
75

14
25

14
75

15
25

15
75

16
25

16
75
17
25
17
75

18
25

18
75

Figure 2. Subsistence ratio for labourers: income relative to cost of subsistence basket
Source: See text.

360 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

fuel, candles, building materials, implements, and a miscellany of other items.
Wages and salaries are often also recorded.The commodities are measured in local
weights and measures, and prices are stated in local units of account, and these
must be converted to international standards. Prices histories have been written for
many European cities, and the research is being extended to Asia. By putting all of
this material in the computer, international comparisons are becoming possible for
the first time, and they are redefining our understanding of economic history. In
particular, they throw new light on the origins of the industrial revolution, as we
shall show.

Figure 1 shows the history of nominal wages of building labourers in leading
European and Asian cities from the middle ages to the industrial revolution. The
various units of account in which the data were recorded have been converted to
grams of silver, since silver coins were the principal medium of exchange. The
figure shows that the divergence in nominal wages was minimal in Europe at the

2

1.5
England

Strasbou

rg

Vienna
1

0.5

16
30

16
80

17
30

17
80

Figure 3. Wage rate relative to price of capital
Source: See text.

6
5
4
3
2
1
0

A
m

ste
rd

am

Lo
nd

on
Pa

ris

St
ra

sb
ou

rg

N
ew

ca
stl

e

Be
iji

ng

Figure 4. Wage rate relative to price of energy
Source: See text.

INDUSTRIAL REVOLUTION 361

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

end of the late middle ages. There was little wage inflation subsequently in eastern
Europe. Wages in western Europe rose during the price revolution (1550–1620).
Thereafter, there was a three-way split, with silver wages falling in southern
Europe, levelling out in the Low Countries, and continuing to rise in London.
From the late seventeenth century onwards, London wages were the highest
recorded.

London wages rose above those elsewhere in Britain in the sixteenth century. By
the late seventeenth century, however, wages in southern English towns like
Oxford were rising to close the gap. Wage movements in northern England were
more erratic. In the late seventeenth century, builders’ wages in cities like York
were as high as those in Oxford. Wage growth ceased in the north in the early
eighteenth century, however, so the region fell behind the south in nominal wages,
although the level was still higher than in most parts of the European Continent.
Fast wage growth towards the end of the eighteenth century brought the north to
the same level as the south, however, and all parts of England had exceptionally
high silver wages.10

Comparisons with Asia further emphasize the high wages in eighteenth-century
Britain. In Beijing, Canton, Japan, and Bengal, labourers earned between one and
two grams of silver per day—less than half the wage in central or eastern Europe
and a smaller fraction of earnings in the advanced economies of the north-west of
the Continent.11

Did Britain’s high nominal wages translate into high living standards or were
they offset by high prices in Britain? To explore this issue, welfare ratios have been
computed for leading cities. Welfare ratios are defined as full-time annual earn-
ings12 divided by the cost of a basket of consumer goods sufficient to keep a family
at a specified standard of comfort—in this case at minimal subsistence. Baskets are
constructed with most spending on the grain that was cheapest in each locality (for
example, oats in northern Europe, polenta in Florence, sorghum in Beijing, and
millet in Delhi).Very small portions of meat, peas or beans, butter or oil, cloth, and
fuel, and a small allowance for housing are also included. Consumption is set at the
low level of 1,940 kilocalories per day for an adult male, with other family
members proportioned accordingly. Calculations with baskets corresponding to a
more affluent lifestyle have also been undertaken, and the relative rankings are
unchanged.

Figure 2 plots the welfare ratios for the cities in figure 1. The population decline
caused by the Black Death meant that real incomes were high everywhere in the
fifteenth century. Welfare ratios in London and the Low Countries were trendless
across the early modern period, although there were oscillations in the series.
Moreover, fully employed workers in these regions earned three to five times the
cost of the subsistence lifestyle. They spent their extra income on a superior diet
(with bread, beer, and much more meat) and more non-food consumer goods
including some of the luxuries of the ‘consumer revolution’ of the eighteenth

10 Gilboy, Wages; Allen, Great divergence; idem, ‘Poverty and progress’.
11 Özmucur and Pamuk, ‘Real wages’; Allen, ‘India in the great divergence’; Allen et al., ‘Wages, prices, and

living standards’; cf. Allen, Bengtsson, and Dribe, Living standards.
12 European building workers were paid by the day, and I assume that 250 days was a full year’s work, making

allowance for Sundays, religious holidays, and erratic employment. Many Asian wages are based on monthly
earnings, and I assume employment for 12 months.

362 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

century.13 In contrast, real living standards fell dramatically across the Continent,
reaching a level of about one. In eighteenth-century Florence and Vienna, fully
employed building workers earned only enough to maintain their families at rock
bottom subsistence. There was no surplus for bread, meat, beer, or wine, let alone
imported luxuries. Real wages also fell sharply in provincial England in the
sixteenth century, but even at the trough labourers in Oxford earned at least 50 per
cent more than bare bones subsistence. The nominal wage inflation of the late
seventeenth century meant that welfare ratios in Oxford were between 2.5 and 3.0
in the eighteenth century.

If we extend the comparisons of living standards to Asia, English performance
looks even more impressive. Low silver wages in the East were not counterbal-
anced by even lower food prices. Welfare ratios for labourers in Canton, Beijing,
and Japan were about one in the eighteenth and nineteenth centuries—as low as
those in the backward parts of Europe. Mass demand for manufactures was very
limited across Asia, since most consumer spending was directed towards basic
necessities.

The earnings of craftsmen (carpenters, masons, and so forth) followed the same
trends as labourers in all countries. Skilled workers, however, earned more than
the unskilled, so their welfare ratios were higher everywhere. Craftsmen in London
or Amsterdam earned six times what was required to purchase the subsistence
basket, while their counterparts in Germany or Italy only 50 per cent more than
that standard. Craftsmen in north-western Europe spent much of their surplus
income on more food and better-quality food. Nonetheless, the mass market for
consumer goods was much larger in Britain and the Low Countries than in most
of Europe.

A third sense in which Britain was a high-wage economy was in terms of the
wage rate relative to the price of capital. Figure 3 plots the ratio of a building
labourer’s daily wage relative to an index of the rental price of capital in northern
England, Strasbourg, and Vienna. The rental price of capital is an average of price
indices for iron, nonferrous metals, wood, and brick, multiplied by an interest rate
plus a depreciation rate. Strasbourg and Vienna were chosen since long series of
wages and prices are available for those cities, and their data look comparable to
those of most of Europe apart from the Low Countries. The series reflect differ-
ences in the price of capital across space as well as over time.

The ratio of the wage relative to the price of capital was similar in all of the cities
in the first half of the seventeenth century. Then the series diverged. In England,
labour became increasingly expensive relative to capital from 1650 onwards. This
rise reflects the inflation of nominal British wages at the time. In contrast, the ratio
of the wage to the price of capital declined gradually in Strasbourg and Vienna
across the seventeenth and eighteenth centuries. The incentive to mechanize
production was much greater in England than in France, Germany, or Austria.

Finally, there is a fourth sense in which labour was costly in industrializing
Britain. That involves a comparison of wages to the price of fuel. Figure 4 is a bar
graph of the ratio of the building wage rate to the price of energy in the early

13 Shammas, Pre-industrial consumer; McKendrick, Brewer, and Plumb, Birth; de Vries, ‘Purchasing power’;
Fairchilds, ‘Production’; Weatherill, Consumer behaviour; Berg and Clifford, Consumers and luxury; Berg, Luxury
and pleasure.

INDUSTRIAL REVOLUTION 363

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

eighteenth century in important cities in Europe and Asia. In this ratio, the price
of a kilogram of fuel was divided by its energy content, so energy prices are
expressed as grams of silver per million BTUs (British Thermal Units). The ratio
is calculated for the cheapest fuel available in each city—coal in London and
Newcastle, peat in Amsterdam, charcoal or firewood in the other cities.

Newcastle stands out as having the highest ratio of labour costs to energy costs
in the world. To a degree the high ratio reflects high British wages, but the low cost
of coal was the decisive factor. Indeed, a similar ratio characterized the situation on
all of the British coalfields and in the industrial cities (Sheffield, Birmingham, and
so forth) built on them.The only place outside of Britain with a similarly high ratio
of labour to energy costs was probably the coal mining district around Liège and
Mons in present-day Belgium. The high cost of labour relative to fuel created a
particularly intense incentive to substitute fuel for labour in Britain. The situation
was the reverse in China, where fuel was dear compared to labour. The Chinese
invented very large kilns for firing their pottery because such kilns had a high ratio
of volume to surface area and so conserved heat. The reverse was true in Britain
where kilns were small and thermally inefficient.

II

Britain’s unusual wages and prices were due to two factors. The first was Britain’s
success in the global economy, which was in part the result of state policy. The
second was geographical—Britain had vast and readily worked coal deposits.

In pre-industrial Europe, real wages moved inversely to the population. As
figure 2 indicates, the real wage rose in Britain and Italy after the Black Death of
1348/9, which cut the population by about one-third. As population growth
resumed, the real wage fell in most of Europe between the fifteenth century and
the eighteenth.The Low Countries were an important exception to this trend. Real
wages fell in rural England in the sixteenth century, but London bucked the trend
in the same way as Antwerp and Amsterdam, and indeed, as we have seen, living
standards rose generally in southern England from 1650 onwards. Why were
England and the Low Countries successful?

The superior real wage performance of north-western Europe was due to a
boom in international trade. The English boom began with the export of ‘new
draperies’ in the late sixteenth century. These were light woollen cloths made in
East Anglia and exported to the Mediterranean through London. Between 1500
and 1600, the population of London grew from about 50,000 to 200,000 in
response to the trade-induced growth in labour demand. During the Common-
wealth, Cromwell initiated an active imperial policy, and it was continued through
the eighteenth century.14 In a mercantilist age, imperialism was necessary to
expand trade, and greater trade led to urbanization. Between 1600 and 1700,
London’s population doubled again, and by 1800 it approached one million. In
the eighteenth century, urbanization picked up throughout England as colonial
trade increased and manufacturing oriented to colonial markets expanded.
Between 1500 and 1800, the fraction of the English population living in settle-
ments of more than 5,000 people increased from 7 per cent to 29 per cent. The

14 P. K. O’Brien, ‘It’s not the economy, silly, it’s the navy’ (unpub. paper, 2006).

364 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

share of the workforce in agriculture dropped from about 75 per cent to 35 per
cent. Only the Low Countries, whose economies were also oriented to interna-
tional trade, experienced similarly sweeping structural transformations. In the
eighteenth century, the Dutch and the English had much more trade per capita
than other countries in Europe. Econometric analysis shows that the greater
volume of trade explains why their wages were maintained (or increased) even as
their populations grew.15

Coal deposits were a second factor contributing to England’s unusual wage and
price structure. Coal has a long pedigree as an explanation for Britain’s industrial
success, and Wrigley put it on the modern research agenda.16 I add two points to
the discussion.

First, coal was not just abundant in Britain—it was cheap, at least in northern
and western Britain on or near the coalfields. Figure 5 shows the price of energy
in leading cities in the early eighteenth century. London did not have particularly
cheap fuel at that time; Newcastle, however, did.The difference in the energy price
between the two cities equals the cost of shipping the coal from the Tyne to the
Thames. Despite an ocean route, transportation accounted for most of the price of
coal in London. Coal prices at other cities in northern and western Britain were
similar to those in Newcastle—at least once canal improvements reduced internal
shipping costs. Except perhaps southern Belgium, no region anywhere in the world
had the same combination of large population and cheap energy. Belgian coal
output, however, was only 13 per cent of Britain’s in 1800, and the return from
inventing coal using technology was correspondingly reduced.

Cheap fuel was important for two reasons. Firstly, inexpensive coal raised the
ratio of the price of labour to the price of energy (figure 4), and thereby contrib-
uted to the demand for energy-using technology. In addition, energy was an
important input in the production of metals and bricks, which dominated the
index of the price of capital services. Cheap energy contributed to the fall in capital
prices relative to wages, and thus contributed to the incentive to substitute capital
for labour.

15 Allen, ‘Poverty and progress’.
16 Jevons, Coal question; Nef, Rise; Hatcher, History; Smil, Energy; Pomeranz, Great divergence; Sieferle, Subter-

ranean forest; Wrigley, Continuity.

6
7
8
9

10

5
4
3
2
1
0

A
m
ste
rd
am
Lo
nd
on
Pa
ris
St
ra
sb
ou
rg
N
ew
ca
stl
e
Be
iji
ng

Figure 5. Price of energy, early 1700s
Source: See text.

INDUSTRIAL REVOLUTION 365

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

Secondly, coal was a ‘natural’ resource, but the coal industry was not a natural
phenomenon. Some coal was mined in the middle ages.17 It was the growth of
London in the late sixteenth century, however, that caused the coal industry to
take off. As London grew, the demand for fuel expanded, and the cost of firewood
and charcoal increased sharply as fuel was brought from greater distances. Coal,
on the other hand, was available in unlimited supply at constant real cost from the
fifteenth to the nineteenth century.18 In the late middle ages, coal and charcoal sold
at about the same price per BTU in London. The market for coal was limited to
blacksmithing and lime burning. In all other uses, sulphur made coal an inferior
fuel. As London’s population exploded in the late sixteenth century, the demand
for fuel rose, as did the prices of charcoal and firewood. By 1585, wood fuel was
selling for twice the price of coal per BTU.That differential made it worthwhile for
buyers to figure out how to substitute coal for wood—in fact, a difficult
problem19—and shipments of coal from Newcastle to London began their rapid
growth. The take-off of the coal industry was thus due to the growth of London.
Since this was due to the growth of international trade, the exploitation of Britain’s
coal resources was the result of the country’s success in the global economy as well
as the presence of coal in the ground.

The Dutch cities provide a contrast that reinforces the point and throws light on
the important question of why the industrial revolution happened in Britain rather
than the Netherlands.20 In an important respect that question is badly formed.The
cities of the Low Countries were the analogues of London, but the industrial
revolution happened on the coalfields of northern and western Britain where
energy was much cheaper.Their counterparts were the coal deposits that stretched
from north-eastern France across Belgium and into Germany. This coal was as
useful and accessible as Britain’s. With the exception of the mines near Mons and
Liège, however, Continental coal was largely ignored before the nineteenth
century. The pivotal question is why city growth in the Netherlands did not
precipitate the exploitation of Ruhr coal in a process parallel to the exploitation of
northern English coal. Urbanization in the Low Countries also led to a rise in the
demand for fuel. In the first instance, however, it was met by exploiting Dutch
peat. This checked the rise in fuel prices, so that there was no economic return to
improving transport on the Ruhr or resolving the political-taxation issues related
to shipping coal down the Rhine. Once the Newcastle industry was established,
coal could be delivered as cheaply to the Low Countries as it could be to London,
and that trade put a ceiling on the price of energy in the Dutch Republic that
forestalled the development of German coal. This was portentous: had German
coal been developed in the sixteenth century rather than the nineteenth, the
industrial revolution might have been a Dutch-German breakthrough rather than
a British achievement.

While the high-wage economy of London led to the exploitation of cheap coal,
the availability of cheap energy also sustained the high-wage economy. British
businesses could pay high wages and compete in the international economy only

17 Hatcher, History.
18 The real price of coal was constant from 1450 to 1850. See also Clark and Jacks, ‘Coal’
19 Nef, Rise.
20 Pounds and Parker, Coal and steel; de Vries and van der Woude, First modern economy; Unger, ‘Energy

sources’.

366 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

if their efficiency was exceptional or if another input was cheap. (This relationship
is the ‘factor price frontier’ of neoclassical economics.) Coal was that other input
that enabled firms to pay a high wage while remaining profitable. Contemporaries
were aware of this advantage. Glassmaking was one industry where the French
were still ahead of the English in the late eighteenth century. Delaunay Deslandes,
the director of Saint-Gobain, was initially sceptical that the English could success-
fully compete against the French since English wages were one-third higher than
French and the standard of living was accordingly superior:

Given the manner in which the French and English lived . . . they could never make
plate [glass] which could enter into competition with ours for the price. Our Frenchmen
eat soup with a little butter and vegetables. They scarcely ever eat meat. They sometimes
drink a little cider but more commonly water. Your Englishmen eat meat, and a great
deal of it, and they drink beer continually in such a fashion that an Englishman spends
three times more than a Frenchman.21

The burden of high wages in England, however, was offset by cheap energy. In
prospectuses of the 1770s, the fuel cost of English glass production was estimated
to be only one-sixth of that in France.22 The same offset occurred in iron produc-
tion. Richard Reynolds of the Coalbrookdale Iron Company wrote to Earl Gower,
President of the Privy Council, in 1784 to object to a proposed tax on coal on the
grounds that ‘coal . . . is the only article that in any degree compensates for our
high price of labour’.23 The shift from charcoal to coal in industrial processes
during the seventeenth and eighteenth centuries—a shift that required the solution
of many technical problems—gradually lowered the average price of energy in the
English economy and underpinned the rise in the average wage.

The remarks of Deslandes and Reynolds have an important further implication.
The high wage in England meant that English workers could buy more food than
many of their counterparts abroad. It is conceivable that eating more food might
have raised the productivity of English labour, offsetting the high wage and
reducing the incentive to mechanize production. However, the import of Deslan-
des’ and Reynolds’ comments is that any such increase in productivity was not
enough to offset the high wage. Deslandes is particularly compelling since his
remarks were not part of a self-interested plea and since he was explicit about the
consumption implications of the wages. In his view, the beef and beer enjoyed by
the English worker did not compensate his employer for his high wage. English
labour was still more expensive than French labour, and the high English wage had
to be offset by some other saving. To that we turn.

III

Britain’s high-wage, cheap-energy economy was an important determinant of the
pace and character of technical change.There were both demand and supply links,
and I begin with the former. I emphasize process innovations. Product innovations
that imitated Asian trade goods like porcelain and cotton cloth were also impor-

21 Quoted in Harris, ‘Saint-Gobain’, p. 67, n. 42.
22 Ibid., p. 38.
23 Quoted by Raistrick, Dynasty, p. 97.

INDUSTRIAL REVOLUTION 367

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

tant, but their manufacture involved process innovations as well, since production
methods had to be redesigned to suit British conditions.24

Britain’s industrial processes diverged from those used elsewhere since Britain’s
high (and rising) wage induced a demand for technology that substituted capital
and energy for labour. At the end of the middle ages, there was little variation
across Europe in capital intensity. As the wage rose relative to the price of capital
in Britain, it was increasingly desirable to substitute capital for labour and that is
what happened. Sir John Hicks had the essential insight: ‘The real reason for the
predominance of labour saving inventions is surely that . . . a change in the relative
prices of the factors of production is itself a spur to innovation and to inventions
of a particular kind—directed at economizing the use of a factor which has become
relatively expensive’.25 Habakkuk used this theory to argue that high wages led
Americans to invent labour-saving technology in the nineteenth century.26 A
similar situation obtained in eighteenth-century Britain.27 It was the prequel to
nineteenth-century America.

We can clarify the influence of prices on invention, if we recognize that it
involved the two stages that Edison called ‘inspiration’ and ‘perspiration’. The first
stage, inspiration, was not the field of action of economics. Today the search for
new ideas may be systematic and driven by commercial considerations, but in the
eighteenth century exogenous factors probably loomed larger. The ideas incorpo-
rated into the inventions of the industrial revolution were either the products of
exogenous scientific advances, or acts of genius, or inadvertent by-products of
normal operations (learning by doing), or they were copied from other activities.

The second stage of invention was R&D—the perspiration that turned a concept
into a new product or a process. Leonardo da Vinci is famous as an ‘inventor’ since
he sketched hundreds of novel machines, but his reputation is overblown in that he
rarely did the hard work needed to turn drawings into functioning prototypes. Our
interest is in the technologies that were used in the industrial revolution, and use
required R&D as well as a ‘eureka’ moment. While new ideas may not have been
economically conditioned, R&D certainly was, since the decision to incur costs to
operationalize a technical idea was an economic one. Prices influenced techno-
logical development through their effect on the profitability of R&D.

The essential idea is that inventors spent money to develop ideas when they
believed the inventions would be useful, and in particular, when their social
benefits exceeded the costs of their invention.When this condition was satisfied, an
inventor with an enforceable patent could recoup the development costs through
royalties. Even when private gain was not the object—for instance, in the case of
Abraham Darby II, who discovered how to make wrought iron from coke pig
iron—social utility was still the aim, so our analysis has force. Whether or not an
inventor got a royalty, a mundane point is crucial: an invention was socially useful
only if it was used. If it was not used, there was no point in inventing it. Invention,

24 Berg, Luxury and pleasure.
25 Hicks, Theory, pp. 124-5.
26 Habakkuk, American and British technology. Economists have since debated how to formalize these ideas

(David, Technical choice, pp. 19–91; Temin, ‘Notes’; Ruttan, Technology; Ruttan and Thirtle, Role of demand;
Acemoglu, ‘Factor prices’).

27 Fremdling, ‘Continental responses’, pp. 168–9, entertains this possibility, as does Mokyr, ‘Editor’s intro-
duction’, pp. 87–9, who also raises many objections to it.

368 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

thus, depended on adoption. Adoption, in turn, depended on factor prices, and
that meant that factor prices influenced R&D and hence invention.

We can see how factor prices affected adoption and R&D with a standard
isoquant model. The model makes five points: (1) a biased technical change saved
one input disproportionately and reduced costs the most where that input was
most expensive; (2) techniques were worth inventing only if they were used; (3) a
new technique was not worth using everywhere; (4) countries with high wages
found it profitable to develop a broader range of techniques with high capital–la-
bour ratios than did low-wage countries; (5) larger markets increased the profit-
ability of R&D and led to more invention.

These points are illustrated in figure 6, which contrasts high-wage and low-wage
firms. The curved isoquant through H and L connects the quantities of capital and
labour needed to produce one unit of output. H is the input combination used by
the high-wage firm, and it has a higher capital–labour ratio than the input com-
bination used by the low-wage firm L. The straight lines tangent to the isoquant at
H and L connect equal cost combinations of capital (K) and labour (N) where the
unit cost in production C = rK + wN and where r and w are the rental price of
capital and the wage rate. Each straight line plotted in figure 6 is of the form
K = C/r + (w/r)N. Its slope equals the wage relative to the price of capital (hence
a steeper line denotes the high-wage firm) and C/r is the point where the line
intersects the K axis. Hence, a higher intersection point indicates higher produc-
tion cost (C). In figure 6 CH/rH indicates the unit cost of the high-wage firm, and
CL/rL the cost of the low-wage firm.

Now consider a potential new technology represented by the point T connecting
a new combination of capital and labour that can produce one unit of output. T is
a biased technical change: it uses more capital and less labour than either H or L.
Would T be used? It would if and only if it lowered costs, and that is the case for

K

CH/rH

T

CL/rL

I H

L

II
III

L

Figure 6. Biased technical change
Source: See text.

INDUSTRIAL REVOLUTION 369

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

the high-wage firm. We know this since a straight line through T that is parallel to
the isocost line through H (hence, representing the same w/r) has a lower inter-
section point on the K axis and, hence, represents lower unit costs. For the
low-wage firm, T would raise costs by the same argument. A technology like T is
worth using—and thus worth inventing—only for the high-wage firm.

The two isocost lines divide the area below them into three spaces. New
technologies in I would be adopted only by the high-wage firm, technologies in III
only by the low-wage firm, and technologies like II by either firm. Some new
technologies are useful to any firm, while others are useful only to firms in
particular factor price situations. Factor prices affect technological evolution
because the adoption and invention of new techniques in sectors I and III depend
on factor prices.

The high-wage and the low-wage firms have opposite incentives to invent
technique T. It would be pointless for the low-wage country to invent it since it
would not be used. It might be worth inventing in the high-wage country, but the
incentive depends on benefits net of development costs. A technique like T in
sector I would lower operating costs for high-wage firms, and that saving gener-
ates the demand for the technology, that is, creates a return for someone to
invent it. But invention requires R&D to actualize the idea. Whether the demand
for the technology is enough to motivate its development depends on the balance
between the saving in operating costs and the cost of the R&D. Scale plays a role
here since the R&D cost must be amortized over the output and compared to the
reduction in unit operating costs. The total cost of production (inclusive of R&D)
with the new technique is C* = C + D/q where D is the development cost and q
is total production over the life of the technology. The total cost line inclusive of
R&D costs is K = C* + (w/r)N = C/r + (D/q)/r + (w/r)N, that is, the K intercept
shifts up by the amortized R&D cost, so the total cost line is above the old one.
The larger is q, the less is the upward shift in the isocost line inclusive of R&D
cost. Two possibilities need to be distinguished at this stage. The first is that the
isocost line rises but remains below the isocost line with the old technique. In
that case, it is profitable to develop (that is, invent) the new technique T. The
second possibility is that the new isocost line rises above the original isocost line.
In that case, it is not profitable to invent the new technique because the market
is too small. Of course, if some other firm or country paid the R&D costs and the
new technology were freely available, it would be adopted because it would cut
operating costs. The size of the market affected the profitability of invention
through the amortization of R&D costs.28

Figure 6 identifies the conditions under which R&D was profitable, and they
drove much private sector R&D. They also highlight the shortcomings of non-
commercial R&D, such as some well-known technology initiatives of the French
state. One was Cugnot’s fardier, a steam tractor developed by the military to
pull cannons across fields. Cugnot built a high-pressure steam engine and
installed it on a vehicle. The fardier was a technical success, but the project was
abandoned since it consumed too much fuel and sank into the mud. High-
pressure steam engines were successfully used for traction only when both prob-
lems were solved by putting them on rails to pull wagons in British coal mines.

28 Acemoglu, ‘Factor prices’.

370 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

A second example was Vaucanson’s fully automated silk loom. This was a tre-
mendous technological achievement, but it was never used commercially since it
was far too capital-intensive.29 These technologies show the force of figure 6
in that they were not profitable to invent because they were not profitable to
use.

I

V

To apply the model to Britain, we must show that eighteenth-century British
inventions were biased towards saving labour and using capital and energy. These
biases meant that they were worth using at British factor prices but not at prices
prevailing elsewhere.

Thanks to Adam Smith, the pin factory is the most famous production process
of the eighteenth century, and this example highlights many of the issues. Smith
argued that high productivity was achieved through a division of labour among
hand workers.30 It is very likely that he derived his knowledge from Diderot and
d’Alembert’s Encyclopédie, since both texts divide the production process into 18
stages, and that cannot be a coincidence.31 Indeed, Smith seems to have used the
Encyclopédie for the exact purpose that Mokyr suggests—to find out about the
latest technology.32

There is a difficulty, however. The Encyclopédie’s account is based on the pro-
duction methods at l’Aigle in Normandy. This was not the state-of-the-art practice
as carried on in Britain. The first high-tech pin factory in England was built by the
Dockwra Copper Company in 1692, and it was followed by the Warmley works
near Bristol in the mid-eighteenth century.33 The latter was a well-known tourist
destination, and Arthur Young visited it.34 Both mills were known for their high
degree of mechanization, and they differed most strikingly from Normandy in the
provision of power. In L’Aigle, machines were propelled by people turning fly
wheels that looked like spinning wheels. In contrast, the Warmley mill was driven
by water power. Since the natural flow of the stream could not be relied on, a
Newcomen steam engine was used to pump water from the outflow of the water
wheel back into the reservoir that supplied it: ‘All the machines and wheels are set
in motions by water; for raising which, there is a prodigious fire engine, which
raises, as it is said, 3000 hogsheads every minute’.35 Powering the mill in this way
immediately eliminated the jobs of the wheel turners (their wages amounted to
one-sixth of the cost of fabricating copper rod into pins) and probably other jobs
as well. Many French workers, for instance, were employed scouring pins. This
activity was done with large machines driven by water power at English needle

29 Doyon and Liaigre, Jacques Vaucanson, pp. 144, 214–15, 230–2.
30 Smith, Wealth of nations, pp. 3–12.
31 Diderot and d’Alembert, eds., Encyclopédie, vol.V, pp. 804–7, vol. XXI, pp. 1–8, ‘épinglier’. Peaucelle (‘Adam

Smith et les encyclopédistes’; ‘Raisonner sur les épingles’; Adam Smith) has examined Smith’s sources very
carefully and identified several additional French publications that he argues Smith relied on. All of these sources
describe production in Normandy.

32 Mokyr, Gifts of Athena, pp. 68–72.
33 Hamilton, English brass, pp. 103, 255–7.
34 Russell, England displayed.
35 Young, Six weeks tour, p. 138.

INDUSTRIAL REVOLUTION 371

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

factories at the time.36 Arthur Young observed that the Warmley works ‘are very
well worth seeing’. It is a pity that Adam Smith relied on the French Encyclopédie
to learn about the latest in technology rather than travelling with Arthur Young.

Why did the English operate with a more capital- and energy-intensive technol-
ogy than the French? L’Aigle was on a river, and water power drove a forge in the
town, so geography was not a bar (indeed, the steam engine at Warmley shows that
water power was possible almost anywhere if you were willing to bear the cost of
a steam engine). The Swedish engineer R. R. Angerstein visited Warmley in the
1750s and noted that ‘the works uses 5000 bushels of coal every week, which,
because they have their own coal mines, only costs three Swedish “styfwer” per
bushel’, which was about half the Newcastle price.37 In addition, English wages
were considerably higher than French wages. Innovation in pin making is an
example of factor prices guiding the evolution of technology.

These considerations operated generally. Much of the technology of the indus-
trial revolution depended on coal. This included many metallurgical applications
(for example, using coke to smelt iron and puddling to refine it) and the steam
engine, invented by Newcomen in the first decade of the eighteenth century.These
technologies all increased coal use relative to other inputs and were only profitable
to use at British factor prices. Fremdling, for instance, has shown that British iron
making technology was not cost-effective in France and Germany before the
middle of the nineteenth century.38 The Newcomen steam engine was profligate in
its use of fuel. Desaguliers, an early enthusiast of steam power, noted that it was
only useful where ‘coals are cheap’ as was the case at Warmley, ‘But it is especially
of immense Service (so as to be now of general use) in the Coal-Works, where the
Power of the Fire is made from the Refuse of the Coals, which would not otherwise
be sold’.39 Steam engines in the eighteenth century were mainly used in coal mines
where coal was effectively free. As a result, they were mainly used in Britain, with
Belgium coming a distant second.40 The coal-using technology was profitable to
invent in Britain but not in France or Germany because the low price of coal
meant that these techniques were only profitable to use in Britain. They were
profitable to use in Belgium, but the small size of the Belgian industry meant that
development costs per unit of output were much higher than in Britain, and this
consideration militated against carrying out the R&D in Belgium.

The other famous inventions of the industrial revolution were the machines to
spin cotton. They were also biased technical changes that raised capital–labour
ratios and were profitable to use—hence to invent—only in Britain.41 Arkwright’s
water frame was the most far-reaching since it inaugurated factory production.
Using 1784 prices to value inputs, the average total cost of 16 count cotton yarn
dropped from 35d. per lb when it was produced in the domestic system with hand
technology to 28 d. per lb when produced in an Arkwright mill of the period. The

36 Early eighteenth-century water-driven scouring machinery is still in operation and can be seen at the Forge
Mill Needle Museum, Redditch.

37 Angerstein, Diary, p. 138. I thank Martin Dribe for help in deciphering the Swedish stwyfer.
38 Fremdling, ‘Transfer patterns’.
39 Desaguliers, Course, vol. II, pp. 464–5.
40 Kanefsky and Robey, ‘Steam engines’, p. 171; Redlich, ‘Leaders’, p. 122; Tann, ‘Makers’, pp. 548, 558;

Hollister-Short, ‘Introduction’, p. 22.
41 The figures discussed in this paragraph and the next are explained more fully in Allen, British industrial

revolution, pp. 182–216

372 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

mill involved mechanical carding as well as spinning, and the system of material
flow worked out in Cromford Mill #2. Labour costs fell 8d. per lb, but that saving
was offset by a rise in capital costs from about 1d. to 2d. per lb. The capital–labour
ratio was almost five times higher with the Arkwight system.

The Arkwright mill was much more profitable in Britain than it would have been
in France. In Britain in the late 1780s, cotton mills were built at a cost of about £3
per spindle. In 1784, the water frames in the Papplewick Mill near Nottingham
had a rated capacity of 0.125 lbs per 12 hour shift or 37.5 lbs per year assuming
they were operated six days per week and 50 weeks per year. Assuming a saving in
operating costs of 8d. per lb and a 10-year life for a cotton mill, the rate of return
was 40 per cent per year. This was considerably greater than 15 per cent, which
investment in fixed capital could realize.42 The rate of return in France equalled the
English rate of return multiplied by the ratio of the English wage relative to the
France wage divided by the price of capital in England relative to the price in
France. The latter was computed using the prices of iron, copper, and timber. The
implied French rate of return was 9 per cent. This calculation probably overstates
French profitability since it takes no account of the local supply of ‘high tech’
components like gears in Lancashire (to be discussed shortly), and their absence
in France. The different profit rates explain why there were about 150 large-scale
mills operating in Britain in the late 1780s and only four in France. The difference
in behaviour does not reflect any difference in ingenuity but rather the choice in
England to save the expensive input labour.

When it was invented, British technology was profitable to use only in Britain (or
Belgium, in the case of coal). That condition did not persist, however. Inventors
modified existing practice to cut costs. This was ‘local learning’, and it led to the
saving of all inputs.43 In the case of the steam pumping engines, for instance, coal
consumption was cut from 44 lbs/hp-hr (horsepower per hour) in 1727 to 2 lbs in
1860 through the efforts of Smeaton, Watt, and Trevethick, as well as by the
collective invention carried out by the owners of copper and tin mines in Corn-
wall.44 The process of technological improvement in this period was neutral and
meant that the steam engine became useful in many activities and places where it
had not been practical previously. The culmination of this process was the com-
pound condensing marine engine that allowed steam vessels to displace sailing
vessels in voyages between Britain and east Asia.45 A similar process characterized
the cotton spinning industry, where Lancashire engineers halved the capital
requirements of Arkwright-style mills as well as cutting labour costs.46

These technological developments facilitated the spread of the industrial revo-
lution beyond Britain during the nineteenth century. Figure 7 illustrates the
process. Improving technology by modifying existing practice meant that both
capital and labour were saved.The trajectory of improvement is represented by the
drift of the point representing the new technology from T towards the origin.
Initially, the improved versions of the new technique were still profitable to use

42 Allen, ‘Engel’s pause’, p. 421; C. K. Harley, ‘Prices and profits in cotton textiles during the industrial
revolution,’ Oxford University, discussions papers in economic and social history, no. 81 (2010).

43 David, ‘Common agency contracting’.
44 Nuvolari, Making; idem, ‘Collective invention’; idem, ‘Making’.
45 Harley, ‘Shift’.
46 Chapman, ‘Fixed capital formation’, p. 253.

INDUSTRIAL REVOLUTION 373

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

only in the high-wage economy. This graphical depiction corresponds to the
historical stage when Britain was using steam engines, mechanical spinning, and
coke smelting, and was, moreover, extending its lead by improving their design. As
Britain pulled further ahead, however, other countries continued to ignore the new
methods. This ‘failure’ easily led to accusations of entrepreneurial failure or inad-
equate engineers, but the real explanation, indicated in figure 7, is that the new
technique was still too expensive to use in a low-wage country. Accordingly, a
critical juncture is represented by the ‘tipping point’ where the line from T to the
origin crosses the isocost line of the low-wage economy. When that happened, it
suddenly became profitable for the low-wage economy to adopt the new
technique—indeed, only in its most advanced form. (The new situation is shown
graphically by the dotted isocost line parallel to, but below, the original isocost line
of the low-wage economy. This new line shows that using a newly discovered input
combination below the tipping point reduced costs in the low-wage economy.)
Suddenly, the industrial revolution spread beyond Britain in a Gerschenkronian
‘great spurt’.

V

The high wages of the British economy in the eighteenth century were an impor-
tant reason why it was profitable to use—and to invent—labour-saving technology
like the spinning jenny. However, there have been other high-wage economies that
did not produce such inventions. Fifteenth-century Europe was one example
(figure 2).Why was eighteenth-century Britain different? A tempting answer is that
the scientific revolution of the seventeenth century led to a greater understanding
of the natural world and allowed new technologies to be invented. In terms of
figure 6, scientific discoveries created new points like T.

Trajectory of micro-improvements

K
CH/rH
T
I H

X
CL/rL

L
II
III
L

Figure 7. The trajectory of micro-improvements
Source: See text.

374 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

One difficulty with this answer is that the role of science in the industrial
revolution was extensively discussed in the 1960s and dismissed by most histori-
ans.47 However, there is a good case for claiming that these historians went too far,
and that scientific discoveries underpinned important technology in the industrial
revolution. The reason that Hall, for instance, could find no link between scientific
discovery and new technology was because he only analysed the period
1760–1830. In the case of Watt, Hall concluded—correctly—that the theory of
latent heat contributed nothing important to the invention of the separate con-
denser. The trouble with this argument is that the scientific discoveries that
mattered for the industrial revolution were made before 1700 and not after 1760.

The most important science related to atmospheric pressure, namely, the find-
ings that the atmosphere had weight and that steam could be condensed to form
a vacuum.48 Galileo first considered the problem of why a suction pump could not
raise water more than about 10 metres and set his secretary Torricelli to work on
it. Torricelli invented the barometer and weighed the atmosphere in 1643. Atmo-
spheric pressure became the hot topic in experimental physics and noteworthy
experiments were carried out by Otto von Guericke, Robert Boyle, Robert Hooke,
Christiaan Huygens, and Denis Papin. Thomas Savery invented a vacuum pump
for draining mines that applied these discoveries. Newcomen applied the same
ideas in his steam engine, which was also intended to drain mines. He began
working on the problem around 1700, apparently built an engine in Cornwall in
1710, and finally erected his well-known engine at Dudley in 1712. Newcomen
could not get a patent in his own right and was forced to do a deal with Savery,
whose pump patent was deemed to cover Newcomen’s engine. The steam engine
was one example of industrial technology derived from science.

A second link from seventeenth-century science involved not only a discovery,
but also the active participation of first-class scientists—Christiaan Huygens and
Robert Hooke—in the production process. Several inventions relating to time-
keeping were involved. The first was the invention of the pendulum clock. Chris-
tiaan Huygens proved mathematically that a cycloid was an isochronous curve so
that a flexible pendulum restrained between cycloid guides would have a regular
swing irrespective of its amplitude. Armed with this insight, he designed the
pendulum clock, which dramatically increased the accuracy of timekeeping.
Huygens was trying to solve the longitude problem, but the pendulum clock did
not work well at sea, so he improvised further. Around 1675, he invented the
balance spring, which made an accurate watch possible and, indeed, installed it in
a watch. Robert Hooke, the Curator for Experiments of the Royal Society and
another scientific luminary, independently conceived of the balance spring perhaps
as early as 1660, although he did not apply the idea until he heard of Huygens’s
work.49 In themselves, clocks and watches were peripheral to the industrial revo-
lution, but their large-scale production had important spin-offs. The improve-
ments in clock and watch design made them more desirable, and their production
grew rapidly.Their moving parts were systems of gears, and each had to be laid out

47 Landes, Unbound Prometheus, pp. 113–14, 323; Mathias, ‘Who unbound Prometheus?’; Hall, ‘Industrial
revolution’.

48 Landes, Unbound Prometheus, p. 104; Cohen, ‘Inside Newcomen’s fire engine’.
49 Weiss, Watch-making, pp. 111–12.

INDUSTRIAL REVOLUTION 375

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

and cut by hand. Hooke designed the first machine to do this.50 The growth of the
watch industry prompted steady improvement in the design of these machines.
The result was the mass production of cheap, accurate gears.

Inexpensive gears revolutionized the design of machinery. Gears replaced levers
and belts (as in the spinning wheel) to control, direct, and transmit power. Mills
had used gears in this way in the middle ages, but these gears were large, crude,
and made of wood. The gears of the industrial revolution were small, refined, and
made of brass or iron. Arkwright referred to the gearing in his water frame as
‘clockwork’ since this system of construction was adapted from clocks and
watches. ‘Clockwork’ was used quite generally to control power in machinery in
the nineteenth century, so gearing was the general purpose technology that
effected the mechanization of industry.51

The watch industry was a key to explaining Arkwright’s success and the growth
of the cotton industry in Lancashire. Arkwright did not sell water frames; entre-
preneurs had to assemble their own engineering departments to construct them.
In the late 1780s and early 1790s, the Quarry Bank Mill employed half-a-dozen
clock and watch makers over the course of many years to construct the ‘clockwork’
for the water frames. At the time, there were over 150 Arkwright mills, so on the
order of 1,000 of these specialists were employed. Where did they come from? As
it happens, most of the world’s watch movements were made in one
place—southern Lancashire. Landes believed the watch industry was British
because the high-wage economy of Britain created a large domestic market for
clocks and watches52. One reason that cotton production was mechanized in
Lancashire (rather than in the Netherlands, for instance) was because the supply
of high-quality, cheap gears was far greater there than elsewhere, as was the supply
of skilled workmen to assemble them. In addition, the machinery for cutting watch
gears was redesigned to produce gears for water frames—first from brass, later
from iron. Standardized gears were made by specialist firms and sold to mills.53

The ‘clock work’ of the water frame was a spin-off of the watch industry. So we can
trace connections from the water frame (and other machinery of the nineteenth
century) back to the discoveries of leading scientists, Huygens and Hooke, in the
mid-seventeenth century.

VI

Cultural shifts also contributed to the industrial revolution. They increased the
quality of would-be inventors and technical personnel, thereby reducing the cost of
R&D and increasing the range of projects that were profitable to undertake. There
were shifts in both elite culture and popular culture.

Mokyr’s model of the industrial enlightenment emphasizes changes in elite
culture as a cause of the industrial revolution.54 On the intellectual plain, the
industrial enlightenment refers to the application of the scientific method (experi-
mentation, generalization, mathematization) to the study of technology. ‘Most

50 Ibid., pp. 153–6.
51 Lipsey, Carlaw, and Bekar, Economic transformations.
52 Landes, Revolution, pp. 238–9.
53 Hills, Power in the industrial revolution, pp. 230–49.
54 Mokyr, Gifts of Athena, pp. 28–77.

376 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

techniques before 1800 emerged as a result of chance discoveries, trial and error,
or good mechanical intuition and often worked quite well despite nobody’s having
much of a clue as to the principles at work’. The industrial enlightenment changed
all that. The scientific study of technology ‘would explain the timing of the
Industrial Revolution following the Enlightenment and—equally important—why
it did not fizzle out like similar bursts of macroinventions in earlier times’.55

On the social plain, the industrial enlightenment involved a small number of
unusual people working in concert. ‘The crucial elements were neither brilliant
individuals nor the impersonal forces governing the masses’—for instance, factor
price movements like those emphasized here—‘but a small group of at most a few
thousand people who formed a creative community based on the exchange of
knowledge’.56 At the highest level, information was exchanged at the Royal
Society. More people were involved in provincial ‘scientific societies, academies,
Masonic lodges, coffee house lectures’,57 and similar venues. Individual
exchanges were important. When Trevithick invented the high-pressure steam
engine, he checked with the mathematician Davies Gilbert to see how much
pressure drop he would lose by not including a separate condenser (the answer
being one atmosphere).58

Mokyr’s examples of inventors who ‘embodied the Industrial Enlightenment’
stand out as cultivated gentlemen committed to science and Enlightenment
culture generally. They were active in learned societies. Benjamin Franklin was an
archetype. He studied science. He conducted his own experiments, notably those
involving electricity. He published his results. He was an inventor (of the lightning
rod and bifocals). He corresponded with leading scientists, and he established the
America Philosophical Society to advance this kind of work.59

Josiah Wedgewood is another example: ‘He was, by all accounts, a compulsive
quantifier, an obsessive experimenter, and an avid reader of scientific literature’.
He was a member of the Royal Society and corresponded with the leading
scientists of the day. There were not many people like him: ‘It might be objected
that Wedgwood was not typical, but the argument of this book is that such
unrepresentativeness is the heart of the process of technological change . . .
averages are . . . not very important: a few critical individuals drive the process’.60

Other famous exemplars were John Smeaton, James Watt, and Edmund
Cartwright.

Britain was different from the US in the social background of its inventors. Khan
and Sokoloff found that great British inventors in the period before 1820 were far
more likely to come from an elite or professional family than were great US
inventors.61 Many of the Enlightenment figures identified by Mokyr provide
examples of this. John Smeaton’s father, for instance, was the son of an attorney
and attended Leeds Grammar School. James Watt was the son of a merchant, who

55 Ibid., pp. 32, 39.
56 Ibid., p. 66.
57 Mokyr, Gifts of Athena, pp. 52–3.
58 Burton, Richard Trevithick, pp. 59–60.
59 Mokyr, Gifts of Athena, pp. 42–3.
60 Ibid., pp. 52–3.
61 Z. Khan and K. Sokoloff, ‘The evolution of useful knowledge: great inventors, science and technology in

British economic development, 1750–1930’, Economic History Society Conference, Exeter, 31 March (2007).

INDUSTRIAL REVOLUTION 377

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

was also mayor of Greenock. Watt attended Greenock Grammar School where he
studied Latin and Greek. Edmund Cartwright came from a wealthy Northamp-
tonshire family and attended Wakefield Grammar School. Henry Cort’s father was
a merchant and at one time was mayor of Kendal. Men like this acquired their
cultivation, and possibly their enlightened ideas, through their family background
and education.62

While inventors from advantaged backgrounds were over-represented in Britain
compared to the US, Khan and Sokoloff found many British inventors from
modest backgrounds.The exemplar of the industrial enlightenment, Josiah Wedge-
wood, was the son of a potter, who walked seven miles each day to a small school,
and was apprenticed to a potter. He acquired his experimental outlook through his
own efforts. Many of the great inventors had similar upbringings, although they
did not have the intellectual accomplishments of Wedgewood. Richard Arkwright
was the son of a poor tailor, attended a night school, and was apprenticed to a
barber. James Hargreaves came from a very poor part of Lancashire, was unedu-
cated, and spent most of his life as a hand loom weaver. Samuel Crompton was the
son of an unsuccessful farmer and part-time hand loom weaver and attended a
local school. As a child, he spun and wove to supplement the family income.
Abraham Darby I was the son of a part-time farmer and nailer and was appren-
ticed to a malt maker. Benjamin Huntsman was the son of a farmer and was
apprenticed to a clock maker. Richard Trevithick was educated in a village school,
and his father was a copper miner.

Several regularities are striking about these inventors. Firstly, they were brought
up in the non-agricultural economy, and indeed their parents (aside from Hunts-
man) had left agriculture in whole or in part. They were part of the urban or
proto-industrial economy. Secondly, none of their parents were labourers. Thirdly,
many of them had trade backgrounds. The typical training from someone of that
social stratum involved several years in a village school where they learned reading,
writing, and arithmetic, followed by an apprenticeship.63 While we do not have full
details on all of these inventors, their formations are consistent with this pattern.

The urban and rural manufacturing economies were created by the commercial
success of Britain in the early modern period. In the middle of the eighteenth
century, about 55 per cent of the British population was non-agricultural. Along
with the Low Countries, England led Europe in this regard.64 The high-wage
economy, of which the inventors were a part, generated the income to purchase the
education and training they received. Indeed, literacy rates in north-western
Europe were much higher than elsewhere since so many people around the North
Sea could afford to send their children to school. Sir Frederick Eden summarized
the spending of a labourer in Ealing, and it included 6d. per week to send his six-
and eight-year-old sons to school.65 If the labourer had had to cut back to live on
an Italian wage, that expenditure would probably have been eliminated. High
wages, town living, a commercial culture, and widespread education constituted a
distinctive popular culture that produced inventors.

62 The biographical details in this and the next few paragraphs are from the new Dictionary of National
Biography.

63 Humphries, Childhood.
64 Allen, ‘Economic structure’, p. 11.
65 Eden, State of the poor, vol. II, pp. 433–4.

378 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

The propensity to invent may also have been strengthened as enlightened
thinking worked its way down the social hierarchy.66 Artisans picked up Newto-
nianism from almanacs, science lecturers, and latitudinarian preaching.67 One
example was John Harrison, the clockmaker, whose chronometer solved the lon-
gitude problem. His father was a carpenter and brought him up to the trade;
otherwise, he was self-taught. He met the Enlightenment when a clergyman lent
him a copy of Nicholas Saunderson’s lectures on natural philosophy. Whether this
exposure to Newtonianism inclined him to pursue the longitude problem is, of
course, one question. Another is how representative he was of artisans in general.
Were they exposed to Enlightenment thinking and influenced by it? Presumably, a
counterpart to the rise of the mechanical worldview was a decline in belief in
witchcraft and magic. There is no consensus among historians of popular culture
that such a decline occurred.68 Sharpe has written that ‘Popular scepticism about
magic, and popular receptiveness to Newtonian science, are problems which are in
urgent need of further research’.69 In this circumstance, the case for a widespread
adoption of the Newtonian worldview must remain conjectural.

How important are the industrial enlightenment and the growth of a literate and
numerate class of commercial artisans in explaining the inventions of the industrial
revolution? This is a very difficult question to answer for Britain, since the effects
of cultural change were intermingled with the powerful incentives created by
Britain’s unique factor prices to invent labour-augmenting technologies. Supply
and demand for R&D were both shifting to encourage invention, so their separate
effects are hard to identify. The Continent, therefore, is a more fruitful laboratory
for studying the effects of cultural developments, for factor prices there were more
stable. Scientific discoveries created some new opportunities for R&D—watch
making is an example—but supply factors probably played a larger role. From this
perspective, Continental inventions in the seventeenth and eighteenth centuries
take on a much greater significance, for they show the effects of culture rather than
factor prices or scientific discoveries on the rate of invention.

While Continental history shows that cultural change played a role in stimulat-
ing invention, it also highlights the limits to culturally-induced invention. The
Dutch economy was a high-wage economy, but it missed the industrial revolution
since it lacked cheap coal, a domestic cotton industry, and a watch industry—all
of which were crucial in stimulating British invention. France was not a high-wage
economy, but the state—animated in part by the Enlightenment confidence that
useful technologies could be produced through purposeful activity—financed
R&D to develop steam tractors and automatic looms. Without a factor price
environment making these techniques cost-effective, they were abandoned.
Culture, by itself, could not make up for an inhospitable economic environment.

VII

I have argued that the famous inventions of the British industrial revolution were
responses to Britain’s unique economic environment and would not have been

66 Burke, Popular culture.
67 Jacob, Scientific culture, pp. 99–115; Stewart, Rise of public science; Sharpe, Early modern England, p. 329.
68 Thomas, Religion, pp. 767–800; Briggs, Witches, pp. 327–30; Burke, Popular culture, pp. 274–5.
69 Sharpe, Early modern England, p. 330.

INDUSTRIAL REVOLUTION 379

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

developed anywhere else. This is one reason that the industrial revolution was
British. But why did those inventions matter? The French were certainly active
inventors, and the scientific revolution was a pan-European phenomenon. Would
the French, or the Germans, or the Italians, not have produced an industrial
revolution by another route? Were there no alternative paths to the twentieth
century?

These questions are closely related to another important question asked by
Mokyr: why did the industrial revolution not peter out after 1815? He is right that
there were previous occasions when important inventions were made. The result,
however, was a one-shot rise in productivity that did not translate into sustained
economic growth. The nineteenth century was different—the First Industrial
Revolution turned into Modern Economic Growth. Why? Mokyr’s answer is that
scientific knowledge increased enough to allow continuous invention. Technologi-
cal improvement was certainly at the heart of the matter, but it was not due to
discoveries in science—at least not before 1900. The reason that incomes contin-
ued to grow in the hundred years after Waterloo was because Britain’s pre-1815
inventions were particularly transformative, much more so than Continental
inventions. That is a second reason that the industrial revolution was British and
also the reason that growth continued throughout the nineteenth century.

Cotton was the wonder industry of the industrial revolution—so much so that
Gerschenkron, for instance, claimed that economic growth in advanced countries
was based on the growth of consumer goods industries, while growth in backward
countries was based on producer goods.70 This is an unfortunate conclusion,
however, for the great achievement of the British industrial revolution was, in fact,
the creation of the first large engineering industry that could mass-produce
productivity-raising machinery. Machinery production was the basis of three
developments that provide the immediate explanations for the continuation of
economic growth until the First World War. Those developments were: (1) the
general mechanization of industry, (2) the railroad, and (3) steam-powered, iron
ships. The first raised productivity in the British economy itself; the second and
third created the global economy and the international division of labour that were
responsible for significant rises in living standards across Europe.71 Steam tech-
nology accounted for close to half of the growth in labour productivity in Britain
in the second half of the nineteenth century.72 The application of gears to machin-
ery design had further productivity growth raising effects beyond these.

The nineteenth-century engineering industry was a spin-off of the coal industry.
All three of the developments that raised productivity in the nineteenth century
depended on two things—the steam engine and cheap iron. Both of these, as we have
seen, were closely related to coal. The steam engine was invented to drain coal
mines, and it burnt coal. Cheap iron required the substitution of coke for charcoal
and was prompted by cheap coal. (A further tie-in with coal was geological—
Britain’s iron deposits were often found in proximity to coal deposits.) There were
more connections: the railroad, in particular, was a spin-off of the coal industry.
Railways were invented in the seventeenth century to haul coal in mines and from

70 Gerschenkron, Economic backwardness.
71 O’Rourke and Williamson, Globalization.
72 Crafts, ‘Steam’.

380 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

mines to canals or rivers. Once established, railways invited continuous experimen-
tation to improve road beds and rails. Iron rails were developed in the eighteenth
century as a result, and alternative dimensions and profiles were explored. Further-
more, the need for traction provided the first market for locomotives.There was no
market for steam-powered land vehicles because roads were unpaved and too
uneven to support a steam vehicle (as Cugnot and Trevithick discovered). Railways,
however, provided a controlled surface on which steam vehicles could function, and
colliery railways were the first purchasers of steam locomotives. When George
Stephenson developed the Rocket for the Rainhill trials, he tested his design ideas
by incorporating them in locomotives he was building for coal railways. In this way,
the commercial operation of primitive versions of technology promoted further
development as R&D expenses were absorbed as normal business costs.

Cotton played a supporting role in the growth of the engineering industry for two
reasons.The first is that it grew to an immense size.This was a consequence of global
competition. In the early eighteenth century, Britain produced only a tiny fraction
of the world’s cotton yarn and cloth. The main producers were in Asia. As a result,
the price elasticity of demand for English cotton cloth was extremely large. If Britain
could become competitive, it could expand production enormously by replacing
Indian and Chinese producers. Mechanization led to that outcome.73 The result was
a huge industry, widespread urbanization (with such external benefits as that
conveyed), and a boost to the high-wage economy. Mechanization in other activities
did not have the same potential. The Jacquard loom, a renowned French invention
of the period, cut production costs in lace and knitwear, and thereby induced some
increase in output. However, knitting was not a global industry, and the price
elasticity of demand was only modest, so output expansion was limited. One reason
that British cotton technology was so transformative was that cotton cloth was a
global industry with more price-responsive demand than other textiles.

The growth and size of the cotton industry in conjunction with its dependence
on machinery sustained the engineering industry by providing it with a large and
growing market for machinery. The history of the cotton industry was one of
relentlessly improving machine design—first with carding and spinning and later
with weaving. Improved machines translated into high investment and demand for
equipment. By the 1840s, the initial dependence of cotton manufacturers on water
power gave way to steam-powered mills.74 By the middle of the nineteenth century,
Britain had a lopsided industrial structure. Cotton was produced in highly mecha-
nized factories, while much of the rest of manufacturing was relatively untrans-
formed. In the mid-nineteenth century, machines spread across the whole of
British manufacturing (one of the causes of the continuing rise in income). Until
then, cotton was important as a major market for the engineering industry.

The reason that the British inventions of the eighteenth century—cheap iron
and the steam engine, in particular—were so transformative was because of the
possibilities they created for the further development of technology. Technologies
invented in France—in paper production, glass, and knitting—did not lead to
general mechanization or globalization. One of the social benefits of an invention

73 S. Broadberry and B. Gupta, ‘Wages, induced innovation and the great divergence: Lancashire, India and
shifting competitive advantage in cotton textiles, 1700–1850’, rev. version, CEPR discussion paper 5183 (2006).

74 von Tunzelmann, Steam power, pp. 175–225.

INDUSTRIAL REVOLUTION 381

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

is the door it opens to further improvements. British technology in the eighteenth
century had much greater possibilities in this regard than French inventions. The
British were not more rational or prescient than the French in developing coal-
based technologies: the British were simply luckier in their geology. The knock-on
effect was large, however. There is no reason to believe that French technology
would have led to the engineering industry, the general mechanization of industrial
processes, the railway, the steam ship, or the global economy. In other words, there
was only one route to the twentieth century—and it traversed northern Britain.

Oxford University

Date submitted 8 March 2007
Revised version submitted 10 May 2009
Accepted 17 July 2009

DOI: 10.1111/j.1468-0289.2010.00532.x

Footnote references
Acemoglu, D., ‘Factor prices and technical change: from induced innovations to recent debates’, in P. Aghion, R.

Frydman, J. Stiglitz, and M. Woodford, eds., Knowledge, information and expectations in modern macroeconomics:
in honor of Edmund Phelps (Princeton, 2003), pp. 464–91.

Acemoglu, D., Johnson, S., and Robinson, J., ‘The rise of Europe: Atlantic trade, institutional change and
economic growth’, American Economic Review, 95 (2005), pp. 546–79.

Allen, R. C., ‘Economic structure and agricultural productivity in Europe, 1300–1800’, European Review of
Economic History, 3 (2000), pp. 1–25.

Allen, R. C., The great divergence in European wages and prices from the middle ages to the First World War’,
Explorations in Economic History, 38 (2001), pp. 411–47.

Allen, R. C., ‘Poverty and progress in early modern Europe’, Economic History Review, LVI (2003), pp. 403–43.
Allen, R. C., ‘Was there a timber crisis in early modern Europe?’, in Economia e energia secc. xiii–xviii, serie ‘II–Atti

delle ‘Settimane di Studi’ e altri Convegni, 34, Instituto Internazionale di Storia Economica ‘F. Datini’, Prato
(2003), pp. 469–82.

Allen, R. C., ‘India in the great divergence’, in T. J. Hatton, K. H. O’Rourke, and A. M. Taylor, eds., The new
comparative economic history: essays in honor of Jeffery G.Williamson (Cambridge, Mass., 2007), pp. 9–32.

Allen, R. C., The British industrial revolution in global perspective (Cambridge, 2009).
Allen, R. C., ‘The industrial revolution in miniature: the spinning Jenny in Britain, France, and India’, Journal of

Economic History, 69 (2009), pp. 901–27.
Allen, R. C., ‘Engels’ pause: technical change, capital accumulation, and inequality in the British industrial

revolution’, Explorations in Economic History, 46 (2009), pp. 418–35.
Allen, R. C., Bengtsson, T., and Dribe, M., Living standards in the past: new perspectives on well-being in Asia and

Europe (Oxford, 2005).
Allen, R. C., Bassino, J.-P., Ma, D., Moll-Murata, C., and van Zanden, J. L., ‘Wages, prices, and living standards

in China, 1738–1925: in comparison with Europe, Japan, and India’, Economic History Review (forthcoming).
Angerstein, R. R., R. R. Angerstein’s illustrated travel diary, 1753–1755, T. Berg and P. Berg, trans. (2001).
Berg, M., Luxury and pleasure in eighteenth century Britain (Oxford, 2005).
Berg, M. and Clifford, H., Consumers and luxury: consumer culture in Europe, 1650–1850 (Manchester, 1999).
Bonney, R., The rise of the fiscal state in Europe, c. 1200–1815 (Oxford, 1999).
Briggs, R., Witches and neighbours (Oxford, 2nd edn. 2002).
Burke, P., Popular culture in early modern Europe (Aldershot, rev. repr. 2006).
Burton, A., Richard Trevithick: giant of steam (2000).
Chapman, S. D., ‘Fixed capital formation in the British cotton industry 1770–1815’, Economic History Review,

2nd ser., XXIII (1970), pp. 235–66.
Clark, G., The political foundations of modern economic growth: England, 1540–1800’, Journal of Interdiscipli-

nary History, 26 (1996), pp. 563–87.
Clark, G. and Jacks, D., ‘Coal and the industrial revolution’, European Review of Economic History, 11, 1 (2007),

pp. 39–72.
Cohen, H. F., ‘Inside Newcomen’s fire engine, or the scientific revolution and the rise of the modern world’,

History of Technology, 25 (2004), pp. 111–32.
Crafts, N. F. R., ‘Industrial revolution in England and France: some thoughts on the question: “Why was England

first?”’, Economic History Review, 2nd ser., XXX (1977), pp. 429–41.

382 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

Crafts, N. F. R., British economic growth during the industrial revolution (Oxford, 1985).
Crafts, N. F. R., ‘Steam as a general purpose technology: a growth accounting perspective’, Economic Journal, 114,

495 (2004), pp. 338–51.
Crafts, N. F. R. and Harley, C. K., ‘Output growth and the British industrial revolution: a restatement of the

Crafts-Harley view,’, Economic History Review, XLV (1992), pp. 703–30.
Crafts, N. F. R. and Harley, C. K., ‘Simulating the two views of the industrial revolution’, Journal of Economic

History, 60 (2000), pp. 819–41.
David, P., Technical choice, innovation, and economic growth: essays on American and British experience in the nineteenth

century (Cambridge, 1975).
David, P., ‘Common agency contracting and the emergence of “open science” institutions’, American Economic

Review, 88 (1998), pp. 15–21.
De Long, J. B. and Schleifer, A., ‘Princes and merchants: European city growth before the industrial revolution’,

Journal of Law and Economics, 36 (1993), pp. 671–702.
Desaguliers, J. T., A course of experimental philosophy, 2 vols. (1734–44).
Diderot, D. and d’Alembert, J. le R., Encyclopédie, ou dictionnaire raisonné des sciences, des arts et des métiers, 28 vols.

(Paris, 1751).
Doyon, A. and Liaigre, L., Jacques Vaucanson: mécanicien de génie (Paris, 1966).
Eden, Sir F. M., The state of the poor, 3 vols. (1797).
Epstein, S. R., Freedom and growth: the rise of states and markets in Europe, 1300–1750 (2000).
Fairchilds, C., ‘The production and marketing of populuxe goods in eighteenth-century Paris’, in J. Brewer and

R. Porter, eds., Consumption and the world of goods (1993), pp. 228–48.
Fremdling, R., ‘Transfer patterns of British technology to the Continent: the case of the iron industry’, European

Review of Economic History, 4 (2000), pp. 195–222.
Fremdling, R., ‘Continental responses to British innovations in the iron industry during the eighteenth and early

nineteenth centuries’, in L. P. de la Escosura, ed., Exceptionalism and industrialisation: Britain and its European
rivals, 1688–1815 (Cambridge, 2004), pp. 145–69.

Gerschenkron, A., Economic backwardness in historical perspective (Cambridge, Mass., 1962).
Gilboy, E. W., Wages in eighteenth century England (Cambridge, Mass., 1934).
Habakkuk, H. J., American and British technology in the nineteenth century (Cambridge, 1962).
Hall, A. R., ‘What did the industrial revolution in Britain owe to science?’, in N. McKendrick, ed., Historical

perspectives: studies in English thought and society (1974) pp. 129–51.
Hamilton, H., The English brass and copper industries to 1800 (1926).
Harley, C. K., ‘The shift from sailing ships to steam ships, 1850–1890: a study in technological change and its

diffusion’, in D. N. McCloskey, ed., Essays on a mature economy: Britain after 1840 (Princeton, 1971),
pp. 215–34.

Harley, C. K., ‘Reassessing the industrial revolution: a macro view’, in J. Mokyr, ed., The British industrial
revolution: an economic perspective (2nd edn. 1999), pp. 160–205.

Harris, J. R., ‘Saint-Gobain and Ravenshead’, in B. M. Ratcliffe, ed., Great Britain and her world, 1750–1914
(Manchester, 1975), pp. 27–70.

Hartwell, R. M., The causes of the industrial revolution (1967).
Hatcher, J., The history of the British coal industry,vol. I: Before 1700: towards the age of coal (Oxford, 1993).
Hicks, J., A theory of wages (1932).
Hills, R. L., Power in the industrial revolution (Manchester, 1970).
Hoffman, P. T. and Norberg, K., Fiscal crises, liberty, and representative government, 1450–1789 (Stanford,

1994).
Hoffman, P. T., Postel-Vinay, G., and Rosenthal, J. L., Priceless markets: the political economy of credit in Paris,

1660–1870 (Chicago, 2000).
Hollister-Short, G. J., ‘The introduction of the Newcomen engine into Europe’, Transactions of the Newcomen

Society, 48 (1976–7).
Humphries, J., Childhood and child labour in the British industrial revolution (Cambridge, 2008).
Jacob, M. C., Scientific culture and the making of the industrial west (New York, 1997).
Jacob, M. and Stewart, L., Practical matter: Newton’s science in the service of industry and empire: 1687–1851

(Cambridge, Mass., 2004).
Jevons, W. S., The coal question: an inquiry concerning the progress of the nation and the probable exhaustion of our coal

mines (1865).
Kanefsky, J. and Robey, J., ‘Steam engines in 18th-century Britain: a quantitative assessment’, Technology and

Culture, 21 (1980), pp. 161–86.
Landes, D. S., The unbound Prometheus: technological change and industrial development in western Europe from 1750

to the present (Cambridge, 1969).
Landes, D. S., Revolution in time: clocks and the making of the modern world (rev. edn. 2000).
LaPorta, R., Lopez-de-Silanes, F., Schleifer, A., and Vishny, R.W., ‘Law and finance’, Journal of Political Economy,

106 (1998), pp. 1113–55.
Lipsey, R. G., Carlaw, K. I., and Bekar, C. T., Economic transformations: general purpose technologies and long-term

economic growth (Oxford, 2005).

INDUSTRIAL REVOLUTION 383

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

McKendrick, N., Brewer, J., and Plumb, J. H., The birth of a consumer society: the commercialization of eighteenth-
century England (1982).

Machlup, F., The production and distribution of knowledge in the United States (Princeton, 1962).
Mathias, P., ‘Who unbound Prometheus? Science and technical change, 1600–1800’, in A. E. Musson, ed.,

Science, technology and economic growth in the eighteenth century (1972), pp. 69–96.
Mathias, P. and O’Brien, P. K., ‘Taxation in England and France, 1715–1810’, Journal of European Economic

History, 5 (1976), pp. 601–50.
Mathias, P. and O’Brien, P. K., ‘The incidence of taxes and the burden of proof’, Journal of European Economic

History, 7 (1978), pp. 211–13.
Mokyr, J., ‘Editor’s introduction: the new economic history and the industrial revolution’, in J. Mokyr, ed., The

British industrial revolution: an economic perspective (Boulder, 1993), pp. 1–131.
Mokyr, J., The gifts of Athena: historical origins of the knowledge economy (Princeton, 2002).
Nef, J. U., The rise of the British coal industry (1932).
North, D. C. and Weingast, B. R., ‘Constitutions and commitment: the evolution of institutions governing public

choice in seventeenth-century England’, Journal of Economic History, XLIX (1989), pp. 803–32.
Nuvolari, A., ‘Collective invention during the British industrial revolution: the case of the Cornish pumping

engine’, Cambridge Journal of Economics, 28 (2004), pp. 347–63.
Nuvolari, A., The making of steam power technology: a study of technical change during the industrial revolution

(Eindhoven, 2004).
Nuvolari, A., ‘The making of steam power technology: a study of technical change during the British industrial

revolution’, Journal of Economic History, 66 (2006), pp. 472–6.
O’Rourke, K. H. and Williamson, J. G., Globalization and history: the evolution of a nineteenth-century Atlantic

economy (Cambridge, Mass., 1999).
Özmucur, S. and Pamuk, S., ‘Real wages and standards of living in the Ottoman Empire, 1489–1914’, Journal of

Economic History, 62 (2002), pp. 293–321.
Pomeranz, K., The great divergence: China, Europe, and the making of the modern world (Princeton, 2000).
Pounds, N. J. G. and Parker, W. N., Coal and steel in western Europe (1957).
Peaucelle, J.-L., ‘La division du travail: Adam Smith et les encyclopédistes observant la fabrication des épingles

en Normandie’, Gérer et Comprendre, 57 (1999), pp. 36–51.
Peaucelle, J.-L., ‘Raisonner sur les épingles, l’exemple de Adam Smith sur la division du travail’, Revue d’Économie

Politique, 4 (2005), pp. 499–519.
Peaucelle, J.-L., Adam Smith, la division du travail, et la fabrication d’épingles (2006).
Quinn, S., ‘The glorious revolution’s effect on English private finance: a microhistory, 1680–1705’, Journal of

Economic History, 61 (2001), pp. 593–615.
Raistrick, A., Dynasty of iron founders: the Darbys and Coalbrookdale (Iron Bridge Gorge Trust, York, 1989).
Redlich, F., ‘The leaders of the German steam-engine industry during the first hundred years’, Journal of

Economic History, IV (1944), pp. 121–48.
Russell, P., England displayed. Being a new, complete, and accurate survey and description of the kingdom of England,

and principality of Wales . . . By a society of gentlemen, 2 vols. (1769).
Ruttan, V. W., Technology, growth, and development: an induced innovation perspective (Oxford, 2001).
Ruttan, V.W. and Thirtle, C., The role of demand and supply in the generation and diffusion of technical change (2001).
Shammas, C., The pre-industrial consumer in England and America (Oxford, 1990).
Sharpe, J. A., Early modern England: a social history, 1550–1760 (2nd edn. 2007).
Sieferle, R., The subterranean forest: energy systems and the industrial revolution (Cambridge, 2001).
Smil, V., Energy in world history (Boulder, 1994).
Smith, A., An inquiry into the nature and causes of the wealth of nations, E. Cannan, ed., (New York [1776], 1937).
Stewart, L., The rise of public science: rhetoric, technology, and natural philosophy in Newtonian Britain, 1660–1750

(Cambridge, 1992).
Tann, J., ‘Makers of improved Newcomen engines in the late 18th century’, Transactions of the Newcomen Society,

50 (1978–9).
Temin, P., ‘Notes on labor scarcity in America’, Journal of Interdisciplinary History, 1 (1971), pp. 251–64.
Temin, P., ‘Two views of the British industrial revolution’, Journal of Economic History, 57 (1997), pp. 63–82.
Thomas, K., Religion and the decline of magic (1971).
von Tunzelmann, G. N., Steam power and British industrialization to 1860 (Oxford, 1978).
Unger, R. W., ‘Energy sources for the Dutch golden age: peat, wind, and coal’, Research in Economic History, 9

(1984), pp. 221–53.
de Vries, J., ‘Between purchasing power and the world of goods: understanding the household economy in early

modern Europe’, in J. Brewer and R. Porter, eds., Consumption and the world of goods (1993), pp. 85–132.
de Vries, J. and van der Woude, A., The first modern economy: success, failure and perseverance of the Dutch economy,

1500–1815 (Cambridge, 1997).
Weatherill, L., Consumer behaviour & material culture in Britain, 1660–1760 (2nd edn. 1996).
Weiss, L., Watch-making in England, 1760–1820 (1982).
Wrigley, E. A., Continuity, chance and change (Cambridge, 1988).

384 R. C. ALLEN

© Economic History Society 2010 Economic History Review, 64, 2 (2011)

Copyright of Economic History Review is the property of Wiley-Blackwell and its content may not be copied or

emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission.

However, users may print, download, or email articles for individual use.