. 1
( 14)



>>

This page intentionally left blank
Gender, Work and Wages in Industrial
Revolution Britain




A major new study of the role of women in the labor market of
Industrial Revolution Britain. It is well known that men and women
usually worked in different occupations, and that women earned lower
wages than men. These differences are usually attributed to custom
but Joyce Burnette here demonstrates instead that gender differences
in occupations and wages were largely driven by market forces. Her
¬ndings reveal that, rather than harming women, competition actually
helped them by eroding the power that male workers needed to restrict
female employment and by minimizing the gender wage gap by sorting
women into the least strength-intensive occupations. Where the strength
requirements of an occupation made women less productive than men,
occupational segregation maximized both economic ef¬ciency and female
incomes. She shows that women™s wages were then market rather than
customary wages and that the gender wage gap resulted from actual
differences in productivity.

J O Y C E B U R N E T T E is Daniel F. Evans Associate Professor of Economics
at Wabash College, Indiana.
Cambridge Studies in Economic History

Editorial Board
Paul Johnson
London School of Economics and Political Science
Sheilagh Ogilvie
University of Cambridge
Avner Offer
All Souls College, Oxford
Gianni Toniolo

Universit a di Roma ˜Tor Vergata™
Gavin Wright
Stanford University

Cambridge Studies in Economic History comprises stimulating and
accessible economic history which actively builds bridges to other
disciplines. Books in the series will illuminate why the issues they
address are important and interesting, place their ¬ndings in a
comparative context, and relate their research to wider debates
and controversies. The series will combine innovative and exciting
new research by younger researchers with new approaches to
major issues by senior scholars. It will publish distinguished work
regardless of chronological period or geographical location.

Titles in the series include:
Robert Millward Private and Public Enterprise in Europe: Energy,
Telecommunications and Transport, 1830“1990
S. D. Smith Slavery, Family, and Gentry Capitalism in the British
Atlantic: The World of the Lascelles, 1648“1834
Stephen Broadberry Market Services and the Productivity Race, 1850“2000:
British Performance in International Perspective.
Gender, Work and Wages in
Industrial Revolution Britain

Joyce Burnette
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521880633

© Joyce Burnette 2008


This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2008


ISBN-13 978-0-511-39350-1 eBook (EBL)

ISBN-13 978-0-521-88063-3 hardback




Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
Contents




page vi
List of ¬gures
vii
List of tables
xi
Preface

Introduction 1
1 Women™s occupations 16
2 Women™s wages 72
3 Explaining occupational sorting 136
4 Testing for occupational barriers in agriculture 186
5 Barriers to women™s employment 221
6 Occupational barriers in self-employment 274
7 Women™s labor force participation 306
8 Conclusion 327
Appendix to Chapter 3 336
Appendix to Chapter 4 342

351
Bibliography
370
Index




v
Figures




1.1 The prevalence of women in commercial directories page 33
1.2 The age distribution of textile factory workers 43
2.1 The female“male wage ratio by age in textile factories 79
2.2 The female“male wage ratio by age in agriculture 80
2.3 Female“male strength ratios: adults 107
2.4 Female“male strength ratios: teens 107
2.5 Female winter wages 127
2.6 Female summer wages 127
2.7 Wage persistence 130
3.1 A general example of a strength“productivity
relationship 141
3.2 A speci¬c example: productivity as a function of
strength in occupations A and B 142
3.3 The ef¬ciency costs of moving workers 143
3.4 A decline in the price of good B 144
3.5 Entry of a new occupation 147
3.6 Determination of the wage ratio in Model B 149
3.7 A change in technology 150
3.8 The distribution of male and female workers across
cloths of various piece-rates 160
3.9 The effect of the lace-making industry on the
market wage ratio 167
5.1 Hicks™s bargaining model 251
5.2 Male wages at the Estcourt Farm in Shipton
Moyne, Gloucestershire 263
5.3 Daily wage of John Rickards at the Estcourt farm 263
7.1 Changes over time in the prevalence of women in
commercial directories 308
7.2 Female“male wage ratio in agriculture 311
7.3 Feinstein™s estimates of real earnings 319



vi
Tables




1.1 Occupations in the 1841 and 1851 censuses:
Great Britain page 19
1.2 The occupations of women workers: Higgs™s
revisions of census data 22
1.3 Employment ratios 26
1.4 Comparison of commercial directories and population 31
1.5 Number of independent tradeswomen, from
commercial directories 32
1.6 The top ten most common occupations for men
and women in commercial directories 35
1.7 Wages in lace-making 47
1.8 The British proprietress 66
2.1 Women™s wages compared to men™s 74
2.2 Payments for reaping at Gooseacre Farm, Radley,
Berkshire 88
2.3 Servants™ wages 96
2.4 Differences in physical performance by sex 109
2.5 Gender gaps in performance for recruits and soldiers 110
2.6 Examples of apprenticeship premiums 119
2.7 Wage persistence, female summer wages 129
3.1 Male and female strength distributions 141
3.2 Wages paid to laborers at the Apley Park farm,
July 15, 1836 151
3.3 Age-speci¬c marital fertility 173
3.4 Women™s wages in cottage industry compared to
wages in other industries, 1833 181
4.1 Descriptive statistics: Arthur Young™s data 191
4.2 Distribution of farm size 192
4.3 OLS and Tobit estimations: speci¬cation one 194
4.4 OLS and Tobit estimations: speci¬cation two 195

vii
viii List of tables

4.5 Elasticities 197
4.6 Two-stage least squares estimates and speci¬cation test 200
4.7 Means of wages: 1770 204
4.8 Correlations of men™s and women™s wages: 1770 204
4.9 Log-log regressions: 1770 205
4.10 Wages in 1833 207
4.11 Correlations of men™s and women™s wages: 1833 208
4.12 Log-log regressions: 1833 209
4.13 Industry regressions: 1833 210
4.14 Correlation of seasonal wage differences: 1833 212
4.15 Difference-of-log regressions: 1833 213
4.16 The effect of unemployment: 1833 213
4.17 Correlations with boys™ wages: 1833 214
4.18 Wages in England and Wales: 1860“1 215
4.19 Correlations of men™s and women™s wages: 1860“1 216
4.20 Log-log regressions: 1860“1 217
4.21 Correlations of wage differences: 1860“1 217
4.22 Difference-of-log regressions: 1860“1 218
4.23 Descriptive statistics: French agricultural day-laborers
in 1839 219
4.24 French agricultural day-laborers: 1839 219
5.1 The percentage of women in selected occupations:
the 1841 census 222
5.2 Occupational sorting in skilled occupations:
Manchester, 1846 223
5.3 The gender division of labor in staymaking 224
5.4 The apprenticeship of girls 237
6.1 Sorting in the garment trades 276
6.2 Capital requirements: Campbell™s estimates
compared to others 282
6.3 Percent female compared to capital requirements 287
6.4 Correlation of minimum capital requirements with
the percentage of business owners who were women 288
6.5 Professional employment in the censuses 293
7.1 Married women™s labor force participation from
census totals 307
7.2 Indexes of occupational segregation from
commercial directories 316
7.3 Trends in female participation in some of the
largest occupations 317
7.4 The predicted effect of changes in real earnings on
married women™s labor force participation 320
ix
List of tables

8.1 Gender division of labor by strength category
of occupation 333
8.2 Men™s hours of housework as a percentage of women™s
hours of housework 334
Preface




Once upon a time women were largely missing from economic history.
Economic historians somehow managed to make claims about the
standard of living without examining women™s wages. Happily, that has
now changed, thanks to the efforts of pioneering feminists who made the
case for the importance of including women in economic history. Since
the value of studying women as well as men is now well established, I do
not feel a need to justify the existence of this book. The subject matter is
contentious, but it is my hope that the book will stimulate, not an all-or-
nothing debate about the existence of gender discrimination, but a
nuanced discussion of where, when, and how gender discrimination may
have operated, and of the relationship between discrimination and
markets.
This book began ¬fteen years ago as a PhD dissertation at Northwestern
University. The origin of the project was a paper I wrote for Joel Mokyr™s
European Economic History class on the correlation between male and
female wages in the “Rural Queries” of 1833. This paper got me thinking
about how the labor market treated women, a process which eventually led
to the ideas expressed here. I am grateful for the input of Joel Mokyr, my
dissertation advisor, and Rebecca Blank and Bruce Meyer, the labor
economists on my committee. A grant from the Mellon Foundation
supported a year of dissertation research, and a Northwestern University
Dissertation Year Grant supported the purchase of micro¬lm from the
archives.
After receiving my PhD, I published parts of my research as articles, but
otherwise put the dissertation aside while I concentrated on collecting data
from farm accounts. I continued to think about the issues raised in this
book, but did not begin to revise it until my sabbatical in 2002“3. I spent
that academic year as a visitor at the London School of Economics,
supported partly by Wabash College and partly by a Sabbatical Fellowship
from the American Philosophical Society. Most of the revisions to the


xi
xii Preface

manuscript were accomplished in the spring of 2005, during a one-
semester leave funded by a grant from the National Science Foundation
(Grant no. 0213954). Any opinions, ¬ndings, and conclusions or recom-
mendations expressed in this book are those of the author and do not
necessarily re¬‚ect the views of the National Science Foundation. I thank
Dan Newlon for working with someone who didn™t understand the grant
process very well.
I am thankful for the many comments I have received from colleagues
when I have presented portions of the material. Colleagues who have
been especially helpful are Greg Clark, Jane Humphries, and Andrew
Seltzer, who have commented on my work multiple times over many
years. I am especially grateful for critics of my work who have forced me
to think more carefully about speci¬c claims. I thank James Henderson
for teaching me to love economics as an undergraduate at Valparaiso
University. Last but not least, I am thankful for the support of my
husband Patrick, both for helping me with my prose, and for running the
household when I was doing other things.
Introduction




Early in the morning of Friday, January 28, 1820, a night watchman at
the Broomward Cotton Mill in Glasgow discovered a ¬re in the carding
room. He:
gave the alarm, and, on going to the spot, found that some Person or Persons had,
by getting up on a tree opposite to, and within three feet of the east side of the
Mill, thrown in, through the opening pane of one of the windows, a Paper Bundle
or Package, ¬lled with Pitch and Gunpowder, and dipped in Oil, which had
exploded, and set Fire to a Basket full of loose Cotton, which communicated to
one of the Carding Engines, and which, unless it had instantly and providentially
been discovered and got under, must have consumed the whole Building.1

James Dunlop, the owner of the mill, was probably not surprised. The
motives of the arsonists were no mystery. On January 31 the Glasgow
Herald reported:
This ¬re, there is good ground to believe, has been occasioned by a gang of
miscreants who, for some time past, have waylaid, and repeatedly assaulted and
severely wounded, the persons employed at the Broomward Cotton Mill, who
are all women, with the view of putting the mill to a stand, and throwing the
workers out of employment.2

A few years later twenty-¬ve mill owners from Glasgow petitioned the
Home Secretary Robert Peel to extend the anti-union Combination
Laws to Scotland. Their petition describes this case in more detail.
Messrs James Dunlop and Sons, some years ago, erected cotton mills in Calton
of Glasgow, on which they expended upwards of 27,000l. forming their spinning
machines (chie¬‚y with the view of ridding themselves of the combination) of
such reduced size as could easily be wrought by women. They employed women
alone, as not being parties to the combination, and thus more easily managed,
and less insubordinate than male spinners. These they paid at the same rate of
wages, as were paid at other works to men. But they were waylaid and attacked,
in going to, and returning from their work; the houses in which they resided,

1 2
The Glasgow Herald, Monday, January 31, 1820, p. 3, col. 2. Ibid., p. 2, col. 4.

1
2 Gender, Work and Wages in Industrial Revolution Britain

were broken open in the night. The women themselves were cruelly beaten and
abused; and the mother of one of them killed; in ¬ne, the works were set on ¬re
in the night, by combustibles thrown into them from without; and the ¬‚ames
were with dif¬culty extinguished; only in consequence of the exertions of the
body of watchmen, employed by the proprietors, for their protection. And these
nefarious attempts were persevered in so systematically, and so long, that
Messrs. Dunlop and Sons, found it necessary to dismiss all female spinners from
their works, and to employ only male spinners, most probably the very men who
had attempted their ruin.3

The women spinners employed by Dunlop lost their jobs as a direct
result of the male workers™ opposition.
The attempt to burn Dunlop™s mill was just one battle in a war
between the cotton spinners™ union and their employers. Other mills
were attacked, and one employer was even shot at in the doorway of his
father-in-law™s house on his wedding night.4 The dispute included,
among other points, an objection to the employment of women. On
November 27, 1822, Patrick McNaught, manager of the Anderston
Cotton Mill in Glasgow, received the following note from the spinners™
union, which emphasized the employment of women:
Sir,
I am authorized to intimate jeoperdy and hazardious prediciment you stand in at
the present time, by the operative cotton spinners, and lower class of mankind,
in and about Glasgow, by keeping them weomen of¬ciating in mens places as
cotton spinners, and plenty of men going idle out of employ, which would I accept
of them for the same price omiting the list which you know is trif¬‚ing. So they
present this proposal as the last, in corresponding terms, so from this date they
give you a fortnight to consider the alternative, whether to accept the ¬rst or the
latter, which will be assassination of body; which you may relie upon no other
thing after the speci¬ed time is run, for you will be watched and dogged by night
and by day, till their ends are accomplished; for you well deserve the torturings
death that man could invent, being so obstinate, more so than any other master
round the town, and seeing poor men going about the street, with familys starving,
and keeping a set of whores, as I may call them, spending their money, drinking
with young fellows, and keeping them up. So mark this warning well, and do not
vaunt over it like you foolish neighbour, Mr. Simpson, in Calton, with his, for he
was soon brought to the test, and you will be the same with murder.5

The writer of this note, identi¬ed only as “Bloodthirst void of fear,” draws
on gender ideology to create a sense of outrage. He calls the women
whores for the offenses of “spending their money” and “drinking with
young fellows,” activities which do not seem to us worthy of condemnation

3
Fifth Report from the Select Committee on Artizans and Machinery, BPP 1824 (51) V, p. 525.
4
Ibid., p. 527. 5 Ibid., p. 531.
Introduction 3

but clearly fall outside what the writer considers to be proper feminine
behavior. One suspects, though, that the real reason for the opposition
to female employment is that the women are working “in men™s places.”
If women were employed, men would be unemployed, or at least would
have to work for lower wages. Employers were somehow immune to
these concerns about proper feminine behavior, and actively sought to
hire women because they could bene¬t economically from doing so. It
was the male workers, who would lose economically from their employ-
ment, who expressed such concerns about proper female behavior. Thus a
man™s opinions on whether women should work in the factory seem to
have been determined by whether he would win or lose economically
from the employment of women. The union™s grievances were not
directed only at women spinners, but also at other forms of competition;
the employment of male workers not approved by the union was also
violently opposed. The violence was economic warfare, aimed at pro-
tecting the spinners™ wages and working conditions. The actions of the
Glasgow mule spinners are just one example of barriers to women™s
employment that were erected because of economic motivations; men
excluded women to reduce competition and raise their own wages.
In the late eighteenth and early nineteenth centuries women and men
generally did not work at the same jobs, and they did not receive the
same wages. These differences are widely known, and the most common
explanation is that they resulted from discrimination or gender ideology.
This book will argue that economic motivations explain the patterns we
observe. In some cases, the occupational sorting was required for eco-
nomic ef¬ciency. Since strength was a scarce resource, the market paid a
premium for it. In other cases occupational sorting was the result of a
powerful group seeking to limit women™s opportunities in order to
improve its own economic position, at the expense of women, and at the
expense of economic ef¬ciency. The case of the Glasgow cotton spinners
illustrates the second case. Women were excluded from the highly paid
occupation of cotton spinning, not because they were incapable of doing
the job, or because employers refused to hire them, or because social
disapproval, combined with violence, kept them at home, but because
the male cotton spinners™ union was effective in excluding them, thus
reducing the supply and increasing the equilibrium wage of cotton
spinners.
In seeking to understand the causes of gender differences in wages and
occupations, this book will focus on actuality rather than ideology. I am
mainly interested in what work women actually did, rather than how
people thought or spoke about this work. Both ideology and actuality are
important topics of study, and one may in¬‚uence the other, but we must
4 Gender, Work and Wages in Industrial Revolution Britain

not confuse the two. Many researchers are primarily interested in the
ideology of the period. For example, Davidoff and Hall note, “The
suitability of ¬eld work, indeed any outdoor work for women, was
almost always discussed in moral terms.”6 This statement provides some
insight into how people in the Industrial Revolution discussed women™s
work. By contrast, I am primarily interested in what people did. Which
jobs did women do, and what were they paid?
We can ask two related but different questions about women™s work:
“What did people think women should do?” and “What work did women
actually do?” What people say does not always match what they actually
do, so evidence on the ¬rst question will not answer the second question.
While social expectations in¬‚uence behavior, they are not the whole story.
People have an amazing ability to say one thing and do another, par-
ticularly when they can bene¬t from doing so. Nineteenth-century
employers could hire married women at the same time they claimed to be
opposed to the employment of married women. For example, in 1876
Frederick Carver, the owner of a lace warehouse, told a parliamentary
committee: “we have as a rule an objection to employing married women,
because we think that every man ought to maintain his wife without the
necessity of her going to work.” However, he seems to have been willing
to break this rule without too much dif¬culty. Carver admitted that “As to
married women, in one particular department of our establishment we
have forty-nine married women and we wish that the present state of
things as regards married women should not be disturbed.”7 Because
preconceived notions of women™s work and actual employment often
con¬‚icted, we must make a clear distinction between the two when trying
to analyze women™s employment opportunities.
Amanda Vickery has warned us against taking Victorian ideology at
face value. She asks:
Did the sermonizers have any personal experience of marriage? Did men and
women actually conform to prescribed models of authority? Did prescriptive
literature contain more than one ideological message? Did women deploy the
rhetoric of submission selectively, with irony, or quite critically? . . . Just because a
volume of domestic advice sat on a woman™s desk, it does not follow that
she took its strictures to heart or whatever her intentions managed to live her life
according to its precepts.8

6
Leonore Davidoff and Catherine Hall, Family Fortunes: Men and Women of the English
Middle Class, 1780“1850 (Chicago: University of Chicago Press, 1987), p. 274.
7
BPP 1876, XIX, p. 258, quoted in Sonya Rose, Limited Livelihoods: Gender and Class in
Nineteenth-Century England (Berkeley: University of California Press, 1992), p. 32.
8
Amanda Vickery, “Golden Age to Separate Spheres? A Review of the Categories and
Chronology of English Women™s History,” Historical Journal 36 (1993), pp. 385, 391.
Introduction 5

This study will heed Vickery™s warning, and will not assume that
statements of gender ideology are evidence of how employers actually
made economic decisions. The fact that some jobs were labeled “men™s
work” is not proof that women were excluded because the gender label
attached to a job and the sex of the person who ¬lled the job did not
necessarily match. An 1833 parliamentary investigation ¬nds that “In
the Northern Counties, the Women engage in Men™s work much more
than in the Southern Districts.”9 While there was a clear category of jobs
designated “men™s work,” it was not true that men always ¬lled those jobs.
Of course, customary expectations often did accurately describe the
gender division of labor. Michael Roberts has suggested that the debate
between custom and market is not productive because the two are
compatible.10 It is true that market ef¬ciency and custom usually pre-
scribed the same outcomes, and I believe that this was no accident, but
the result of the close relationship between the two. In theory the rela-
tionship between custom and market could run in either direction.
Custom could determine the work that people did, or the work that
people did could determine which customs would emerge, or both. Most
historians believe that custom shaped economic outcomes. Some believe
that economic outcomes shaped custom. Heidi Hartmann, for example,
claims that women™s low social status has its roots in the gender division
of labor and can only be ended by ending occupational segregation.11
I believe that economic outcomes matched custom so closely because
custom was created to explain and justify the existing patterns of work
and pay. In some cases the gender division of labor resulted from eco-
nomic forces that promoted the most ef¬cient outcome. However, since
most people did not understand those economic forces, they relied on
gender ideology to explain the patterns they observed. In other cases the
gender division of labor was not ef¬cient but bene¬ted a particular
group; in these cases the group bene¬ting from occupational segregation
created and used gender ideology to promote their own economic
interests.
By emphasizing the economic motivations for gender differences, I am
providing a materialist explanation for the gender division of labor. This
is meant to be an alternative to the prevailing ideological explanation,
which gives priority to ideas about gender roles. I do believe that such

9
BPP 1834 (44) XXX, Whitburn, Durham, p. 169.
10
Michael Roberts, “Sickles and Scythes Revisited: Harvest Work, Wages and Symbolic
Meanings,” in P. Lane, N. Raven, and K. D. M. Snell, eds., Women, Work and Wages in
England, 1600“1850 (Woodbridge: Boydell Press, 2004), p. 89.
11
Heidi Hartmann, “Capitalism, Patriarchy, and Job Segregation by Sex,” Signs 1 (1976),
pp. 137“69.
6 Gender, Work and Wages in Industrial Revolution Britain

ideologies were present, but I don™t think they were the driving cause of
the differences we observe. Distributional coalitions could take advan-
tage of such ideologies, and even expand them, in order to justify their
inef¬cient policies. The Glasgow cotton spinners called the women
spinners whores, not because they were driven by a concern for sexual
purity, but because, by generating outrage, they could increase public
support for their campaign to remove the women from their jobs. The
question is not whether gender ideology existed, but whether it was the
engine driving the train or just the caboose. Most research on the subject
makes ideology the engine; I think it was the caboose.12
Even if patterns of work and pay were determined by economic forces,
that does not mean that people understood them that way. Customary
explanations are created partly because people do not understand eco-
nomic forces. During the Industrial Revolution sudden changes in
technology caused custom and the market to diverge, creating discom-
fort for the people involved when new realities did not match the cus-
tomary explanations that had been created for a different reality. We can
see an example of this discomfort in a passage by Friedrich Engels
describing the husband of a factory worker:
[a] working-man, being on tramp, came to St. Helens, in Lancashire, and there
looked up an old friend. He found him in a miserable, damp cellar, scarcely
furnished; and when my poor friend went in, there sat Jack near the ¬re, and
what did he, think you? why he sat and mended his wife™s stockings with the
bodkin; as soon as he saw his old friend at the door-post, he tried to hide them.
But Joe, that is my friend™s name, had seen it, and said: “Jack, what the devil art
thou doing? Where is the missus? Why, is that thy work?” and poor Jack was
ashamed and said: “No, I know that this is not my work, but my poor missus is
i™ th™ factory; she has to leave at half-past ¬ve and works till eight at night, and
then she is so knocked up that she cannot do aught when she gets home, so I have
to do everything for her what I can, for I have no work, nor had any for more nor
three years . . . There is work enough for women folks and childer hereabouts,
but none for men; thou mayest sooner ¬nd a hundred pound in the road than
work for men . . . when I got married I had work plenty . . . and Mary need not
go out to work. I could work for the two of us; but now the world is upside down.
Mary has to work and I have to stop at home, mind the childer, sweep and wash,
bake and mend.” . . . And then Jack began to cry again, and he wished he had
never married.13

Both gender ideology and market forces were very real for Jack. Gender
ideology told him that he should earn the income while his wife worked

12
For an alternative view, see Rose, Limited Livelihoods, pp. 12“13.
13
Frederick Engels, Condition of the Working Class in England in 1844 (London: George
Allen and Unwin, [1845] 1926), pp. 145“6.
Introduction 7

in the home, and the fact that this ideology did not match his situation
made him miserable. Market forces, however, determined the actual pat-
tern of work; his wife worked at the factory while Jack worked in the home.
Many studies of women™s work have chosen to focus on ideology, on
how people thought and talked about women workers.14 This focus may
arise from an interest in ideology for its own sake, or from a belief that
ideology drives action, that what people actually do is determined by the
categories of how they think. My focus on actuality comes from a belief
that the chain of causation more often runs the other way, that actuality
drives ideology. Economic actors respond to economic incentives, and
use ideology as a cover for their naked self-interest.
The relative strength of ideological and economic motivations is best
seen when the two con¬‚ict. Humphries has suggested that occupational
segregation was supported because concerns about sexuality required
keeping the sexes apart.15 In spite of this concern, however, men were
admitted to the intimate setting of childbirth. Though midwifery had
historically been a female activity, men began to enter the profession as
man-midwives in the seventeenth century. By the nineteenth century
male physicians were favored as birth attendants in spite of the Victorians™
prudishness that considered it “indelicate” for a father to be present
at the birth of his own child.16 Men who otherwise would consider it
dangerous to allow men and women to work together hired men to
attend at the births of their children. The medical profession de¬‚ected
any concerns about indelicacy by stressing male skill and supposed
female incompetence. Where male jobs were at stake, impropriety did
not seem to be a problem.
The existence of gender ideology sometimes makes it more dif¬cult to
discover the actuality of what work women did. Unfortunately, the
ideologies that were present affected the accuracy of the historical
records. Because a woman™s social status was determined by her rela-
tionship to men, the census does not accurately describe the work
women did. Many working women were not listed as having any
occupation. The 1841 census instructed enumerators to ignore the
occupations of a large fraction of women; its instructions state, “The
professions &c. of wives, or of sons or daughters living with and assisting

14
For example, see Deborah Simonton, A History of European Women™s Work (London:
Routledge, 1988) and Pamela Sharpe, “Commentary,” in P. Sharpe, ed., Women™s
Work: The English Experience 1650“1914 (London: Arnold, 1998), pp. 71“2.
Jane Humphries, “ ˜ . . . The Most Free from Objection . . . ™ The Sexual Division of
15

Labor and Women™s Work in Nineteenth-Century England,” Journal of Economic
History 47 (1987), pp. 929“50.
16
Jean Donnison, Midwives and Medical Men (London: Historical Publications, 1988), p. 64.
8 Gender, Work and Wages in Industrial Revolution Britain

their parents but not apprenticed or receiving wages, need not be
inserted.”17 In practice, census enumerators seem to have ignored
women™s employment even when they were receiving wages; Miller and
Verdon have both found examples of women who were paid wages for
agricultural labor but had no occupation listed in the census.18 Whether
an occupation was categorized as “skilled” was also socially determined.
Bridget Hill found that census of¬cials were unwilling to categorize
occupations employing women and children as skilled.
Albe Edwards, the man responsible for the reclassi¬cation, met with a problem
when he found certain occupations which technically were classi¬ed as “skilled”
had to be down-graded to “semi-skilled,” “because the enumerators returned so
many children, young persons, and women as pursuing these occupations.”
Edwards did not hesitate to lower the status of certain occupations when he
found women and young people worked in them in large numbers.19

In this case the categorization of occupations as skilled or semi-skilled
re¬‚ects ideology rather than characteristics of the job.
The ability of ideology to alter the historical record is not limited to
the nineteenth century. Sanderson ¬nds that in Edinburgh women were
actively involved in many skilled occupations, and that historians have
devalued their contributions by assuming that women™s occupations
were “merely extensions of domestic skills” or by failing to recognize
that women™s occupations were skilled occupations. The most telling
example of such devaluation of women™s work is from:
the entry in the printed Marriage Register for eighteenth-century Edinburgh where
the advocate John Polson is recorded as married to “Ann Strachan, merchant
(sic)”. The fact is that Ann Strachan was a merchant, but the modern editor,
because he assumed that an advocate was unlikely to have a working wife, recorded
this as an error. In a Commissary Court process it was stated during evidence on
behalf of the defender, that Polson had married Ann Strachan, the defender™s sister-
in-law, “who at that time had a great business and served the highest in the land.”20

We must avoid making the same mistake as the editor of the marriage
register, who took the gender ideology so seriously that he assumed Ann


17
Quoted in Edward Higgs, Making Sense of the Census (London: HMSO, 1989), p. 81.
18
C. Miller, “The Hidden Workforce: Female Fieldworkers in Gloucestershire, 1870“1901,”
Southern History 6 (1984), 139“61, and Nicola Verdon, Rural Women Workers in
Nineteenth-Century England: Gender, Work and Wages (Woodbridge: Boydell Press,
2002), pp. 117“19.
19
Bridget Hill, “Women, Work and the Census: A Problem for Historians of Women,”
History Workshop Journal 35 (1993), p. 90.
20
Elizabeth Sanderson, Women and Work in Eighteenth-Century Edinburgh (New York:
St. Martin™s Press, 1996), p. 105.
Introduction 9

Strachan™s occupational title must be a mistake. If Ann Strachan the
merchant disappears from history, we have lost any hope of discovering
the true place of women in the economy. Because what people said
about work is liable to be ¬ltered through the lens of ideology, I will try
wherever possible to use other types of evidence, such as statistical
evidence, to determine what people actually did.
Part of this book will be devoted to documenting the gender differ-
ences in wages and occupations. However, the main question I wish to
address is not whether differences occurred, but why they occurred.
What caused the gender differences in wages and occupations that we
observe? The question is not new, and many answers have been offered.
The most common explanation for gender differences in the labor
market is ideology: social institutions enforced socially determined
gender roles, and women were con¬ned to low-paid and low-status
work. These social constraints could operate even if people were not
aware of them.21 Differences between the genders were socially con-
structed. Both the gender division of labor and women™s lower wages
were determined by gender ideology. For example, Deborah Simonton
claims that “customary practices and ideas about gender and appro-
priate roles were instrumental in delineating tasks as male work and
female work.”22 Sonya Rose focuses on the expectation that women
were not supporting a family, and therefore did not need to be paid as
much as a man; she claims that “Women were workers who could be
paid low wages because of an ideology which portrayed them as sup-
plementary wage earners dependent on men for subsistence.”23
The ideological explanation of gender differences has some strengths.
People did express ideas about femininity and masculinity that implied
women should do certain jobs, and men others. We can observe these
ideas being expressed. And we have seen abrupt changes in the gender
division of labor that suggest arti¬cial barriers existed in the past. If the
percentage of law degrees earned by women increased from 5 percent in
1970 to 30 percent just ten years later, this suggests that women were
eager to become lawyers, and some barrier besides interest or inclination
kept the number of female lawyers low in 1970.24 Surely gender ideology

21
Sonya Rose notes that “Social actors often are unaware that these assumptions are
guiding their activities.” Limited Livelihoods, p. 13.
22
Simonton, European Women™s Work, p. 35
23
Sonya Rose, “ ˜Gender at Work™: Sex, Class and Industrial Capitalism,” History
Workshop Journal, 21 (1986), p. 117.
24
The percentage of law degrees earned by women continued to rise, reaching 42 percent
in 1990 and 47 percent in 2001. US Census Bureau, Statistical Abstract of the United
States: 2003 (Washington, DC: Government Printing Of¬ce, 2003), p. 194.
10 Gender, Work and Wages in Industrial Revolution Britain

played some part in the Church of England™s prohibition on the
ordination of women, which lasted until 1994. However, while I do
think that gender ideology is part of the story, in this book it will be cast
as a supporting character rather than as the protagonist.
At the other extreme, Kingsley Browne has embraced biological dif-
ference as an explanation for all differences in labor market outcomes
between men and women.25 Evolution, through sexual selection, cre-
ated differences between men and women. Women, who can have only a
few offspring, developed characteristics that led them to nurture these
offspring, maximizing the chances of survival. Men, who can father a
nearly unlimited number of children, developed strategies for winning
competitions that would allow them to have access to more females.
Scienti¬c studies have shown that the sex hormones cause differences in
aggressiveness, risk-taking, and nurturing behaviors. Kingsley Browne
has argued that these differences between the sexes explain why men are
more successful in the labor market than women. Men take more risks,
are more aggressive, and choose to spend less time with their families. He
argues that these are biological traits, against which it is futile to ¬ght, and
that they cause the observed differences in wages and occupations.
Even if Browne is right that evolution gives men a more competitive
character, his explanation provides at best part of the story. His main
focus is the “glass ceiling,” the gap in success at the highest levels. He
claims that men are more competitive and take more risks, and therefore
are more likely to reach the top. However, this explanation doesn™t tell
us why there is so much occupational segregation farther down the
occupational ladder. Also, Browne™s explanation cannot account for
sudden changes in the occupational structure. If there was something in
the female character, created by evolutionary sexual selection, that made
women reluctant to be lawyers, the number of women entering law
would not have changed so radically in the space of a couple of decades.
Happily, we have recently seen a few authors who neither assume men
and women must be biologically identical because they wish it to be so,
nor suggest that biological differences make any attempts to change the
status quo futile. Steven Pinker notes the emergence of a new left that
acknowledges both human nature and the possibility of improving our
social institutions.26 In his chapter on gender differences, Pinker acknow-
ledges biological differences that might lead men and women to choose

25
Kingsley Browne, Divided Labours: An Evolutionary View of Women at Work (New Haven:
Yale University Press, 1999).
26
Steven Pinker, The Blank Slate: The Modern Denial of Human Nature (New York:
Penguin Books, 2003), pp. 299“300.
Introduction 11

different occupations, but also acknowledges the existence of gender
discrimination.27 Acknowledging differences does not imply that one sex
is better than the other or must dominate over the other. Leonard Sax
notes that
The bottom line is that the brain is just organized differently in females and
males. The tired argument about which sex is more intelligent or which sex has
the “better” brain is about as meaningful as arguing about which utensil is
“better,” a knife or a spoon. The only correct answer to such a question is:
“Better for what?”28

Sax suggests that the outcomes are more likely to be equal if we admit
gender differences than if we don™t.
[Y]ou can teach the same math course in different ways. You can make math
appealing to girls by teaching it one way, or you can make it appealing to boys by
teaching it in another way. Girls and boys can both learn math equally well if you
understand those gender differences.29

However, ignoring gender differences and teaching math only one way is
likely to disadvantage one gender. Differences between the sexes are
important and must be acknowledged if we are to understand our world
and work to improve it.
There are also economic historians who allow biology to have a role in
shaping economic activity, without admitting it the power to determine
every observed difference. Some historians allow strength to have a role
in determining the sexual division of labor. Judy Gielgud notes that
“there are understandable reasons for a wage differential. For example, a
man™s strength might enable him to accomplish more of a given task
than could a woman in the same time, where both were working at full
stretch.”30 Merry Wiesner claims that the gender division of labor in
agriculture in the early modern period was partly, though not com-
pletely, due to differences in physical strength, “with men generally
doing tasks that required a great deal of upper-body strength, such as
cutting grain with a scythe.”31 Mary Friefeld™s story about the male
domination of mule-spinning points to the male union as the factor
excluding women after 1834, but acknowledges strength as the excluding

27
Ibid., pp. 354“7.
28
Leonard Sax, Why Gender Matters: What Parents and Teachers Need To Know about the
Emerging Science of Sex Differences (New York: Broadway Books, 2005), p. 32.
29
Ibid., p. 33
30
Judy Gielgud, “Nineteenth Century Farm Women in Northumberland and Cumbria:
The Neglected Workforce,” unpublished PhD thesis, University of Sussex, 1992, p. 85.
31
Merry Wiesner, Women and Gender in Early Modern Europe, 2nd edn (Cambridge:
Cambridge University Press, 2000), p. 106.
12 Gender, Work and Wages in Industrial Revolution Britain

factor in the early period. Pamela Sharpe admits a role for strength in the
occupation of wool-combing.32 Other historians have noted the effect of
women™s role in child-bearing on their work opportunities. Brenner and
Ramas, for example, note that “[b]iological facts of reproduction “
pregnancy, childbirth, lactation “ are not readily compatible with cap-
italist production,” so that as factories replaced home production
women were marginalized.33 These explanations all allow biology an
important role, without making the current division of labor the only one
biologically possible.
This book is also located between the extremes; it neither refuses to
acknowledge biological differences, nor sees observed gender differences
as completely determined by biology. I believe the importance of bio-
logical differences must be acknowledged if we are to have any hope of
understanding the gender division of labor, but I do not attempt to
ascribe all differences to biology. There is exclusion in this story, but it™s
not the whole story. We don™t have to deny the importance of biological
differences, or minimize their importance in the labor market, but nei-
ther do we have to accept all observed differences as the inevitable result
of our evolutionary heritage.
Men and women are different in ways that affect their productivity, so
we must not assume that differences in wages and occupations are
necessarily due to discrimination. If we accept even the least contro-
versial differences between men and women, much of the difference in
wages is explained. The biological differences that I focus on are the
least controversial. Kingsley Browne has argued that gender differences
in personality, created by the evolutionary process of sexual selection,
explain the differential success of men and women, but it may be dif¬-
cult to say whether traits such as competitiveness are determined by
biology or by culture. My argument does not rely on differences in
cognition or personality, and requires only two differences between the
sexes, neither of which is controversial. First, men are stronger than
women, and second, women give birth and breast-feed their infants,
while men do not. These two differences are suf¬cient to explain much
of the occupational segregation and gender wage gap that we observe in
Industrial Revolution Britain. While I do suggest that in many cases the
gender gap in wages was the result of biological differences between men
and women, that does not mean that I oppose attempts to reduce the

32
Pamela Sharpe, Adapting to Capitalism: Working Women in the English Economy, 1700“
1850 (London: Macmillan, 1996), p. 24. She notes that both strength and guild
restrictions kept this occupation male.
33
Johanna Brenner and Maria Ramas, “Rethinking Women™s Oppression,” New Left
Review 144 (1984), pp. 33“71.
Introduction 13

gender gap. Referring to the assumption that biological explanations of
the gender gap must support the status quo, Steven Pinker points out
that, “This makes about as much sense as saying that a scientist who
studies why women live longer than men ˜wants old men to die™.”34
While I take biology seriously, I don™t think it can be the whole story.
I differ from Kingsley Browne in not accepting that all differences in
labor market outcomes are simply the result of biology, and therefore
good. I am skeptical of claims that women will never choose career over
family, especially when I see so many women doing so today. Kingsley
Browne claims, rather broadly, that
Women care less about climbing hierarchies and about objective forms of
recognition such as money, status, and power than men. They place more
importance on a high level of involvement with their children. These conclusions
are consistent with evolutionary theory, biological fact, and psychological data.
It is simply the case that women tend to ¬t work to families, while men ¬t
families to work.35

However, this statement clearly does not describe all women. I read
the following in the Guardian: “I always expected to regret not having
children . . . So it comes as something of a surprise to discover that now,
in my 40s, I do not regret that I never gave birth . . . Instead, I feel more
liberated than I could ever have imagined.”36 It could be that the col-
umnist, Laura Marcus, is an unusual case, but it could also be that
Browne has overestimated the role of evolutionary biology in deter-
mining women™s choices.
The main conclusion of this book is that economic motivations caused
the gender differences we observe in the labor market of Industrial
Revolution Britain. In some cases these economic forces were bene¬cial,
and in other cases they were harmful, but in either case both women and
the economy in general would have bene¬ted from more competitive
markets. In the relatively competitive sectors of the labor market,
strength was an important input in production, and men™s higher wages
represent the premium paid for strength. In order to economize on the
scarce resource of strength, men were sorted into occupations requiring
more strength, and women into occupations requiring less strength.
Economic motivations led employers to hire men for jobs requiring
strength, and hire women for jobs requiring less strength. When tech-
nology changed, the gender division of labor changed too, always allo-
cating men to the more strength-intensive jobs. Employers were not

34
Pinker, The Blank Slate, p. 353. 35 Browne, Divided Labours, p. 53
36
Laura Marcus, “The Joys of Childlessness,” The Guardian August 22, 2002, p. 18.
14 Gender, Work and Wages in Industrial Revolution Britain

constrained by gender roles, but switched between men and women
workers when prices signaled that they should. While these forces did
result in gender differences in wages and occupations, they were bene-
¬cial in the sense of improving the ef¬ciency of the economy, and in the
sense that they minimized the gender wage gap. Women™s role in child-
bearing reduced the time women had available for market work, and
probably encouraged them to remain in the low-wage cottage industry
sector, but overall child-bearing was probably not as important as
strength in determining women™s productivity.
Unfortunately, economic motivations were not always bene¬cial. The
desire for gain sometimes leads groups with economic power to alter the
market to favor themselves at the expense of others. Mancur Olson
called such groups distributional coalitions.37 While such groups take
many forms, common forms are unions and professional organizations.
These organizations often attempt to limit the supply of their services
and thus raise their own wages. One way that occupational groups tried
to limit labor supply was by excluding women from the occupation.
While those in the occupation would bene¬t from high wages, society as
a whole would suffer a loss of ef¬ciency, and women would be harmed
by having their occupational choices restricted. Heidi Hartmann has also
argued that women were excluded from certain occupations because
men wanted to protect their own economic interest.38 Hartmann adds
that men wanted not only to maintain their own high wages, but also to
protect their own power within the family by ensuring that women
remained dependent. I agree with Hartmann, and will argue that most of
the real discriminatory constraints that women faced were restrictions
put in place by men who were trying to protect their own economic
position. Of course, not every group of men was able to enforce
restrictions against women. Only those occupations with some source of
market power, such as possession of a specialized skill, were successful in
excluding women.
I offer different explanations for different parts of the labor market,
but the explanations have a common strain: the importance of economic
self-interest. I do not believe that self-interest is always good. In fact, one
half of my story illustrates how self-interest could be harmful to both
women and the economy. Self-interest is bene¬cial if disciplined by
competition, but most economic actors would prefer to take the easier

37
Mancur Olson, The Rise and Decline of Nations: Economic Growth, Stag¬‚ation, and Social
Rigidities (New Haven: Yale University Press: 1982).
38
Hartmann, “Capitalism, Patriarchy, and Job Segregation.” See also Cynthia Cockburn,
Brothers: Male Dominance and Technological Change, 2nd edn (London: Pluto Press,
1991), pp. 34“5.
Introduction 15

route of monopoly and, if allowed, will use their power to bene¬t
themselves at the expense of others. Competition was the most powerful
force protecting women™s opportunities, and barriers to women™s
employment appeared where competition was weakest. In competitive
labor markets, market forces led to occupational sorting, but this sorting
bene¬ted women because it minimized the economic costs of their lesser
strength. The main source of barriers to women™s employment was
groups of men, or “distributional coalitions” to use Mancur Olson™s
term, who wished to monopolize an occupation to raise their own wages.
Where competition was strong these rules were ineffective; only where
competition was limited would unions and professional organizations
effectively bar women from employment. If there had been more com-
petition, women would have been able to work in a wider variety of
occupations, and would have had opportunities to earn higher wages.
In Industrial Revolution Britain men and women tended to work in
different occupations, and received different wages. This book explores
the reasons for those differences. I conclude that gender ideology played
a supporting role, but was not the driving force behind most of the
occupational segregation or wage gaps. Gender ideology had the most
in¬‚uence in institutions that did not have to compete to survive, such as
the family and the government. Comparative advantage and product-
ivity differences determined the division of labor and wages in the most
competitive sectors of the labor market. In other sectors, where one
group was able to amass enough economic power to sti¬‚e competition,
men erected barriers to the employment of women in order to reduce the
competition for their jobs. These men used gender ideology to increase
public support for the entry barriers they erected, but their primary
motivations were economic.
1 Women™s occupations




Before we can discuss the causes of occupational segregation, we must
¬rst have an accurate understanding of what work women did. While
this may seem to be a simple task, it presents some challenges to the
historian. Measures of occupational distribution are less than perfect,
and occupational patterns were changing rapidly during the Industrial
Revolution. Census data on individuals begins only in 1841, and when it
does exist it is not an accurate measure of women™s employment. This
leaves us without any aggregate measures of employment, so a glance at
the statistical abstract will not suf¬ce; instead, we must build a picture of
women™s employment from numerous incomplete sources. This chapter
will examine the evidence and determine what work women did during
the Industrial Revolution. Section I will discuss the limited statistical
evidence available on the pattern of occupational sorting by gender, and
Section II will examine the anecdotal evidence on women™s occupations.
Though the evidence is neither comprehensive nor perfectly reliable, it is
clear that men and women tended to work in different occupations.
However, it is also clear that the sorting was not perfect, and that women
were frequently found in occupations not generally considered to be
“women™s work.”
When examining women™s employment, we must keep in mind that
many of women™s productive contributions remain invisible to the his-
torian. Women at all levels of the labor market assisted their husbands
but received no of¬cial recognition for their productive contributions.
Frequently a marriage was also a business partnership, sometimes
explicitly. An advertisement in the Dorset County Chronicle speci¬ed,
“Wanted, A Man and his Wife, to manage a Dairy of Sixteen Cows.”1 In
the parish workhouses, which separated all inmates by sex, the master
took charge over the male inmates and the matron over the female
inmates. The workhouse of Melton, Suffolk, paid a salary of £50 a year

1
Dorset County Chronicle, December, 1860, quoted in Pamela Horn, “The Dorset Dairy
System,” Agricultural History Review 26 (1978), p. 100.

16
Women™s occupations 17

to the “governor and his wife.”2 In this case, a married couple shared
these responsibilities and received a joint salary. We do not know how
often the salary was simply given to the husband, with the understanding
that the wife would contribute her services too. In many cases where a
husband and wife worked as partners, the contribution of the wife was
not of¬cially acknowledged. One eighteenth-century observer noted a
farmer who was assisted by his wife: “a large occupier of £17,000 a year,
who was able to manage without a steward or bailiff, because he had the
assistance of ˜his lady, who keeps his accounts™ .”3 A farmer™s wife was
frequently his business partner, taking over the management of the dairy
and the poultry. Wool manufacture was also a family business; Joseph
Coope, a Yorkshire clothier, noted that he had a servant and two
apprentices, “which is the whole I employ, except my wife and myself.”4
We have enough evidence of this type to con¬rm that many wives
worked with their husbands. In cottage industry the value of the output,
such as a piece of cloth woven, was often counted as the man™s earnings,
even though much of the work was actually done by his wife or children.
Unfortunately, we do not have the means to measure the extent of this
work. In most cases the contribution of the wife to the family business
went unnoticed and unrecorded.

I. Measuring occupational segregation
The ¬rst problem I will address is how to measure occupational sorting.
The statistical evidence is unfortunately inadequate; the only aggregate
data on employment comes from the census, which does not list occu-
pations of individuals before 1841. Even at this late date, the census
systematically underrecords female employment. Left without a com-
prehensive measure of employment, I use other measures to establish
occupational sorting by gender. First, I show that the percentage of
women employed varied greatly by industry. Then I use commercial
directories to measure occupational segregation for a speci¬c segment of
the labor market “ business owners. Both of these measures con¬rm that
men and women tended to work in different occupations.

2
F. M. Eden, State of the Poor (London: Davis, 1797), vol. II, p. 687. In other cases,
married couples working as governor and governess received separate salaries. It was
fairly common, however, to give one salary to a husband and wife team. John Moss and
his wife received £50 a year to be master and mistress of the Preston workhouse. BPP
1816 (397) III, p. 181.
3
Ivy Pinchbeck, Women Workers and the Industrial Revolution 1750“1850 (London:
Routledge, 1930), p. 8. The observer was Marshall, Rural Economy of Norfolk, 1782.
4
BPP 1806 (268) III, p. 31.
18 Gender, Work and Wages in Industrial Revolution Britain

A. The census
The census is usually the ¬rst place a historian looks for information on
employment patterns because it provides the only complete measures of
employment in the entire economy. Table 1.1 shows the occupational
distribution from both the 1841 and 1851 censuses. These numbers
suggest low rates of female labor force participation: in the 1841 census
only 25 percent of females over age 10 had an occupation, and in the
1851 census only 35 percent. Women who did work were heavily con-
centrated in a few occupations. Three categories “ domestic services,
textiles, and clothing “ accounted for 85 percent of the female workers in
1841 and 80 percent in 1851. The same categories held only 22 percent
of male workers in 1841 and 20 percent in 1851. This stark contrast has
been noted by many historians.5
Unfortunately, the census numbers are not an accurate measure of
women™s employment. While Hatton and Bailey conclude that the
censuses of the early twentieth century accurately measured women™s
labor force participation, the same cannot be said of the 1841 and 1851
censuses.6 Edward Higgs has studied the censuses extensively and
concluded that the census numbers should not be considered raw data,
but rather cultural objects generated by ideology.7 The census data were
collected by men who built some of their cultural ideology into the data.
The assumption that the household, rather than the individual, was the
working unit is re¬‚ected in the way the census data were collected. The
1811 to 1831 censuses collected information on the number of families,
not individuals, in three broad occupational categories.8 Individual
enumeration began with the 1841 census, but knowledge of the occu-
pation of the household head was considered suf¬cient. The 1841
census instructed the enumerators to ignore a large fraction of women
workers; the instructions state, “The professions &c. of wives, or of sons
or daughters living with and assisting their parents but not apprenticed

5
For example, see Elizabeth Roberts, Women™s Work, 1840“1940 (Cambridge:
Cambridge University Press, 1988), ch. 2, and Jane Rendall, Women in an
Industrializing Society: England 1750“1880 (Oxford: Basil Blackwell, 1990), pp. 55“6.
6
Timothy Hatton and Roy Bailey, “Women™s Work in Census and Survey, 1911“1931,”
Economic History Review 54 (Feb. 2001), pp. 87“107.
7
“If the census reveals itself as part of the process by which gender divisions were de¬ned,
it cannot be used uncritically to study gender divisions in Victorian society. Such
quantitative data is not necessarily ˜raw material™ for unbiased scienti¬c analysis, it is also
a human construct and therefore a worthy, and indeed necessary, subject for historical
analysis.” Edward Higgs, “Women, Occupations and Work in the Nineteenth Century
Censuses,” History Workshop Journal 23 (1987), pp. 76“7.
8
The categories were “agriculture; trade, manufactures, and handicraft, and the number
not occupied in the preceding classes.” Higgs, Making Sense of the Census, pp. 22“3.
Table 1.1. Occupations in the 1841 and 1851 censuses: Great Britain

1841 census 1851 census

Males Females Males Females

Occupational Percent Percent
category 1000s % 1000s % female 1000s % 1000s % female

40 3 64 3
Public administration 0.8 0.2 7.0 1.0 0.1 4.5
51 0 63 0
Armed forces 1.0 0 0.0 1.0 0.0 0.0
113 49 162 103
Professions 2.2 2.7 30.2 2.5 3.6 38.9
255 989 193 1135
Domestic services 5.0 54.5 79.5 2.9 40.1 85.5
94 1 91 0
Commercial 1.8 0.1 1.1 1.4 0.0 0.0
196 4 433 13
Transport and 3.8 0.2 2.0 6.6 0.5 2.9
communications
1434 81 1788 229
Agriculture 28.2 4.5 5.3 27.3 8.1 11.4
24 0 36 1
Fishing 0.5 0.0 0.0 0.6 0.0 2.7
218 7 383 11
Mining 4.3 0.4 3.1 5.9 0.4 2.8
396 14 536 36
Metal manufacture 7.8 0.8 3.4 8.2 1.3 6.3
376 1 496 1
Building and 7.4 0.1 0.3 7.6 0.0 0.2
construction
107 5 152 8
Wood and furniture 2.1 0.3 4.5 2.3 0.3 5.0
48 10 75 15
Bricks, cement, pottery, 0.9 0.6 17.2 1.1 0.5 16.7
glass
23 1 42 4
Chemicals 0.5 0.1 4.2 0.6 0.1 8.7
47 3 55 5
Leather and skins 0.9 0.2 6.0 0.8 0.2 8.3
44 6 62 16
Paper and printing 0.9 0.3 12.0 0.9 0.6 20.5
525 358 661 635
Textiles 10.3 19.7 40.5 10.1 22.4 49.0
Table 1.1. (cont.)

1841 census 1851 census

Males Females Males Females

Occupational Percent Percent
category 1000s % 1000s % female 1000s % 1000s % female

358 200 418 491
Clothing 7.0 11.0 35.8 6.4 17.3 54.0
268 42 348 53
Food, drink, lodging 5.3 2.3 13.5 5.3 1.9 13.2
476 41 445 75
Other 9.3 2.3 7.9 6.8 2.6 14.4
5093 1815 6545 2832
Total occupied 100.0 100.0 26.3 100.0 100.0 30.2
1604 5369 1060 5294
Total unoccupied
6697 7184 7605 8126
Total individuals over 51.8 51.7
age 10
Labor force 76.0 25.3 86.1 34.9
participation rate

Note: % ¼ percentage of all occupied males or females in this occupational category.
Percent female ¼ percentage of individuals in this occupational category who were female.
Source: B. R. Mitchell, Abstract of British Historical Statistics (Cambridge: Cambridge University Press, 1962), p. 60.
Women™s occupations 21

or receiving wages, need not be inserted.”9 Because of this aspect of the
culture, the work of women was seriously undercounted, particularly in
1841. Table 1.1 suggests that female labor force participation rates were
25 percent in 1841, but 35 percent in 1851. On the surface this dif-
ference looks like a large increase in labor force participation, but it
would be an error to conclude that this represents a real change, or that
three-fourths of women did not work in 1841. The apparent increase
just re¬‚ects how drastically women were undercounted in 1841. The
1851 census is an improvement in this respect, since it does ask that the
occupations of wives be included. Even in 1851, however, the problem
was not eliminated; women workers continued to be undercounted
because women workers were more likely than men to be part-time,
seasonal, and home workers, and because census enumerators expected
women to be dependents.
Historians have debated the extent of errors in the census counts.
Edward Higgs has suggested there are serious errors in the counting of
domestic servants that would make the occupational distribution of
females appear more skewed than it actually was, while Michael
Anderson claims the problem is overstated by Higgs. In a survey of the
returns of Rochdale, Lancashire, Higgs found that only 56 percent of
people recorded as servants were “servants in relationship to the head of
the household in which they lived.”10 Some of these people were
probably servants working elsewhere but living at home. Many of these,
however, would be better described as housewives; they were just female
family members who did the housework. Higgs found that “For some
enumerators ˜housekeeper™ and ˜housewife™ were synonymous.”11 While
these women were clearly workers, they were not domestic servants in the
sense in which we generally use the term. While the exact amount of
overcounting is not known, the potential for error is very large. For
example, if the number of servants was reduced by taking out family
members designated as “servants,” the number of servants in Rochdale in
1851 would be reduced by one-third.12 Even among those who were
actually hired servants, many were allocated to the wrong industry; many
of the female servants recorded in the domestic service industry spent
more time working in agriculture or trade rather than in domestic work.13

9
Quoted in ibid., p. 81.
10
Edward Higgs, “Domestic Service and Household Production,” in Angela John, ed.,
Unequal Opportunities (Oxford: Basil Blackwell, 1986), p. 130.
11
Ibid., p. 131. 12 Ibid., p. 132.
13
Higgs, “Women, Occupations and Work.” Among farm servants, men were most likely
to be allocated to the agricultural sector, while women were likely to be classi¬ed as
domestic servants.
22 Gender, Work and Wages in Industrial Revolution Britain
Table 1.2. The occupations of women workers: Higgs™s revisions of census data (percentage of
occupied women)

1841 1851

Sector Census Revised Census Revised

Agriculture 3.9 33.2 7.0 27.4
Mining 0.4 0.3 0.4 0.3
Building 0.1 0.1 0.0 0.0
Manufacture 32.0 28.0 42.7 39.4
Transport 0.2 0.1 0.5 0.4
Dealing 3.8 11.4 4.5 10.3
General laborers 0.8 0.7 0.3 0.3
Public service/professions 3.1 2.7 4.1 3.8
Domestic service 55.7 23.4 40.4 17.9

Note: The “census” ¬gures are not directly from the census, but were revised to allow
comparability across all nineteenth-century censuses. The revision adds corrections for the
wives of tradesmen and the wives of agricultural workers, who are assumed to work one-
sixth of the year, and moves some women from the “domestic service” category to the
agricultural, retailing, and “dependent” sectors.
Source: Higgs, “Women, Occupations and Work,” Tables 4 and 5.


While a male servant hired by a farmer would be counted as an agri-
cultural worker, a female servant hired by a farmer might be counted as a
domestic servant even if she did agricultural work. Thus Higgs suggests
that the census data understate the participation of women and overstate
the skewedness of the occupational distribution. Higgs revised the
census ¬gures to correct for seasonal work in agriculture, the under-
counting of working wives, and the overcounting and mis-allocation of
domestic servants. The results of this revision, shown in Table 1.2, tell a
much different story. If Higgs is correct, the occupational distribution
was not so heavily skewed toward domestic service, and had more
women in agriculture, which was the most common occupation for men.
Michael Anderson, however, has questioned whether the problem is
as bad as Higgs suggests. Rochdale does not seem to be representative of
the entire country. Anderson ¬nds that a national sample of census
enumeration books suggests much lower numbers of women related to
the household head who were recorded as “servant” or “housekeeper.”14
Servants who were related to the household head may have been visiting
their families, since the 1851 census was taken on Mothering Sunday.15

14
Michael Anderson, “Mis-Speci¬cation of Servant Occupations in the 1851 Census: A
Problem Revisited,” Local Population Studies 60 (1998), pp. 59“60.
15
Ibid., p. 61.
Women™s occupations 23

Anderson™s evidence suggests that the overcounting of servants was
much smaller than Higgs suggested, but not entirely absent. Anderson
estimates that 11 percent of those listed as “domestic servant” and
58 percent of those listed as “housekeeper” were related to the head of
household.16 Higgs™s corrections, then, are too extreme, and should not
be taken as an accurate measure of the occupational distribution, but
they do demonstrate that the errors present in the census data could
potentially distort the occupational distribution.
Overcounting of domestic servants is not the only problem with the
census data. There is reliable evidence that many women who were
employed outside the home for wages were not listed as employed in
the censuses. Andrew Walker notes that, while the owner of a Dar¬eld
stone quarry is listed in the 1881 census as employing nine women, no
women in that enumeration district are listed as having the occupation
of stone worker, suggesting that the census enumerator probably failed
to record the occupations of some women.17 Miller used evidence from
Gloucestershire farm wage books to show that female employment in
agriculture was underenumerated in the censuses of the late nineteenth
century. Individual women who were clearly employed in agriculture,
and received wages that were recorded in an account book, are not
recognized by the census as employed. Miller matched the names of
females in the farm wage books to the 1871 censuses and found that
eleven of the seventeen women matched were returned by the census as
having no occupation. For example, Anne Westbury worked 221½ days
at a farm in Fairford, but the 1871 census does not list an occupation for
her.18 Nicola Verdon has done the same for a farm in the East Riding of
Yorkshire; fourteen women were employed on this farm but not listed as
agricultural laborers in the 1881 census. My own estimates suggest that
the 1851 census records less than half of the female out-door laborers in
agriculture.19 Leigh Shaw-Taylor has defended the reliability of census
on female employment, claiming that the employment of women who
worked regularly was well recorded. He notes that irregular employment
was underrecorded, but does not consider that a serious fault because


16
Ibid., p. 63.
17
Andrew Walker, “ ˜Pleasurable Homes™? Victorian Model Miners™ Wives and the
Family Wage in a South Yorkshire Colliery District,” Women™s History Review 6 (1997),
pp. 317“36.
18
Miller, “The Hidden Workforce,” p. 146. See also Helen Speechley, “Female and
Child Agricultural Day Labourers in Somerset, c. 1685“1870,” unpublished PhD
thesis, University of Exeter, 1999.
19
Joyce Burnette, “The Wages and Employment of Female Day-Labourers in English
agriculture, 1740“1850,” Economic History Review 57 (2004), pp. 664“90.
24 Gender, Work and Wages in Industrial Revolution Britain

the censuses were not meant to measure irregular work.20 However, if
we wish to obtain an accurate picture of women™s employment we
cannot afford to ignore irregular work. Much of the work women did
was irregular, and con¬ning ourselves to regular work will produce a
skewed picture of female participation in the labor market.
The nature of women™s work during the Industrial Revolution means
that it could not be well recorded by the census. The censuses recorded
each individual as either having an occupation or not, and generally only
one occupation was listed per person.21 This was not a good system for
recording women™s work during the Industrial Revolution period, which
has been described as an “economy of makeshifts.”22 Many women did
not pursue one type of employment exclusively, but survived by com-
bining many different kinds of employment with other sources of
income. Peter King estimated that, by gleaning, women and children
could earn between 3 and 14 percent of a laborer™s family income, and
Steven King has argued that poor women combined poor law payments
with work income in order to make ends meet.23 Women who worked as
agricultural day-laborers usually worked only a few days in a year. Of
the seventy-one different women who appear in the wage book of the
Estcourt farm in Gloucestershire between 1828 and 1849, ¬fty-nine
women (83 percent) were casual workers in the sense that they worked
fewer than sixty days in a year.24 At the Oakes farm in Derbyshire,
approximately half of all days worked by women were worked during the
two-week hay harvest, so the vast majority of women hired at this farm



20
“However, it is very clear that irregular work by women was under-recorded in
1851, but largely because the G.R.O. did not want to know about such work.” Leigh
Shaw-Taylor, “Diverse Experience: The Geography of Adult Female Employment in
England and the 1851 Census,” in Nigel Goose, ed., Women™s Work in Industrial
England: Regional and Local Perspectives (Hat¬eld: Local Population Studies, 2007),
p. 40.
21
This point was made by Andrew August, “How Separate a Sphere? Poor Women and

. 1
( 14)



>>