. 9
( 14)


BPP 1844 (587) XXVII. The summary tables for male occupations from the 1851
census do not list staymaker as a separate category.
Barriers to women™s employment 225

proper Lengths, it is thrust in between the Rows of Stitching: This requires a
good deal of Strength.4
In this passage Campbell both expresses the ideology which led to
gender segregation in the garment trades and explains why staymaking
was the exception. In staymaking, the strength requirement was great
enough to overcome the fear of physical intimacy between the sexes, and
in the eighteenth century most staymakers were men. However, in the
later eighteenth century the work became easier. Lane notes that “After
the mid-eighteenth century, stays were lighter, using less whalebone,
and the craft came within a woman™s capabilities.”5 Once staymaking no
longer required the strength of a man, women took over the trade. Thus
a dramatic change in the gender division of labor seems to have been the
result of changes in the amount of strength required.
Strength requirements, however, are not suf¬cient to explain the sorting.
Women did not have equal access to all occupations where strength was
unnecessary. Among the occupations in Table 5.1 not requiring strength,
women are clearly sorted into a few occupations. Nearly all the dress-
makers and milliners were women, and about half of the teachers, but
other occupations were dominated by men. Many of the occupations
requiring no strength, such as accountant, attorney, and clerk, had even
fewer women than occupations such as blacksmith and brass founder,
which is not consistent with physical comparative advantage. The absence
of women from white-collar occupations such as law cannot be explained
by physical strength and must have been the result of exclusion. The
clearest gender division is in the garment trades; women were dressmakers
but not tailors, though the two occupations are very similar. Wages were
much lower in dressmaking or millinery than in tailoring, suggesting that
women were not free to choose the more lucrative trade.6 The same
occupational patterns are evident in the commercial directories. Table 5.2
shows that, while women are in general less likely to appear in occupations
requiring strength, there were many occupations requiring no strength that
had no women at all, or even fewer women than the blacksmith trade.
Manchester had no female accountants, auctioneers, or hairdressers.
Strength certainly does not explain all the occupational sorting.
Human capital, while it is clearly important in occupational choice, is
not suf¬cient to explain the sorting either. Even when they had similar

R. Campbell, The London Tradesman (New York: Augustus M. Kelley, [1747] 1969),
pp. 224“5.
Lane, Apprenticeship in England, p. 123.
In 1800 journeywomen milliners earned 6s. a week in Colchester, while journeymen
tailors in London earned 27s. a week. Davidoff and Hall, Family Fortunes, p. 302, and
BPP 1840 (43) XXIII, p. 582.
226 Gender, Work and Wages in Industrial Revolution Britain

levels of human capital, men and women worked in different occupations.
Women often did invest in human capital, but these women generally
ended up in teaching or dressmaking. Though dressmaking and tailoring
required similar human capital investments, the gender segregation bet-
ween the two trades was almost complete. In 1788 all the tailors in
Manchester were male, and all the dressmakers were female.7 In 1846
only one out of 240 tailors was female, and only 12 out of 257 milliners
and dressmakers were male.8 The large number of women who entered
the dressmaking trades, whose apprenticeship fees were substantial,9 is
evidence that many were willing to invest in human capital. However,
wages in dressmaking were low compared to wages in tailoring, suggesting
that the concentration of women in dressmaking was due to exclusion
rather than choice.
Choice is not a suf¬cient explanation for the absence of women from
certain skilled trades, because some women attempted to work in skilled
trades but were rebuffed by men who excluded them from the occupa-
tions. Tailors in London went on strike to prevent women from entering
their trade, as did mule-spinners in Glasgow.10 The hatters of Stockport
agreed to strike against any employer who hired women.11 The opposition
of male workers to the employment of women will be examined later in
this chapter. For now it is suf¬cient to note that women who wished to
work in certain skilled trades were prevented from doing so. The women
who were affected resented the constraint. In a letter to The Pioneer in
1834, a woman criticized the tailors™ union, which was engaged in a strike
to keep women out of the tailoring trade:
Surely the men might think of a better method of bene¬ting themselves than that
of driving so many industrious women out of employment. Surely, while they
loudly complain of oppression, they will not turn oppressors themselves. Surely
they will not give their enemies cause to say, when a woman and her offspring are
seen begging in the streets, “ This is the work of union.12

Lewis™s Manchester Directory for 1788.
These calculations ignore eighteen tailors and eighteen milliners and dressmakers
whose gender could not be determined. Slater™s National Commercial Directory of Ireland,
1846; Slater™s Royal National and Commercial Directory, 1850.
Neff, Victorian Working Women, p. 117, quotes premiums of £40 to £60 for
dressmaking in the mid-nineteenth century, and Sally Alexander, Women™s Work in
Nineteenth-Century London: A Study of the Years 1820“1850 (London: Journeyman
Press, 1983), p. 34, quotes premiums of £30 to £50.
Barbara Taylor, Eve and the New Jerusalem: Socialism and Feminism in the Nineteenth
Century (New York: Pantheon Books, 1983), ch. 4, and BPP 1824 (51) V, p. 525.
A. Aspinall, The Early English Trade Unions: Documents from the Home Of¬ce Papers in the
Public Record Of¬ce (London: Batchworth Press, 1949), p. 107.
The Pioneer, March 19, 1834, quoted in Taylor, Eve and the New Jerusalem, p. 108.
Barriers to women™s employment 227

Women demonstrated a desire to work, but they were not allowed to
enter many trades. Thus the low numbers of women in these occupa-
tions are partially due to occupational segregation constraints. In the
skilled labor market, in contrast to the unskilled labor market, gender
discrimination was an important part of the story.
Many factors contributed to the lower numbers of women in most of
the skilled craft, white-collar, and professional occupations. First, the
fact that many women workers were unrecorded means that their
measured participation is lower than their actual participation. Second,
women may have chosen not to invest in the human capital necessary for
skilled work because they expected to spend less time in the labor force.
This choice was the indirect result of the assignment of women to
household tasks, which may be the result of either comparative advan-
tage or gender discrimination within the family. Third, women in higher
social classes may have felt more social pressure to stay out of the labor
force, resulting in fewer women with the means to enter business willing
to do so. Fourth, women may simply have been excluded from certain
occupations, and not permitted entry when they tried to enter skilled
occupations. While all of these factors had some in¬‚uence on the relative
absence of women from high-paid occupations, in this chapter I will
focus mainly on the fourth, the direct method of exclusion. I do
acknowledge, though, that the other factors may have played an
important role. Evidence of explicit barriers is not hard to ¬nd, and in
light of such evidence we can conclude that gender discrimination
reduced the number of women in skilled and professional occupations.
Having established that the sorting models presented in Chapter 3 do
not fully explain the gender division of labor for skilled craft, white-
collar, and professional occupations, I turn to examining the possible
sources of the occupational barriers that prevented women from working
in occupations where they could have been productive. I divide the
discussion of barriers to entry into two parts: occupational barriers for
employees in skilled crafts, and occupational barriers in business and
professional occupations. This chapter examines the barriers faced by
employees, and the following chapter examines the barriers faced by
women in self-employment.

II. Sources of occupational barriers among employees
We have seen, in Chapter 3, that in some parts of the labor market,
economic forces operated freely and the division of labor between the
sexes was not the result of discriminatory barriers. In other occupations,
there clearly were barriers keeping women out, even women who
228 Gender, Work and Wages in Industrial Revolution Britain

expressed a desire to work in those occupations. What made these
segments of the labor market different? This section will argue that
exclusionary barriers appeared in less competitive parts of the labor
market, where control of an important skill allowed the workers in that
occupation to limit competition. Where work was unskilled, or the skills
needed were easily learned, competitive markets ensured that workers
were allocated according to their comparative advantage.
Some of the potential sources of exclusion were not, in fact, important
in creating occupational segregation. I argue in this section that gov-
ernment regulations, guilds, and employers, while they may have
occasionally contributed to the exclusion of women, were not important
sources of occupational barriers. None of these factors was capable of
explaining a substantial amount of occupational segregation. By con-
trast, the actions of employee organizations such as unions explain a
great deal of occupational segregation. Male workers actively fought to
exclude women from their occupations, and often won. The role of
unions in limiting women™s employment opportunities explains why
unskilled occupations were less likely to have barriers to women™s
employment, and why skilled occupations were more likely to have such
barriers. Women were not excluded from unskilled occupations because
unions had no power in those occupations. Only in skilled occupations,
where the unions had the power to enforce their desires, could the
exclusionary desires of the male workers overrule the desire of employers
to hire women. An important part of this explanation is the claim that
unions were successful in skilled occupations, but not in unskilled
occupations. To establish this fact, I will turn to the historical record and
examine unions in different occupations. Once I have shown that unions
were not successful in unskilled occupations, and that unions were the
most important source of exclusion, I can explain why women were not
excluded from unskilled occupations.

A. Government regulation
While government regulation has been blamed for gender segregation,13
these laws were not an important source of occupational segregation
during the Industrial Revolution. Government restrictions on women™s
employment only appeared at the very end of the time period, and thus

Honeyman and Goodman, “Women™s Work, Gender Con¬‚ict,” p. 622, claim that
protective legislation “was designed to reinforce the position of women as wives and
mothers” and “provided the capitalist with the opportunity to remove women from the
factory into the more economical environment of the sweatshop.”
Barriers to women™s employment 229

are a characteristic of the Victorian era, but not of the Industrial
Revolution. Even when they were in place, these laws had relatively little
effect on women™s employment.
Laws limiting women™s work did not appear until the 1840s. The ¬rst
act restricting women™s employment was the Mines and Collieries Act
of 1842, which prohibited women from working underground in col-
lieries.14 This was the only act to completely exclude women from an
occupation. The ¬rst factory act to apply to adult women was the Bill of
1844, which limited their hours of work to twelve.15 Women™s hours
were further limited to ten in 1847. Thus government regulation cannot
be blamed for any occupational segregation that occurred before 1842.
Also, we must not assume that government regulations were always
effective. To determine the effect of laws restricting women™s employ-
ment we must look at their actual effect, rather than just their wording or
intent. The 1842 Mines Act, which outlawed the employment of women
underground, seems to have reduced the opportunities for women to
work in mines, but did not completely exclude women. The 1851 census
reports 2535 female coal miners, a slight increase from the 2350
reported in the 1841 census. During the same period, though, the
number of male coal miners increased 58 percent, so the law seems to
have slowed the growth of female employment in coal mining.16 How-
ever, it does not seem to have completely prevented women from
working in mines. The 1842 law did not prevent women from working
above ground, so the 2535 female coal miners reported in 1851 may
have been working legally, but there is other evidence that the law was
not always well enforced. Angela John ¬nds that women continued to
work underground, some of them dressing as men.17 In some cases, the
existence of illegal women workers was revealed only when they died in
mine accidents.18 The government provided only one commissioner for
the enforcement of this law, and the ¬ne for an offense was too low to be
a serious deterrent (£10 maximum).19 In 1845 the inspector estimated
that 200 women were working illegally in Wigan. One employer even

5 & 6 Vict. c. 99. Laws regulating child labor began earlier, in 1802.
B. L. Hutchins and A. Harrison, A History of Factory Legislation (Westminster: P. S. King &
Son, 1903), ch. 4.
BPP 1844 (587) XXVII and 1852“3 (1691) LXXXVIII.
Angela V. John, By the Sweat of their Brow: Women Workers at Victorian Coal Mines
(London: Croom Helm, 1980), pp. 55“8, and “Colliery Legislation and Its
Consequences: 1842 and the Women Miners of Lancashire,” Bulletin of the John
Rylands Library 61 (1978), pp. 78“114.
For example, Hannah Hatharington was killed in February 1845 when a mine roof fell.
Ibid., p. 99.
Ibid., p. 79.
230 Gender, Work and Wages in Industrial Revolution Britain

complained that some men refused to work for him because he would
not employ their wives and daughters.20 The 1842 law seems to have
reduced the demand for female mine workers without completely
eliminating their use. Petitions from workers asking parliament to repeal
the prohibition suggest that the workers felt they were harmed, and at
least one woman noted that she earned lower wages after the law.21
The factory acts limiting the hours of employment of women and
children were also widely evaded. The early acts applied only to children
and were not very effective. By the time women were included in the
factory acts in the 1840s, enforcement had been improved somewhat.
Factory inspectors were appointed in 1831. Still, the acts of 1844 and
1847, which were the ¬rst to limit women™s hours of employment, were
dif¬cult to enforce. Employers worked women and children on shifts,
while the men worked longer hours, and “under such a system the Ten
Hours Act was completely nulli¬ed, and it was impossible for the
inspectors to detect overtime employment.”22 If the factory was oper-
ated more than ten hours, it was dif¬cult to monitor whether any women
and children worked more than ten hours. One factory inspector com-
plained that if the factories used shift work, “no practical system of
inspection could prevent extensive fraudulent overworking.”23 The laws
limiting women™s employment were not effective until 1850, when
parliament limited the number of hours the factory machinery could
Even when they were enforced, factory acts did not prohibit women
from working, but simply limited the number of hours they could work
and thus made them less useful to the employer than male workers.
Employers might have chosen to employ only men, who could work
longer hours, but they did not do so. Employers continued to employ
women, and the percentage of women employed in factories even rose
slightly between 1838 and 1856, from 55 to 57 percent of the work-
force.24 This increase may have been less than what would have
occurred in the absence of the law, but the law clearly did not lead to
reductions in female factory employment. The hours limitation may still
have harmed women if it prevented them from taking the more highly
paid factory jobs. Rose claims that the effect of the acts was to keep
women out of certain higher-wage jobs: “although women and men both

Ibid., p. 104.
This woman claimed that she earned 3s. less per week as a result of the law. Ibid., p. 10.
Hutchins and Harrison, A History of Factory Legislation, p. 102.
BPP 1849, XXII, p. 135, quoted in ibid., p. 102.
Women over 13 were 55.2 percent of the factory labor force in 1838, 55.9 percent in
1850, and 57.0 percent in 1856. Ibid., ch. 6.
Barriers to women™s employment 231

worked as cotton weavers and earned equal piece rates, legally men, and
men only, could clean their machines after working hours, so some men
could and did earn higher weekly wages than women.”25 So it is possible
that hours regulations hurt women™s opportunities.
While government regulations had some marginal impact on
women™s employment opportunities, such effects were limited to a few
industries (mining and textile factories). When in place, laws limiting
women™s employment did not prevent the use of women workers. More
importantly, government regulations cannot have been an important
source of occupational barriers during the Industrial Revolution because
the regulations appeared too late. None of the laws passed before 1842
applied to adult women. We must look elsewhere for the causes of
occupational segregation.

B. Gender ideology
During the nineteenth century we see the rise of a number of different
ideologies which may have in¬‚uenced women™s opportunities. The
family wage ideal suggested that the male head should earn a wage high
enough to support his whole family, without his wife or children working
to contribute to the family™s income.26 This idea was new to the nine-
teenth century; earlier generations expected women and children to
contribute.27 By the late nineteenth century, the ideal of the family wage
was strong enough that men felt a loss of status if their wives worked.
The nineteenth century also saw the rise of domestic ideology, which
assumed that a woman™s place was in the home.28 While households in
earlier generations often included servants or apprentices, the household
was increasingly restricted to the nuclear family (even apprentices no
longer lived “in”), and the home was increasingly seen as a retreat from
the world. The woman™s role was to be the “angel of the house.” The
woman™s sphere of in¬‚uence, the home and family, was separate from
the man™s sphere of in¬‚uence, which included the outside world.29

Rose, Limited Livelihoods, p. 74.
See Hilary Land, “The Family Wage,” Feminist Review 6 (1980), pp. 55“77; Rose,
“Gender at Work”; and Wally Seccombe, “Patriarchy Stabilized: The Construction of
the Male Breadwinner Wage Norm in Nineteenth-Century Britain,” Social History 11
(1986), pp. 53“76.
Pinchbeck, Women Workers, pp. 1“2.
See Joan Scott and Louise Tilly, “Women™s Work and the Family in Nineteenth-
Century Europe,” Comparative Studies in Society and History 17 (1975), pp. 36“64.
See Roberts, Women™s Work, 1840“1940, pp. 4“5; Simonton, European Women™s Work,
pp. 87“90.
232 Gender, Work and Wages in Industrial Revolution Britain

Another attitude that may have in¬‚uenced women™s work opportun-
ities was concern about sexual morality. Parliamentary reports of the
early nineteenth century show great concern with sexual purity and the
moral results of mixing men and women. The 1843 parliamentary report
on Women and Children in Agriculture shows that the clergy and gentry
were greatly concerned with the moral effects of crowded cottages and
single men and women working together. For example, parliamentary
investigator Alfred Austin worried that:
The sleeping of boys and girls, and young men and young women, in the same
room, in beds almost touching one another, must have the effect of breaking
down the great barriers between the sexes, “ the sense of modesty and decency
on the part of women, and respect for the other sex on the part of the men. The
consequences of the want of proper accommodation for sleeping in the cottages
are seen in the early licentiousness of the rural districts.30

Humphries has suggested that concerns about the consequences of
sexual activity led to a separation of the sexes at work.31
We do observe people using gender ideology to explain constraints
on women™s employment. In 1845 the male potters™ union justi¬ed
its opposition to female employment with this appeal: “To maidens,
mothers, and wives, we say machinery is your deadliest enemy . . . It will
destroy your natural claims to home and domestic duties.”32 However,
the question still remains whether the domestic ideology these men
appealed to was the real cause of the barriers, or simply justi¬cation for
barriers that had other causes. Other historians have noted that men may
have been using the ideology as a cover for pursuing their economic
interests. For example, Hilary Land, in her discussion of the concept of
the family wage, a wage high enough for a man to support his family
without his wife working, notes that:
It is dif¬cult to know how far skilled and organized working-class men, that is the
labor aristocracy, accepted this form of marriage relationship as an ideal or merely
couched their arguments in terms which would appeal to the social reformers and
some sections of the capitalist class in order to further their own ends.33

While gender ideology may have played a role in declining participation
(see Chapter 7), I do not think it was the most important cause of
occupational barriers, because it could be disregarded when more
powerful economic interests were at stake. While both workers and

Women and Children in Agriculture, p. 24. 31 Humphries, “Most Free from Objection.”
Quoted by Barbara Drake, Women in Trade Unions (London: Virago Press, [1920] 1984),
p. 6.
Land, “The Family Wage,” p. 57.
Barriers to women™s employment 233

employers shared similar gender ideology, they took different sides in the
battle over whether women could be hired for certain jobs. As described
in the introduction, Glasgow mule spinners fought a violent battle with
their employers about whether women could be hired as mule spinners.
The workers fought for restrictions because such restrictions limited the
supply of workers and thus allowed them to earn higher wages.
Employers, on the other hand, fought for the right to hire women
because the larger labor supply and lower wages would increase their
pro¬ts. Gender ideology, then, was not suf¬cient to enforce restrictions
on women™s employment, which were only effective when a group with
market power had an economic incentive to enforce such restrictions.

C. Guilds
Guilds have been an important source of barriers to women™s employ-
ment in certain times and places. Sheilagh Ogilvie has shown that they
were a powerful force in limiting German women to a few low-paid
occupations.34 Women were allowed to work in skilled trades only as
wives and widows, indicating that they could and did acquire the
necessary skills, but females were not admitted to guilds as apprentices.
In England, however, guilds were both less powerful and more open to
females. Even if guilds did create barriers excluding women in the pre-
industrial period, by the nineteenth century they had lost their monopoly
power over employment and thus had lost any power they might have
once had to exclude women from well-paying occupations.
Guilds were a product of the pre-industrial economic system. They
organized workers into three classes: apprentices, who were bound to a
master for a number of years; journeymen, who worked for wages; and
masters, who set up in business for themselves. Both the employers
(masters) and the employees (journeymen) were members of the same
guild. Before the Industrial Revolution, journeymen could usually
expect to becomes masters themselves in the space of a few years
because capital requirements for setting up independent shops were low.
Guilds included both masters and journeymen because there was rela-
tively little distinction between the two. The Webbs note that “it was the
prospect of economic advancement that hindered the formation of per-
manent combinations among the hired journeymen of the Middle Ages.”35

Sheilagh Ogilvie, A Bitter Living, and “Guilds, Ef¬ciency, and Social Capital: Evidence
from German Proto-Industry,” Economic History Review 57 (2004), pp. 286“333.
Sidney Webb and Beatrice Webb, The History of Trade Unionism (London: Longmans,
Green, & Co, 1894), p. 7.
234 Gender, Work and Wages in Industrial Revolution Britain

In other words, the masters and journeymen did not have industrial
disputes because their interests coincided. Masters as well as jour-
neymen saw new workers as potential competitors. With the Industrial
Revolution, however, the economic interests of the two groups
diverged. Capital requirements increased, and it became harder for
journeymen to become masters. If journeymen were not able to set up
as masters and had to remain wage earners their whole lives, masters
no longer had an interest in limiting their numbers. Journeymen,
however, wished to maintain strict limits on entry, in order to keep
their wages up. When the interests of masters and journeymen began
to diverge, the guild became obsolete, and journeymen organized
themselves into unions whose purpose was to oppose the masters.
Since unions represented a different industrial structure from guilds
and had different interests, they will be investigated separately.
Guilds included both employers and employees and thus were not
likely to take actions that bene¬ted employees but hurt employers. The
interests protected by the guild were those of the small master. Entry to
the trade was restricted, to limit competition. The main tools the guild
used to limit the supply of its product were to allow only apprenticed
individuals to practice the trade, and to limit the number of apprentices.
If the guild successfully enforced both of these limits, then it effectively
held monopoly power over the trade and could enjoy monopoly rents.
While usually justi¬ed as training, apprenticeship rules were also about
restricting entry. Apprenticeships of up to seven years were required in
trades that could be learned in a few months. Citing Defoe™s claim that
clock- and watch-making could be learned in a few weeks, Simonton
suggests that “the training had less to do with expertise and rather more
to do with the status carried by the trade.”36 Dunlop concludes that in
sixteenth-century England apprenticeship was “a formidable weapon in
the hands of the guilds” and that it “could be employed as an instrument
of monopoly.”37 In England the prohibition against unapprenticed
workers was supported by law; under the Elizabethan Statute of Arti¬-
cers only those apprenticed to a trade could practice it. Masters could
limit entry to the trade by limiting the number of apprentices they took.
There was still a free rider problem, since individual masters could
bene¬t individually from cheap labor if they took more apprentices,
while all masters shared the costs of the increase in supply. For this

Deborah Simonton, “Apprenticeship: Training and Gender in Eighteenth-Century
England,” in Maxine Berg, ed., Markets and Manufactures in Early Industrial Europe
(London: Routledge, 1991), p. 230.
O. Jocelyn Dunlop, “Some Aspects of Early English Apprenticeship,” Transactions of the
Royal Historical Society 3rd series, 5 (1911), pp. 193“208.
Barriers to women™s employment 235

reason guilds usually had explicit rules about how many apprentices a
master could have.
In England, these exclusionary barriers were not aimed speci¬cally at
women. Women could and did participate in guilded trades. One way
that a woman could do so was as the wife or widow of a male member of
the trade. The trade was carried on by the family unit, and women were
an important part of that unit. Women worked as assistants to their
husbands, keeping the shop or supervising the apprentices. The fact that
a tradesman™s wife was his partner in business is evident in the custom of
admitting a widow to the guild upon the death of her husband. The
widow of a guild member could enter the guild, practice the trade, and
take apprentices, even if she had never been apprenticed herself. A
widow had the ability to carry on the business in her own right because
she had been actively involved in the trade while her husband lived.
Many English guilds, unlike continental guilds, also allowed females
to be apprenticed.38 Girls, as well as boys, could be apprenticed and
could enter the guild as “freemen.” Dunlop and Denman note that girls
were apprenticed in the carpenters™, wheelwrights™, and clockmakers™
trades.39 In most guilds males had an advantage over females because
the son of a master could enter the trade without a formal apprenticeship
while a daughter could not, but in a number of guilds men did not even
have this advantage. In London the butchers, carpenters, and drapers all
allowed the daughter as well as the son of a freeman to gain her freedom
(guild membership) by patrimony.40 Not all guilds admitted female
apprentices; Lane claims that girls “were rigorously excluded from
apprenticeships in the building or leathers trades, and the heavy metal
skills (wheelwright or blacksmith),” but these are trades that girls would
not have chosen anyway because of the high strength requirement.41
Guild rules were frequently written in inclusionary language. In the
records of the carpenters™ guild we ¬nd, “If any Apprentice or
Apprentices Marry or Absent themselves from their Master or Mistress
during their Apprenticehood, then within one month the Master or
Mistress is to Bring their Indenture to the hall to be Registered and

German guilds admitted widows, but did not allow girls to be apprenticed. See Ogilvie,
“Guilds, Ef¬ciency, and Social Capital.”
O. Jocelyn Dunlop and Richard Denman, English Apprenticeship and Child Labor: A
History (London: Unwin, 1912), p. 151.
Philip Jones, The Butchers of London (London: Secker and Warburg, 1976), p. 21; Jupp
and Pocock, The Worshipful Company of Carpenters, p. 544; Percival Boyd, Roll of the
Drapers™ Company of London: Collected from the Company Records and Other Sources
(Croydon: J. A. Gordon, 1934).
Lane, Apprenticeship in England, p. 39.
236 Gender, Work and Wages in Industrial Revolution Britain

Entered.”42 In 1704 the Curriers™ Company agreed that no “foreigners”
should be allowed to work if members of the guild were unemployed.
They decreed:
that ye Beadle shall goe with any ffreeman that is out of worke to any of ye places
where a fforiner is at worke & acquait ye Master or Mistress that it is ordered by
this Court that he or she turn away ye fforiner & sett ye freeman to worke which
if they disobey ye Master and wardens are to take such . . . with them as is
provided against disobedient members by ye Orders of this Company.43

The rule suggests that women were employers and members of the
Women who were guild members seem to have carried on their trade
in the same way as the men. We ¬nd women taking apprentices. In
records of the Carpenters™ Company we ¬nd “Richard Stevenson sonne
of Robt. Stevenson late of Dublin in the Kingedome of Ireland Pavier
bound to Anne Nicholson Widowe the Relict of Anthony Nicholson, for
eight years,” in April of 1686, and “Robert Harper sonne of William
Harper of Notchford in the county of Chesheire, bound to Abigail
Taylor for Seaven Yeares,” in June of 1692.44 Katherine Eyre, a member
of the London Carpenters™ Company, took three apprentices between
the years of 1701 and 1707.45 Records of the Witney Blanket Weavers™
Company show that on October 24, 1733, Richard Ash¬eld was bound
to Eliza Jefferson for seven years.46 In 1824 a mistress shipwright from
Liverpool, Mrs. Simpson, had four apprentices.47 Simonton ¬nds that
3 percent of those taking apprentices were women.48
Apprenticeship requirements did function as barriers to entry, but
these barriers were not particularly aimed at females. In England, girls
could be apprenticed. In 1742 the butchers of London admitted to the
freedom of the company Hester Maynard, who had been apprenticed to
Francis Baker.49 In 1815 Caroline Atherton was admitted to the
Drapers™ Company of London by right of apprenticeship.50 While he

Records of the Worshipful Company of Carpenters, 1913, vol. I, p. vii, quoted in Clark,
Working Life of Women, p. 173.
Curriers™ Company, Fair Copy Extracts from Court Minute Books, vol. I, fos. 216“20,
quoted in C. R. Dobson, Masters and Journeymen: A Prehistory of Industrial Relations
(London: Croom Helm, 1980), pp. 47“8.
Records of the Worshipful Company of Carpenters, 1913, vol. I, p. 189, quoted in Clark,
Working Life of Women, p. 174. See also Jupp and Pocock, The Worshipful Company of
Carpenters, p. 161.
Jupp and Pocock, The Worshipful Company of Carpenters, pp. 543“4.
Alfred Plummer, The Witney Blanket Industry: The Records of the Witney Blanket Weavers
(London: Routledge, 1934), p. 161.
BPP 1824 (51) V, p. 227. 48 Simonton, “Apprenticeship,” p. 245.
Jones, The Butchers of London, p. 21. 50 Boyd, Roll of the Drapers™ Company, p. 7.
Barriers to women™s employment 237
Table 5.4. The apprenticeship of girls

Years Town Percent female Source

1532“65 Bristol 4 a
1542“53 Bristol 3 b
1563“1713 Kingston upon Hull 0.4 a
1603“14 Salisbury 3.6 b
1600“45 Bristol 2.2 c
1710“31 Surrey 5.2 d
1710“52 Sussex 3.2 d
1710“60 Warwickshire 3.6 d
1710“60 Wiltshire 7.4 d
1711“20 Bedfordshire 5.0 d
1710“60 Sussex 2.9 e
1710“60 Warwickshire 3.1 e
1710“60 Wiltshire 6.9 e

a. Roberts, “Words They Are Women.”
b. Wright, “Churmaids, Huswyfes and Hucksters.”
c. Ben-Amos, “Women Apprentices.”
d. Snell, Annals of the Labouring Poor, Table 6.4.
e. Lane, Apprenticeship in England, p. 40.

observed that it was not the general practice, Stephen Smith, a
Gloucestershire weaver, claimed that girls were often apprenticed as
weavers. He noted a current apprentice, Rachael Smith, who lived near
him.51 Female apprentices were taught the trade; if an apprentice was
employed only in household work, the contract could be broken. From
1715 court records we learn of “Sarah Gibson discharged from her
apprenticeship to Joanna Worthington of St. Andrew™s Holborn widow,
mantua-maker [dressmaker], upon proof that the said Sarah, instead of
learning the trade of mantua-maker had been employed in common
household work.”52
However, while girls clearly could be apprenticed, few actually were. If
we focus on private apprenticeship, as opposed to parish apprenticeship,
only about 4 percent of apprentices were girls. Simonton ¬nds that
4 percent of private apprentices were girls in the late eighteenth century,53
and the other estimates presented in Table 5.4 suggest similar rates,

BPP 1806 (268) III, p. 346.
George, London Life in the Eighteenth Century, pp. 234, 418. See also Snell, Annals of the
Laboring Poor.
Simonton, “Apprenticeship,” p. 245.
238 Gender, Work and Wages in Industrial Revolution Britain

with some variation. While some guilds may have discouraged women
from entering, the low number of female apprentices was probably
more the result of family decisions than guild rules. Parents deciding
how to invest their scarce resources were more likely to apprentice
sons. Partially this re¬‚ects the expected return of this human capital
investment. Daughters would bring lower returns to the investment
over their lifetime because they spent less time in the labor force.
Partially, however, this decision also re¬‚ects gender discrimination. To
the extent that discrimination worked to lower women™s opportunities
here, it was discrimination that operated through the family.
More girls were bound as parish apprentices; about a third of parish
apprentices were girls.54 However, these apprenticeships did not provide
a route to higher-paying occupations because fewer parish apprentices
were bound to higher-paying trades. Training was not the only goal of
parish apprenticeship; the system also served to provide support for
children dependent on the parish.55 Because overseers were reluctant to
pay high premiums out of the poor rate to bind parish apprentices, they
were rarely bound to trades with good earnings prospects. Some parishes
set an upper limit on the premium that could be paid for a parish
apprentice.56 Female parish apprentices were often apprenticed to
“housewifery.”57 Thus the larger percentage of girls among the parish
apprenticeship does not mean that large numbers of girls were being
trained in skills valued in the labor market.
Even if they wished to, guilds did not always have the power to
exclude women workers. English guilds were weaker than continental
guilds in the pre-industrial era. Hutton notes that in the fourteenth
century, “Gild and civic regulations were not necessarily strictly
obeyed . . . The most one can say about such regulations is that they rep-
resent a situation which the civic and craft elite would have liked to bring
into existence.”58 Ogilvie suggests that English guilds began to decline in
the sixteenth century, and notes that in the West Riding of Yorkshire “rural
people took up ¬rst woollen weaving, then worsted weaving, without
lengthy (and often without any) apprenticeship . . . when guilds did
manage to secure apprenticeship legislation, it was widely ignored.”59

Snell ¬nds that in the eighteenth century 34 percent of parish apprentices were female,
while in the nineteenth century 31 percent were female. Snell, Annals of the Laboring
Poor, Table 6.1. See also Hindle, “˜Waste™ Children?,” p. 34.
Hindle, “˜Waste™ Children?” 56 Lane, Apprenticeship in England, p. 25.
Hindle, “˜Waste™ Children?,” p. 37.
Diane Hutton, “Women in Fourteenth Century Shrewsbury,” in Lindsay Charles and
Lorna Duf¬n, eds., Women and Work in Pre-Industrial England (London: Croom Helm,
1985), p. 83“4.
Ogilvie, A Bitter Living, p. 96, and “Guilds, Ef¬ciency, and Social Capital,” p. 303.
Barriers to women™s employment 239

In 1702 only about half the weavers in Taunton had served an
What little power guilds did have was lost during the Industrial
Revolution. Certainly by the nineteenth century English guilds had little
power and apprenticeship had ceased to be an important barrier to entry
into a trade. The Statute of Arti¬cers, which forbade non-apprenticed
workers, was repealed in 1814, but the institution of apprenticeship had
fallen into decline before this. Where apprenticeship was not necessary
to learn the skills, non-apprenticed workers often worked in the trade. In
weaving, apprenticeship was not necessary; in 1803 only 13 percent of
the weavers employed by a Gloucester clothier had served an appren-
ticeship.61 Richard Fawcett, a Bradford manufacturer, noted of weavers
“I believe nineteen out of twenty have not served regular ap-
prenticeships.”62 In Leeds, only apprenticed clothiers were allowed to
sell cloth in the two main cloth halls, but there was a third hall, known as
Tom Paine Hall, where anyone could sell cloth, and many persons not
apprenticed to the trade sold cloth there.63 A Leeds clothier noted that
in this third hall, “any persons who have not served an apprenticeship go
and shew a coloured piece there.”64 Under the pressure of this com-
petition, the white cloth hall agreed in 1803 to accept cloth from non-
apprenticed persons.65 One indicator of the declining power of guilds
was the declining portion of youth who were apprenticed. Simonton
notes that “apprentices were clearly a declining proportion from 1771,
with a fairly steep drop from 1786.”66 Where apprenticeship was not
truly necessary to learn the skill, guilds found themselves unable to
restrict the trade to apprenticed workers.
English guilds were not an important source of occupational segre-
gation during the Industrial Revolution. Guilds allowed girls to be
apprenticed, so the fact that few girls were apprenticed re¬‚ects choices
made in the family rather than discriminatory barriers within the guild
system. The fact that fewer girls were apprenticed became increasingly
irrelevant as the guild system disintegrated. If apprenticeship was no
longer necessary for employment, girls could not be disadvantaged by
their lack of apprenticeship. During the Industrial Revolution, the
interests of employers and employees diverged, leading to the decline of
the guild system and the rise of unions, which were important in
excluding women from certain occupations.

Ibid. 61 Hammond and Hammond,The Skilled Labourer, p. 170.
BPP 1806 (268) III, p. 184.
See Adrian Randall, Before the Luddites (Cambridge: Cambridge University Press,
1991), p. 211, and BPP 1806 (268) III.
BPP 1806 (268) III, p. 10. 65 Ibid., p. 201. 66 Simonton, “Apprenticeship,” p. 238.
240 Gender, Work and Wages in Industrial Revolution Britain

D. Employers
While they are obvious candidates for excluding women from waged
work, employers were not an important source of occupational segre-
gation during the Industrial Revolution.67 Their actions were generally
the opposite: employers fought for the right to hire women. One mistake
that could lead us to blame employers for occupational segregation is
accepting the ideological statements of employers as accurate descrip-
tions of their actions. In assessing the role of employers in excluding
women, we must remember that statements of cultural ideals do not
always match actions. Employers™ statements that married women
should not work are not proof that such women were not hired. One
employer claimed to oppose the employment of married women at the
same time he admitted to employing them himself:

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 . . . but 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.68

Employers seemed to have no trouble saying one thing and doing
another. Struggling to survive in a competitive market, they could not
afford to indulge their personal preferences.
A brief look at labor history reveals that employers were not an
important source of occupational segregation. Most employers con-
sidered it their right to hire women if they could bene¬t by doing so, and
opposed restrictions on women™s employment. Employers who wished
to maximize pro¬ts wanted to employ women and were willing to hire
women if it meant increased pro¬ts. M™Connel and Kennedy, for
instance, found its male workforce troublesome and began to hire
women as mule spinners in 1810. The experiment did not work;
problems with increased wastage of raw material and high turnover
rates, both resulting from the fact that the women did not recruit and
discipline their own assistants like the men did, meant that the shift to
female workers did not increase the ¬rm™s pro¬ts, even though the
women were paid lower piece-rate wages. Huberman notes that:

An example of a historian who blames employers for occupational segregation is Jordan,
“The Exclusion of Women,” who claims that employers refused to hire women simply
out of “androcentric blindness.”
Frederick Carver of Nottingham, BPP 1876, XIX, p. 258, quoted in Rose, Limited
Livelihoods, p. 32.
Barriers to women™s employment 241

The ¬rm might have persevered with the policy if pro¬ts remained healthy for like
other ¬rms, it saw the long-term bene¬t of getting rid of recalcitrant male workers
and replacing them with women. However, after 1820 when the fall in margins
signaled a squeeze on pro¬ts, the ¬rm was compelled to ¬nd a way to reduce costs
or to raise productivity by some means other than the hiring of women.69
It was falling pro¬ts, not gender ideology, that caused M™Connel and
Kennedy to cease hiring female mule spinners. The willingness of the
employers to hire women when they thought they could bene¬t from
doing so indicates that it was productivity rather than gender which
prevented women from being spinners.
When it was in their economic interest to hire women, employers were
willing to ¬ght unions for the right to do so. Both James Dunlop and
William Kelly of Glasgow attempted to hire women as mule spinners,
but the women did not stay because of the violent reactions on the part
of the male workers.70 From 1810 to 1834, the journeymen and master
tailors of London fought over whether women could be employed.71 In
1861 employers attempted to hire women to work stocking frames, but
the male workers went on strike.72 Sonya Rose has studied many of these
worker“employer con¬‚icts in the later nineteenth century, and con-
cludes that “when it was possible to do so, employers attempted to hire
women in place of men.”73 As we shall see when examining unions,
employers were more likely to ¬ght for the right to hire women than to
exclude them from employment.
While most theories of employer discrimination assume that gender
roles motivated constraints, a somewhat different theory suggests that
occupational constraints were based on the need to prevent contact
between the sexes that could result in sexual misconduct. Humphries
claims that concerns about sexual behavior motivated occupational
segregation.74 Concern about sexual propriety, however, seems to have
motivated action in the upper classes (and particularly in parliamentary
committees), more often than among the laboring or employer classes.75
Employers showed by their actions that they were willing to mix the
sexes. Men and women often worked side by side. In some regions
much agricultural work was done by gangs, which included workers of

Huberman, Escape from the Market, pp. 28“9.
BPP 1824 (51) V, p. 525, and Aspinall, The Early English Trade Unions, p. 390.
Taylor, Eve and the New Jerusalem, ch. 4. 72 Rose, “Gender Segregation,” p. 171.
Rose, “Gender Antagonism,” p. 195. 74 Humphries, “Most Free from Objection.”
Humphries claims that upper-class concerns about morality extended down into the
working classes. However, to establish this point she uses statements from workers
about what they thought was proper rather than evidence of actual work patterns. If we
look at work patterns, revealed preference tells us that the lower classes were willing to
allow men and women to work together.
242 Gender, Work and Wages in Industrial Revolution Britain

both sexes.76 Farmers were not averse to allowing men and women to
work together, as indicated by the adage advising mixed groups of
workers: “One man among women, one woman among men.”77 Men
and women often worked in the same location even when they were
occupationally segregated. Young women were piecers but were never
mule spinners; concern about morality cannot explain this occupational
difference because spinners and piecers necessarily worked side by side.
Even if employers had wished to separate the sexes, they could have
established different factories for each sex. James Dunlop of Glasgow,
for example, installed smaller mules in his factory so that he could hire
women exclusively.
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 . . . of such reduced size as could easily be wrought by women. They
employed women alone . . . These they paid the same [piece] rate of wages, as
were paid at other works to men.78

Even the desire for a segregated work environment would not prevent
both men and women from working in the same occupation.
Economic theory tells us that discriminatory employers who refuse to
hire women on ideological grounds are less ef¬cient and can only suc-
ceed in monopolistic markets. Becker™s model of discrimination predicts
that in a competitive market employers who discriminate against women
should fail.79 Only monopolistic employers should be able to discrim-
inate against women. If Becker is correct, we should observe that
employers in competitive markets hire women willingly, while only
employers in monopolistic markets are able to exclude women. Davidoff
and Hall point to the chocolate manufacturer George Cadbury, who in
the 1870s “was strongly opposed to the employment of married women
and refused to have them working at Bournville.”80 George Cadbury
could afford to indulge his gender ideology, and refuse to hire married
women, because his brand gave him a certain amount of monopoly

Pinchbeck, Women Workers, p. 86. 77 Kussmaul, Servants in Husbandry, p. 47.
BPP 1824 (51) V, p. 525. Men and women were paid the same piece-rate wages, so
they were treated equally but may have had different earnings. Dunlop eventually
stopped hiring women because of the violent attacks of the male union: “But they [the
women] were waylaid and attacked, in going to, and returning from their work; the
houses in which they resided, were broken open at 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.”
Becker, The Economics of Discrimination, ch. 3.
Davidoff and Hall, Family Fortunes, p. 58.
Barriers to women™s employment 243

power. Employers in more competitive markets may have shared
Cadbury™s ideology, but we observe them ¬ghting for the right to hire
While employers may have been the source of discrimination in a few
cases, employer discrimination cannot explain widespread occupational
segregation. The importance of employers in occupational segregation
has been overstated both because a few examples are taken as repre-
sentative of the whole and because statements of employers are too
readily accepted as statements of fact. Looking at their actions, we ¬nd
that employers allowed men and women to work together and were even
willing to ¬ght for the right to hire women.

E. Unions
The nineteenth century saw the emergence and growth of trade
unions.81 As we have seen, guilds organized both employers and
employees into one trade organization. With the Industrial Revolution,
however, it became harder for a journeyman to move from being a wage-
earner to being a self-employed producer. As the potential threat of a
new worker as a competitor became more remote, the employer lost the
incentive to limit the number of wage-earners in the trade, but the wage-
earners still bene¬ted from limiting entry to the occupation, and they
organized unions and threatened their employers with strikes in order to
do so. Employers were not an important force in limiting women™s
opportunities because their economic interests con¬‚icted with, and
usually overruled, their ideas about gender roles. Because their gender
ideology coincided with their economic interest, male workers became
the most important force excluding women from wage-earning occu-
pations. When unions were successful, they were able to enforce con-
straints on women™s employment, and occupational segregation thrived
because the market mechanism was overruled.
Though they were illegal until 1824, unions existed early in the
Industrial Revolution. “Conspiracies in restraint of trade” were illegal
under common law, and the Combination Acts of 1799 and 1800
speci¬cally outlawed collective action. The law gave the masters more
power to prosecute, but it certainly did not prevent unions from forming
and acting. The law was only partially effective because employers did

By “union” I mean any group of employees working together to promote their own
interests. Women theoretically could be members of such groups, but in practice were
not. Unions of the Industrial Revolution period did not have the same formal structures
as unions today because the law was different; union activity was illegal until 1824.
244 Gender, Work and Wages in Industrial Revolution Britain

not always ¬nd it in their interests to enforce the law, and when they did
it was a slow and costly process. Unions existed and exerted in¬‚uence
even when they were illegal. The Combination Acts must have had some
effect on reducing union activity, though, since there was a surge of
union activity after they were repealed. After 1824 unions grew in
number and importance. The 1830s saw two failed attempts at organ-
izing a general union.82 The Webbs identi¬ed three periods of union
expansion: 1833“4, 1873“4, and 1889“90.83 By the later half of the
nineteenth century, union power was ¬rmly established in Britain.
While guilds used apprenticeship rules to maintain their incomes,
unions relied on other rules to limit the supply of workers and thus
maintain high wages. When they were able, unions enforced closed
shops. A Bradford magistrate noted in 1802 that “the shearmen will not
suffer any man to work who has not got a ticket,” the ticket being proof
of membership in the Shearmen™s Club.84 Unions also limited the
number of new workers entering the trade. The most powerful unions
admitted only relatives of current members. The Lancashire mule
spinners only allowed sons, brothers, or orphaned nephews of current
members to enter the trade.85 The Dublin carpenters also allowed only
sons, brothers, and nephews of current members to be apprenticed.86 In
these cases unions enforced rules that were more restrictive than the
guilds™ apprenticeship rules had been.
Gender ideology and economic incentives played a joint role in
leading unions to demand the exclusion of women from their trades.87
Unions desired to limit access to their trades, in order to reduce labor
supply and increase their wage. Gender ideology made women a natural
target.88 Without either one of these factors things would have turned
out differently. Without gender ideology, women would not have been
an easily identi¬able group, and exclusion would have been based on

The National Association for the Protection of Labor, and the Grand National
Consolidated Trades Union. See G. D. H. Cole, Attempts at General Union: A Study in
British Trade Union History, 1818“1834 (London: Macmillan, 1953).
Webb and Webb, The History of Trade Unionism, p. 314.
Aspinall, The Early English Trade Unions, p. 50.
Rose, Limited Livelihoods, p. 143. 86 BPP 1824 (51) V, p. 430.
Similarly, racist ideology and economic incentives played a joint role in creating the
system of apartheid in South Africa. Frederick Johnstone, Class, Race and Gold: A Study
of Class Relations and Racial Discrimination in South Africa (London: Routledge and
Kegan Paul, 1976), emphasizes the role that white workers played in instituting the
color bars that prevented non-white workers from entering skilled occupations. White
workers worked to institute racial discrimination because they bene¬ted economically.
On gender ideology, see Davidoff and Hall, Family Fortunes; Hartmann, “Capitalism,
Patriarchy, and Job Segregation”; Rose, “Gender at Work”; and Scott and Tilly,
“Women™s Work and the Family.”
Barriers to women™s employment 245

other factors such as nationality or family connection. Without an
economic incentive driving the exclusion, it is doubtful that unions
would have pushed hard to exclude women. Employers had the same
gender ideology, but they were willing to hire women anyway because
they bene¬ted economically from doing so.
Thus, unions of the Industrial Revolution did everything they could to
exclude women from employment. The Bookbinders™ Trade Society
excluded women in 1810.89 Barbara Drake reports that women were
initially allowed to be members of the Manchester Spinners™ Union, but
were excluded after 1818.90 In 1820 the Glasgow mule spinners went on
strike, demanding an end to the employment of women. One employer
received an anonymous letter threatening:
I am authorized to intimate [the] jeoperdy and hazardious prediciment you stand
in at the present time . . . by keeping them weomen of¬ciating in mens places as
cotton spinners, and plenty of men going idle out of employ.91

Spital¬elds silk weavers combined to restrict women to the cheaper
work. In 1769 they succeeded in getting the masters to agree to a book of
prices stipulating, “No woman or girl to be employed in making any
kind of work except such works as are herein ¬xed and settled at 5d. per
ell . . . And no woman or girl is to be employed in making any sort of
handkerchief of above the usual or settled price of 4s.6d. per dozen.”92
The Stockport Hatmakers™ Society excluded women in 1808.93 Their
rules included an agreement to strike (“knock off”) against women
workers: “And it is unanimously agreed that all women are to be
knocked off against, to knock one woman off at one shop at a time, till it
is gone round the trade, and so on till they are all done away with.”94
Other historians have noted the fact that male workers fought for

Felicity Hunt, “Opportunities Lost and Gained: Mechanization and Women™s Work in
the London Bookbinding and Printing Trades,” in Angela John, ed., Unequal
Opportunities (Oxford: Basil Blackwell, 1986), p. 74.
“The Manchester Spinners™ and the Manchester Small Ware Weavers™ Societies are
known to have had women members in the 18th century. During the spinners™ strike of
1818, men and women drew equal strike pay; but, owing it would seem to their failure
to observe trade union conditions, the women were afterwards excluded.” Drake,
Women in Trade Unions, p. 4.
BPP 1824 (51) V, p. 531.
A List of Prices in those Branches of the Weaving Manufactory called the Black Branch, and the
Fancy Branch . . . , 1769, quoted in George, London Life in the Eighteenth Century, p. 182.
Maxine Berg, “Women™s Work, Mechanisation and the Early Phases of Industrialisation
in England,” in Patrick Joyce, ed., The Historical Meanings of Work (Cambridge:
Cambridge University Press, 1987), pp. 64“98.
Aspinall, The Early English Trade Unions, p. 107. Rules agreed to, September 19, 1808.
Here they agree to use a rolling strike, that is, to strike against only one master at a time.
246 Gender, Work and Wages in Industrial Revolution Britain

restrictions on female employment that bene¬ted them economically.
Rose emphasizes the “active role played by skilled male workers in cre-
ating gender segregation as they attempted to preserve their own jobs.”95
Male workers used gender ideology to argue that a woman™s place was
in the home, and that the male wage should be high enough to support
his wife and children, that men should earn a “family wage.”96 They
justi¬ed the exclusion of female workers as a method to obtain a family
wage. Humphries suggests that working-class women as well as men
bene¬ted from such restrictions because they increased male wages
enough to keep total family earnings constant, while women gained
leisure.97 It is theoretically possible that a removal of women from labor
supply could raise wages enough to keep the wage bill constant, if the
labor demand curve was unit elastic, but Humphries provides no evi-
dence that this was so.98 More importantly, even if family income did
not decline, women lost bargaining power within the household. Con-
cern about the gender gap arises not just because low earnings leave
women poor, but also because the difference between male and female
earnings ensures that men have more power than women. Even if total
family income did not change, a shift in the composition of income that
increased male earnings and reduced or eliminated female earnings
would have increased the gender gap and made women more dependent
on men. Hartmann suggests that men used occupational segregation
not only to maintain their wages, but also to maintain their patriarchal
power within the home.99 Men wanted to keep their jobs, and to con-
tinue to enjoy the bene¬ts of women™s domestic labor, so they excluded
women from skilled occupations.
Deborah Simonton claims that gender, not economics, was the pri-
mary reason for unions™ rules against women, and she supports this
claim by arguing that if their motivation had been primarily economic,
then unions would have excluded other men as well.
On the one hand, it was not about women, but about protecting their craft
position and independence vis-a-vis capitalists who wished to control labor

Rose, “Gender at Work,” p. 120.
Seccombe, “Patriarchy Stabilized”; Land, “The Family Wage.”
Jane Humphries, “Class Struggle and the Persistence of the Working-Class Family,”
Cambridge Journal of Economics 1 (1977), p. 251.
Humphries uses Marx™s theory of wage determination and opinions of the workers
themselves to support her claim that total family income would remain constant.
Neither is convincing evidence that labor demand was elastic or unit elastic. Modern
studies sometimes ¬nd that labor demand is elastic, but most estimates suggest inelastic
demand. Daniel Hammermesh, Labor Demand (Princeton: Princeton University Press,
1993), pp. 78“9.
Hartmann, “Capitalism, Patriarchy, and Job Segregation.”
Barriers to women™s employment 247

supply. On the other hand, it clearly was about gender, because the workers
against whom it was aimed were women, since they perceived female labor, not
unskilled male labor, as the threat.100

However, it is not true that exclusion was aimed only at women. While
women were an identi¬able group likely to be targeted, they were not the
only group excluded. At the extreme, mule spinners allowed only male
relatives of current workmen to enter the trade. In this case most men
experienced the same exclusion as women. Alexander Erskine, a worker
at a Glasgow cotton mill, gave evidence that:
he would wish to be a spinner, and earn the same high wages as they do, but he
was not brought up a spinner, and the spinners would not therefore allow him to
learn, for the spinners have an association to prevent this, by refusing to teach
anybody to spin who has not been brought up with them.101

Clearly Alexander was excluded from employment as effectively as any
female. Restrictions on employment prevented workers in dying trades
such as handloom weaving, men as well as women, from entering other
trades. In 1841 a parliamentary investigator noted:
I am perfectly convinced that the distress of the hand loom weavers is mainly and
almost entirely to be ascribed to the exclusive monopoly established by the
forcible conduct of the trades in all other lines, which prevents their sons getting
into any other line . . . every trade is fenced round by prohibitions, which render
it impossible for a person to get into it, except a son or a brother, or some near
relation of an already existing member.102

Other groups besides women were also targeted for exclusion. In 1812 the
wool-combers™ union agreed not to admit Irishmen.103 The fact that
women were occasionally the insiders also supports the claim that gender
was not the only line dividing insiders from outsiders. In powerloom
weaving in Glasgow, women operated their own exclusionary society.
Mary Donald, age 11, gave evidence in 1833 that she was not a weaver,
since “Her mother wants to get her into the mill, but can™t afford to pay
10s.6d. to the association.”104 This girl found her employment oppor-
tunities limited, not by men, but by other women.
Not every combination sought to exclude women. On one occasion,
journeymen pipe-makers argued for maintaining women™s jobs. They
asked the public to avoid “marked Pipes” because the new type of pipe

Simonton, European Women™s Work, p. 172. 101 BPP 1833 (450) XX, A1, p. 81.
BPP 1841 (296) X, p. 108.
The rules agreed to in 1812 include this statement: “No Irishmen to be admitted to
society, after the date of these Articles.” Aspinall, The Early English Trade Unions, p. 135.
BPP 1833 (450) XX, A2, p. 54.
248 Gender, Work and Wages in Industrial Revolution Britain

resulted in unemployment for women: “a great many Women are in
Want of Business that were always brought up to it; for by the aforesaid
Pipes, two Women can do as much Work as would require four.”105 In
this case, women were not the threat to employment; new pipes were.
Since the interest of the journeymen coincided with that of women
workers, they promoted the employment of women. However, when
they found it in their interest to do so “ when doing so would decrease
competition and maintain their high wages “ male unions attempted to
exclude women workers.
In most cases the ¬erce resistance of employers to rules against hiring
women suggests that these rules were binding. The presence of a strong
union was important in determining whether women worked in a trade,
as can be observed in a few trades where unions had differing success in
different cities. After the self-acting mule was invented in 1833, strength
was no longer necessary and women were able to operate mules pro-
ductively.106 In Glasgow, where the union was not effective, women
sometimes worked as mule spinners.107 In Lancashire, however, the
mule spinners maintained an effective union, and no women worked
mules there. Similarly, in the later nineteenth century, Edinburgh had
numerous women compositors but England had few. Bradley and Black
explain this by pointing to the difference in the strength of the com-
positors™ unions:
Why are women employed so largely in Edinburgh, so little in England? Factory
law is the same for the whole of Britain, so the cause must be sought elsewhere.
The answer would seem to lie in the attitude adopted by men™s Trade Unions.
The Union in Edinburgh has never recovered from the blow dealt it in 1872“3,
and is not now in a position to make any effective stand against the inroads of the
army of women compositors. In London and Lancashire, on the contrary, the
unions are strong.108
The large numbers of women compositors in Edinburgh are explained
by the fact that the Edinburgh union lost a strike in 1872“3 and had not
recovered the strength necessary to prevent the employment of women.
Thus, the presence of an effective union does seem to be a constraining
We have seen that neither government regulations, guilds, nor employers
explain the limits to women™s employment. Government regulations

An advertisement from 1745, quoted by Dobson, Masters and Journeymen, p. 41.
See below.
In particular, James Dunlop and Mr. Crombie employed women. BPP 1824 (51) V,
pp. 615, 618.
Barbara Bradley and Anne Black, “Women Compositors and the Factory Acts,”
Economic Journal 9 (1899), p. 264.
Barriers to women™s employment 249

limiting female employment did not appear until the 1840s, and limits
on hours do not seem to have reduced female factory employment.
Unlike German guilds, which severely limited women™s opportunities,
English guilds did not have suf¬cient power to exclude women
workers.109 Employers were not an important source of occupational
constraints because they were eager to hire women when they could
pro¬t from doing so. I conclude that the main source of employment
barriers for wage-earning women was male unions. Male unions
bene¬ted economically from excluding women from their occupations,
and did so whenever they could. In most cases gender ideology coin-
cided with unions™ economic interests, making it dif¬cult to determine
which force was driving the desire to exclude women. I believe that
economic interest was the primary force because employers, who
shared similar gender ideology but had opposing economic interests,
chose to side with their economic interests. Gender ideology was a
convenient rhetorical device for unions, but their primary motivation
was economic.
If unions were the cause of the exclusion of women from certain
trades, they needed not only the desire to exclude women, but also the
ability to enforce this desire. Why were some unions successful while
others were not? Jordan does not accept the union explanation for the
exclusion of women because “only a few of the strongest craft unions
had suf¬cient power to impose their wishes on employers.”110 She is
right that we should be skeptical about the ability of a union to impose
its will on the employer. Not all unions were successful in imposing
employment restrictions. As we shall see in the next section, the amount
of power a union possessed depended on the skill level of the job; unions
in skilled trades were more successful. This difference in union power,
then, provides the explanation of why we observe occupation segrega-
tion in more highly skilled trades, but not in low-skilled trades.

III. Where unions could be successful
The previous section concluded that unions were the most important
cause of occupational segregation constraints. This section will show
that unions had the economic power to enforce their demands only in
skilled occupations, thus providing an explanation of why there were

On restrictions imposed by German guilds, see Ogilvie, A Bitter Living.
Jordan, “The Exclusion of Women,” p. 286. Where unions were weak, Jordan ascribes
occupational segregation to employers, while I ascribe it to sorting based on strength,
as described in Chapter 3.
250 Gender, Work and Wages in Industrial Revolution Britain

occupational segregation constraints in skilled occupations but not in
unskilled occupations. Where unions were not strong enough to erect
barriers to women™s employment, the division of labor was determined
by comparative advantage, as described in Chapter 3.
Both contemporaries and historians have noted that in Industrial
Revolution Britain there was a clear distinction between skilled and
unskilled occupations in terms of union success. In 1836 Andrew Ure
noted that “it is, moreover, a well established fact, that those artisans
who are the worst paid seldom combine, and never with any force; but
only those who enjoy the best wages, such as cotton spinners, engin-
eering mechanics, founders . . . &c.”111 More recently, Rule noted that
skilled laborers had the power to win disputes, but that “this power
depended upon the defense of the skill, both against deskilling innov-
ation and as a frontier against the unskilled, including large numbers of
women workers.”112 To establish that the success of unions depended
on skill, I will present both a model explaining why skill should matter
and examples from a number of different unions showing that skilled
workers were in fact more successful.
The classic model of union bargaining was presented by Hicks in
1932.113 Figure 5.1 shows the essence of the model. Hicks predicted
that the outcome of wage bargaining would depend on the “employer™s
concession curve” and the “union™s resistance curve,” both of which
map wage offers against length of strike. The employer™s concession
curve gives the wage the employer would be willing to pay to prevent a
strike of a given length. The union™s resistance curve gives the length of
time the workers would be willing to strike to obtain a given wage. When
bargaining, each party will foresee the outcome, and thus both will agree
on w*, the wage where the two curves meet. In this model, strikes occur
only if the decision-makers incorrectly predict the curves, leading each
side to different estimates of the curves and thus different estimates of w*.
Neither the workers nor the employer can hold out forever. The
workers have to eat, and when their assets and credit are exhausted they
will be forced to return to work. The employer has ¬xed costs to pay,
and cannot endure a strike inde¬nitely because he is constrained by the
possibility of bankruptcy. Anything that allows the striking workers to
hold out longer, such as a large strike fund, would shift the union™s

Andrew Ure, The Cotton Manufacture of Great Britain (London: Charles Knight, 1836),
vol. I, p. xxv.
John Rule, “The Formative Years of British Trade Unionism: An Overview,” in John
Rule, ed., British Trade Unionism, 1750“1850: The Formative Years (New York:
Longman, 1988), p. 6.
J. R. Hicks, The Theory of Wages (New York: Peter Smith, [1932] 1948), ch. 7.
Barriers to women™s employment 251

Employer™s Concession Curve


Union™s Resistance Curve

Expected Length of Strike
Figure 5.1 Hicks™s bargaining model

resistance curve to the right and increase the wage the union can secure.
A strike may not be necessary to win the higher wage, because the
employer will foresee a longer strike and will be willing to give a higher
wage to avoid the strike.
Hicks drew some speci¬c conclusions about what factors would
in¬‚uence the position of the two curves. The union™s resistance depends
on its ability to support its members:
The actual duration of resistance depends on ability as much as on willingness.
Strikers™ ability to hold out depends, in its turn, partly on the size of the union™s
accumulated funds (the amount of strike pay it can give), partly on the savings of
the members (which enable them to be content with a low rate of strike pay, or to
hold out when strike pay has disappeared), partly on the attitude toward the
strike of parties not directly concerned (the willingness of shopkeepers to give
credit, the willingness of other unions or independent well-wishers to give loans
or donations to the union). The greater the extent of such resources, the stronger
the union will be.114

In order to win a strike, a union must be able to build up a strike fund
large enough to support itself. Unskilled workers, because they were so
poorly paid, had dif¬culty building up large strike funds and lost many
strikes when their funds ran out. Skilled workers had an advantage, in
the form of a higher resistance curve, because they were more highly
paid, and could more easily save money and build up strike funds.

Ibid., pp. 153“4.
252 Gender, Work and Wages in Industrial Revolution Britain

Occasionally a strike was won on the charity of the townspeople, who
offered loans to the strikers, but dependence on others could not provide
the basis of consistent power.
Skilled unions also found it easier to maintain discipline. While the
group as a whole bene¬ts from maintaining the strike, an individual
worker could do better by working. To win a strike, a union must pre-
vent its members from agreeing to go back to work. As Mancur Olson
points out, this is more dif¬cult in large groups because social pressure is
less effective. The larger the group, the greater the chance that free-
riding will prevent any action.115 Skilled occupations have the advantage
because the number of workers is usually smaller. Also, skilled occu-
pations can impose discipline by threatening to prevent a person from
working in the trade. This was a real threat because being able to work in
the skilled trade was valuable; the alternative was unskilled work, at a
much lower wage.
Anything that makes the strike more costly to ¬rm pro¬ts will shift the
employer™s concession curve to the left, and increase the wage that the
union can secure. Hicks noted that the employer™s concession curve
depends on “the degree to which the union can make the strike effective
in causing a stoppage of the employer™s business” and on the costs of the
stoppage to the ¬rm.116 If the business can continue to operate in spite
of the strike, the employer can hold out longer. Here again skilled
workers had the advantage over unskilled workers. Strikes were more
costly to the ¬rm if the business of the ¬rm had to be suspended during
the strike. Unskilled workers could be replaced, and the business could
continue to operate with minimal loss of pro¬t. Skilled workers, how-
ever, were dif¬cult to replace, and a strike of skilled workers might stop
production. The employer needed to recruit skilled workers who were
not in the union or train new workers. If the work was highly skilled,
training took much too long and could not be used to win a strike. A
union is the strongest if all workers with a particular skill are part of the
union. If the union has a monopoly on the skill and there is no good
substitute for the skill, that union will be able to extract monopoly wages
because the employer cannot hope to train strike-breakers at all. One of
the important sources of power for these unions was the fact that they
controlled training. If the union could control the skill “ determining
who could acquire it and who could not “ it could control employment.
Since the method of training for most skilled wage-earners was

Mancur Olson, The Logic of Collective Action (Cambridge, Mass.: Harvard University
Press, 1965).
Hicks, The Theory of Wages, pp. 154“5.
Barriers to women™s employment 253

apprenticeship with a current worker, skilled unions had effective power
over who was trained and thus who would work.
Sometimes unions with insuf¬cient economic power to win their
demands tried to improve their bargaining position by using violence.
This was the one important factor in which skilled trades did not have
the advantage. By using violence, a union could win a strike even if it did
not have economic power. James Dunlop of Glasgow let go his women
spinners because the protests of the male union were so violent. His
women spinners:
were waylaid and attacked, in going to, and returning from their work; the
houses in which they resided, were broken open in the night. The women
themselves were cruelly beaten and abused; 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.117

In 1824 John Martineau testi¬ed before a parliamentary committee that
“Last Monday, while I was at Liverpool, a man was murdered in the
streets, for having refused to join a combination.”118 Violence could be
very important in determining the outcome in a particular situation.
Thus, predictions of success based on whether the trade was skilled will
predict the outcome only with a certain amount of error.
Violence, however, was rarely a source of consistent power, and its use
was generally the sign of a weak union. The most successful unions did
not resort to violence. If the union could obtain its demands using
economic pressure only, there was no need for violence. The unskilled


. 9
( 14)