THE COURSE OF WAGE DIFFERENTIALS BETWEEN SKILLED AND NONSKILLED MANUAL WORKERS IN BRITAIN BETWEEN 1856 AND 1964
| DOI | http://doi.org/10.1111/j.1467-8543.1983.tb00121.x |
| Author | Roger Penn |
| Date | 01 March 1983 |
| Published date | 01 March 1983 |
THE COURSE
OF
WAGE DIFFERENTIALS BETWEEN SKILLED
AND NONSKILLED MANUAL WORKERS IN BRITAIN
BETWEEN
1856
AND
1964
ROGER PENN*
THIS
article investigates empirical material on wage differentials between skilled and
nonskilled manual workers during the period between
1856
and 1964. The evidence
was gathered
in
Rochdale, Lancashire, as part of a wider project that analysed the
changing class situation of skilled manual workers (See Penn 1981; forthcoming). The
data pertains to the engineering and cotton industries that have provided the major
sectors of employment
in
the town during the period under review.
TABLE
I
Numbers
of
Workers (Male and Female) Employed in the Cotton and Engineering Industries
in
Rochdale,
186
1-
196
1
1861
l8Yl
I921
lY61
Employed
in
Engineering 930
1.461
5.770
5,060
Population 38.114
76,161
90.816
85.787
Employed
in
Cotton
1
I
.584
14.31
1
2
1.656
9.390
Source:
The
Decennial
Censuses,
1861-IY61
(see Chapter
5
of
Penn 1981. for a
full
discussion
of
the
generation
of
the tables).
A) THE NATURE
OF
THE
EVIDENCE
The materials on differentials used in this article have been presented in their
original form in two appendices of my doctoral dissertation. Before proceeding with
their analysis, it is necessary to describe their features in some detail. The sources for
information
on
wages in the engineering and cotton industries are very similar prior to
1914. The figures for 1859 and 1877 on engineering derive from the ‘Returns of Wages
1830-1886’ (see Labour Statistics), which were assembled to provide an historical
background to the first national ‘Wages Census’ in 1886. In both years the evidence
relates to ‘Manchester and Neighbourhood’ and gives the wage rate per week. Both
features present problems of interpretation for this analysis. Firstly, it is not clear
whether the area, ‘Manchester and Neighbourhood’, included Rochdale. However,
even
if
it did not, it is unlikely that the rates would differ very much from those
presented, since Rochdale was certainly within the close economic hinterland
of
Manchester. Any difference would probably be in the order of a few pence per week
and would not significantly affect the overall averages constructed. Information in the
form
of
a weekly wage rate presents the first major conceptual difficulty in the analysis
of
differentials, since the weekly wage rate does
not
necessarily correspond to weekly
earnings, because
it
does not include extra payments like bonuses and overtime.
However, this difficulty, which as will be seen, has severely handicapped many recent
attempts to examine the course of twentieth century differentials in engineering, is not
a major problem
in
the nineteenth century engineering industry, since as Hart
&
McKay (1975, p.
33)
have pointed out, ‘wage rates accounted for the bulk of earnings
up to 1914’.
*Department
of
Sociology, University
of
Lancaster.
69
70
BRITISH JOURNAL
OF
INDUSTRIAL RELATIONS
The evidence for the cotton industry in
1859
and
1877
also derives from the ‘Returns
of Wages,
1830-1886’
(see Labour Statistics), which themselves were transcripts from
previous Government statistics on wages (see Miscellaneous Statistics Pt
I1
&
X).
However, there is a divergence between the
1887
compendium volume and Bowley’s
(1900)
influential subsequent analysis of wages in the cotton industry which purports
to have been derived from the same original sources. It would appear as if Bowley
introduced data from hand-mule spinning in the construction
of
his overall average for
spinners on fine counts and also evidence for ‘little’ and ‘big’ piecers which are not
included
in
the statistical sources which he cites.
The spinners’ average in
1859
in Bowley’s text appears to include as a component
spinners
on
fine counts which are, in fact, hand-mule spinners. They vary from
23s
Od
to
45s
Od per week. Piecers in
1859
earned between
6s
Od and
10s
Od per week which
fails to explain Bowley’s estimate
of
6s
Od for small piecers. Indeed, the distinction
between ‘big’ and ‘little’ piecers was not made in the returns.
Nor
was it made in the
1879
returns which do not provide any data on hand-mule piecers. The earnings
of
hand-mule spinners ranged from between
40s
Od and
55s
Od per week, and again seem
to have featured in Bowley’s estimates of spinners’ overall average earnings.
I
have followed the
1887
compendium data, as does Wood
(1910)
in his own analysis
of wages in the spinning sector, in the construction of the tables presented in this
article.
The evidence for
1886
for both industries derives from the first comprehensive wage
census conducted in that year (see Wage Census
1886)
and presents data on average
weekly wage rates. The material on cotton is presented for ‘Rochdale, Heywood and
Neighbourhood’ and that on engineering for ‘South Lancashire (excluding
Manchester)’. Since the material gives the precise number employed
I
have used the
weighted
averages in the construction of the tables in this article. This procedure was
followed wherever possible in this research. The data for
1906
comes from the Wage
Census of that same year which was the most comprehensive ever taken nationally. It
gives information for both the cotton and engineering industries on numbers
employed, forms of payment (piece-rate versus time-rate) and average earnings.
In the period from
1906
onward, the evidence for the two industries becomes
separate and the major sources become the respective employers’ organisations. The
material for the engineering industry has been collected by the Engineering
Employers’ Federation and relates to the Rochdale district
of
the Federation. The
evidence is collected for the purposes
of
wage negotiations and is regarded as very
accurate (see Hart
&
Mackay
1975).
Evidence was collected for the years
1914, 1918,
1925, 1930, 1935, 1955
and
1964;
which gives, overall, a series of eleven points
in
time
from which to construct the picture of wage differentials within the engineering
industry.
Unfortunately, the material
on
cotton is not
so
rich. A major reason for the gaps
was the
loss
of many records held by the Textile Employers’ Federation in wartime
air-raids, including the inter-war wage inquiries. However, some useful information,
itself derived from the destroyed records concerning the
1930s,
is contained in Jewkes
&
Gray
(1935)
and Gray
(1937).
The difficulties
in
assembling adequate
post-1945
evidence lie in the general lack of relevant data which is partly the result of the
dramatic decline
of
the industry and the subsequent demoralisation
of
industrial
relations. There is some material in the Evershed Commission Report
(1945),
but it is
not very thorough, and consequently
I
have relied exclusively on data furnished
privately to me by the U.K. Textile Manufacturers Association.
Although the data are of mixed quality, with variations in the areas covered and the
type of information, they are also quite powerful. What has been collected is a unique
series
of
precise evidence on wage differentials within the working class over the last
hundred years. which is both industrially and geographically specific. The averages
of
To continue reading
Request your trial