Book Reviews
| Published date | 01 March 1970 |
| Date | 01 March 1970 |
| DOI | http://doi.org/10.1111/j.1467-8543.1970.tb00577.x |
BOOK
REVIEWS
A
Cmhrry
of
Pay
by
E.
H.
Phelps Brown with Margaret
H.
Browne.
PROFESSOR
Phelps Brown has expanded his well-known study of wage-rates
in five industrial countries since 1860 into a full-scale examination of the
historical record over the past century. He assembles for each of the five
countries (France, Germany, Sweden, the United Kingdom
and
the
United States) a mass of statistical data on wages, incomes, prices,
employment and productivity, some in the form of time series, some for the
purpose of making international comparisons at particular times. The
century is divided into three
periods-1860-1913,1920-38,
and 1946-60-
and each period is examined separately, first from the point of view
of
international differences in levels of income at the date selected within the
period--1905-9, 1930-31, and 1953-and then from the
point
of view of
changes over the period. Woven into the statistical account is a com-
mentary and analysis, directed towards establishing the movements that
took place, and providing a ‘framework of explanation’ of their causes.
These explanations (or theories) are interesting for the light which they
throw
on
the operation of the economic system in the various countries
and on the elements in the system common to each. There have been few
more outstanding and sophisticated examples of what Profasor
Phelps
Brown calls ‘reasoned history’.
The task of preparing the statistical record, testing its reliability and
ensuring,
as
far
as
possible, consistency and comparability
is
a vast one. It is
difficult enough to accomplish this on a national scale and a great deal
more difficult where several different countries are involved and the data
extend over
a
century. There must always be a large element of judgment
(not to say speculation) in many of the key figures, particularly those relat-
ing to the first two periods, and this makes it necessary to offer explanations
of the observed changes with due caution since,
as
the authors point out,
explanations are no more trustworthy than the record on which they are
based. Some illustrations of the need for caution, drawn exclusively from
the first of the three periods (1860-1913) are given below.
Many readers
will
regret that the conspectus offered in the statistical
tables and charts remains fragmented by the division into three periods
so
that there are no continuous time series covering the entire century.
It
would also have been helpful
if
the estimates of GNP used
in
the
Appendix
had been reproduced,
as
a
continuous series or otherwise. In one table
only (Table 33)
is
a measure provided of the change over the whole span of
years. This is for real wages in each
of
the five countries and
shows
a
Macmillan and
Co.
Ltd., London, 1969, 476 pp., 105s.
117
118
BRITISH JOURNAL
OF
INDUSTRIAL RELATIONS
fourfold increase in France, Germany and the United Kingdom, a fivefold
increase in the United States and a multiplication by a factor of 7.5 for
Sweden. But no comparable table appears for other variables such as out-
put, prices and productivity. The focus of interest is on the process of
development within each period rather than on the changes accomplished
over the century.
Professor Phelps Brown seems to have most difficulty with the figures
for France and finds the estimates for national income and income gene-
rated by industry for the inter-war period (Period Two) in France
so
conflicting that he is unable to adopt any of them. This makes it all the
more curious that he feels able to adopt estimates for Period One that may
well be in no better case. The chart of industrial productivity (p. 123)
which shows
a
rise of over 60 per cent between 1894 and 1913 ought to
arouse even more scepticism than the authors profess. Over the same period
real wages in France rose by
7
per cent (p. 433).
So
wide a divergence
between the changes in productivity and real wages is highly unusual and
in marked contrast to the experience of the United Kingdom over the
same period. The British figures show an increase of 2 per cent in industrial
productivity (Appendix 3) and
a
fall of about
2
per cent in real wages
(ibid).
I
should lay long odds that the British figures under-state and the
French figures greatly over-state the improvement in productivity.
The German figures also seem defective. The big drop shown in in-
dustrial productivity from 85 in 1878 to
78
in 1880 is quite inconsistent
with the jump in output that took place in 1880. As Sir Arthur Lewis has
pointed out, Hoffman’s index of industrial production, which forms the
basis of Professor Phelps Brown’s estimates, is much too high throughout
the seventies because
of
a
change in 1879 in the method used by the
German Customs authorities for recording raw material imports (on which
the estimates of industrial production are based). The impression given by
the chart on
p.
116 of
a
prolonged pause between the middle seventies and
middle eighties in German development is thus mistaken. There
is
also
something not altogether plausible about the figures for money wages,
which show
a
drop of 18-19 per cent between 1873 and 1877-not far
short of the fall in 1929-33-although industrial production continued to
increase. The fall in real wages by 8-9 per cent over the same period also
looks suspicious.
When we turn to the United Kingdom we are immediately struck by
the extraordinary slowing down at the beginning of the century. That real
wages ceased to improve is not in question. But is this true also of pro-
ductivity? Is it really conceivable that GDP per occupied person fell by
7
per cent and industrial productivity by
4
per cent between 1899 and 1913?
It may be conceivable but it seems highly unlikely. Maddison (who is not
mentioned in the bibliography) shows output per man-hour rising from 93
in
1900
to
100
in 1913. There are three grounds for scepticism: other
estimates of GDP show a bigger increase than Feinstein’s; the rise in
BOOK
REVIEWS
119
industrial production as estimated by Hoffmann and Lewis also points to a
bigger increase; and there is considerable doubt about the available price
indices.
To
take the last point first.
If
what we are seeking to establish is the
movement of productivity we need
a
price index of British output.
Professor Phelps Brown, however, used a GNP deflator to convert net
domestic income into real terms and
so
introduces a terms of trade effect
into the final result. His deflator gives much too high a weight to food and
very little weight to industrial output (the Rousseaux index of industrial
wholesale prices which he uses relates almost entirely to imported
materials). Over the whole of the period 1860-1913
it
shows little
or
no net
change, falling by 23 per cent
to
1895 and rising thereafter by
28
per cent.
The price deflator used by Professor Phelps Brown for industrial produc-
tion shows broadly similar movements: the rise from 1895 to 1913 works
out at 25 per cent. But this index too is unrepresentative of the value
added by British industry since it takes no account of the swing in the
price of imported materials or of economy in their use.
A
GDP deflator
recently prepared by Sir Arthur
Lewis
shows
a
more marked upward
trend over the period, rising by 7 per cent up to 1895
and
then by
16
per
cent from 1895 to 191 3. If this deflator were used instead of either of those
used by Professor Phelps Brown the improvement in productivity would
be less than half
as
rapid up to 1895 and would become positive (but
slower) over the rest of the period. The comparison would run
as
follows
:-
Real domestic income per occupied person
Phelps Brown
Lewis
deflator deflator
1860 54 68
1895 103 94
1913 102 102
If we take next the increase in industrial production
this
works out,
using Hoffmann’s index excluding building,
at
nearly 50 per cent for the
period 1895-1913-fasterY indeed, than in the 18 years 1881-99.
No
doubt
there was a progressive slowing down from about 1870 onwards, and
there was certainly a very marked contrast between the two periods in the
trend in building. But there is nothing in these figures to suggest stagnation
after 1900.
Finally, there
is
a
conflict
of
evidence
as
to
what happened to GDP at
current prices.
Sir
Arthur Lewis, in the unpublished work on which
I
have
drawn, gets an increase of 62 per cent between 1895 and 1913 while
Feinstein gets only 53 per cent for the rise in net national income over the
same period.
I
conclude from all this that the deceleration of growth at the turn of
the century may well have been
a
good deal less sharp than Professor
To continue reading
Request your trial