This PDF is a selection from an out-of-print volume from the National

Bureau of Economic Research

Volume Title: Wages in Germany, 1871-1945

Volume Author/Editor: Gerhard Bry assisted by Charlotte Boschan

Volume Publisher: Princeton University Press

Volume ISBN: 0-87014-067-1

Volume URL: http://www.nber.org/books/bry_60-1

Publication Date: 1960

Chapter Title: Wages in Germany, Great Britain, and the United States

Chapter Author: Gerhard Bry

Chapter URL: http://www.nber.org/chapters/c2510

Chapter pages in book: (p. 266 - 322)

 CHAPTER 6

Wages in Germany, Great Britain,

and the United States
General
IT IS the purpose of this chapter to summarize the major findings in
German wage behavior and to confront them, wherever possible, with the
results of corresponding inquiries for Great Britain and the United States.
Such a comparison should help to determine which features of German
wage behavior can be regarded as characteristic of wages in general, and
which must be explained by historical circumstances peculiar to Germany.
Ideally, a number of other industrial countries should be included in the
international comparisons, but only two are dealt with in order to keep
the discussion within manageable limits. Great Britain and the United
States were selected, first, because the industrial histories of the three
countries are roughly comparable; second, because they formed the core of
world industrialism during the three-quarters of a century ending in 1945;
and third, because wage series for each of them are relatively plentiful.
It stands to reason that, within the confines of a brief chapter, the
comparisons must often rest on the results of investigations made by
students of wages in Great Britain and the United States. At times,
readily available summary measures in the form of wage and price indexes
were used. Since these measures for Great Britain and the United States
are used "as available," that is, without adjustments to assure compar-
ability with the German data, most comparisons must be regarded as
rough and ready, indicating only broad similarities and differences.
Before we turn to comparison of wage behavior, a brief review of the
economic development of the three countries is indicated. By the time
Germany had attained political unity and launched its career as an indus-
trial nation, Great Britain had already achieved a high level of industriali-
zation and a commanding role in the markets for manufactured goods.
Britain's advanced industrial development, the limitations of its domestic
market for industrial products, and its need to import both raw materials
and foodstuffs made that country highly dependent upon industrial
exports. To a considerable extent, Germany's industrialization progressed
in an atmosphere of economic competition with Great Britain. During
the first decades of Germany's economic development, when the industriali-
zation process itself created a rapidly mounting demand, German industrial
products began to replace imports from Britain in the domestic market—
especially if the market was protected from British competition. The
conifict became sharper when Germany was well established as an in-
dustrial nation, with the needs for raw materials and foreign markets
266
 wW

GERMANY, GREAT BRITAIN, AND THE U. S. 267

characteristic of an industrially mature country. In foreign markets
Germany soon challenged Britain's virtual monopoly. As for the United
States, it was blessed with a unique combination of advantages. A relative
late-corner to the industrial field, at least compared to Britain, it enjoyed
the benefits of rapid industrial growth. But unlike the other two countries,
the United States encompassed a huge territory rich in resources and with
a large and growing domestic market—an "empire" within its own borders.
This made possible extensive industrial expansion without an immediate
need to invade the markets of competing nations or to defend American
markets politically.
The differences in the economic and political development of the three
nations affected the pace of their industrial progress as well as their
economic and military fortunes during and after their first overt conifict—
World War I. Great Britain's victory after heavy losses, America's late
entry into the struggle and her important contribution to the Allied cause
to some extent re-established the array of power as it had existed before the
war. Britain maintained an essentially defensive economic position based
on the resources of her empire, and the United States continued a relatively
unimpeded economic growth. Germany, after its defeat, was thrown back
to the position of the tardy aspirant who must start anew. But by 1929
it had largely recovered from the worst effects of defeat and had modern-
ized its productive apparatus. The ravages of the Great Depression
provided the political opportunities for preparing a new all-out challenge
to the international status quo. The National Socialists devoted Germany's
industrial potential to rearmament, and tried to adjust the European
political scene for a second time to what must have appeared to them as
a new balance of economic power. World War II was the result of this
challenge. And in that war Germany's defeat was decided—to a far
greater extent than in World War I—by the ever-growing might of two
nations, the United States and the Soviet Union. Germany had not been
able to attain industrial predominance by being first in the field like
Great Britain, or by being large and protected like the United States.
German ambitions, if they were to be realized, must overcome strong
economic and political resistance. The first attempt to break this resistance
ended in a major setback, the second in Germany's destruction as a
political unit.
The patterns of the relationship among the three industrial powers
have been important determinants of trends in economic growth and in
wage behavior. At an earlier point in this study we have followed real
per capita income changes in Germany, Great Britain, and the United
States, for most of the period with which we are concerned (see Table 5).
Let us recapitulate the broad findings insofar as they bear upon wage
developments. Between 1871 and 1913, real per capita income doubled
in Germany and Great Britain, trebled in the United States. Between 1913
and 1939 the increase was about one-third in the United States and a little
268 WAGES IN GERMANY
less in the other countries. The climate of the interwar period as a whole,
however, differed sharply among the three countries. In Germany, real
per capita income for 1925-32 was 7 percent below the 1913 level, in
Great Britain 7 percent above, and in the United States 26 percent above.
There were notable differences also in trade union organization. The
following tabulation shows total union membership in percent of the
gainfully occupied population of each country. Before World War I,
the degree of organization was relatively low in all three countries, with

Trade Union Membership, in Percent of Labor Force

Germany Great Britaina United States

1910 8 14 6
1920 42 43 12
1930 24 22 7
a Great Britain's union membership in 1910, 1920, and 1930 is compared with the
working population for 1911, 1921, and 1931. If union membership data in the latter
three years are used for comparison, the percentages are 17, 34, and 21, respectively.
souRcE: For Germany, our estimates: union membership, three big unions (Table 11)
plus estimated other unions (see Table 12 for interwar period); labor force, interpolated
from data for census years (see Table 6). For Great Britain and the United States, see
Leo Wolman, Union Membership in Great Britain and the United States, Bulletin 68,
National Bureau of Economic Research, 1937, p. 10.

Great Britain ranking first, Germany in an intermediate position, and

the United States last. After World War I, Germany attained a degree of
organization roughly comparable to that of Great Britain. In all three
countries unionizatio.n grew rapidly between 1910 and 1920, and declined
sharply between 1920 and 1930. By 1920 unions in Germany and Great
Britain accounted for more than 40 percent of the working populations,
but in the United States for only 12 percent. In 1930, however, in both
Germany and Great Britain, only 20 to 25 percent of the gainfully occupied
population belonged to trade unions, and in the United States less than
7 percent. Unions in the United States did not attain their major growth
until the later years of the Great Depression. By the end of the 1930's
the degree of unionization in the United States was probably about 15
percent and in Great Britain 28 percent.' In the meantime, however,
German trade unions had been ingulfed by the Nazi Labor Front.

Long-Term Trends
MONEY WAGES
During the three-quarters of a century under review, money wage levels
in all three countries showed clearly defined and substantial growth
trends. The rises were not uninterrupted, but the factors making for
growth in each of the countries were persistent enough to bring about,
1 EstimateS by Leo Wolman, personal communication.
 GERMANY, GREAT BRITAIN, AND THE U. S. 269
for the period 187 1-1944, a quadrupling of hourly wages in Germany and
Great Britain, and a sevenfold increase in the United States.
Appendix Table A-48, and Charts 33 and 34 present indexes depicting
the approximate course of wage levels in Germany, Great Britain, and the
United States.2 We observe from these exhibits that up to World War I,
hourly money wages increased most rapidly in Germany, next in the
United States, and most slowly in Great Britain. After 1913 the picture
changed radically, with Germany now showing the least rapid advance in
hourly wages. British rates rose on the whole more steeply after 1913 than
did German rates, and United States earnings made by far the strongest
gains (see Chart 33). During the three decades following the outbreak of
World War I, hourly wages in Germany rose by about one-half, tripled
in Britain, and increased four-to-five times in the United States.3
Average hourly wage levels, relative to 1913, are shown in summary
measures for selected periods in the following tabulation. For all three
Germany Great Britain United States
Period Rates Earnings Rates - Earnings
1924-32 153 166 194 243
1924.39 148 162 194 249
192,4-44 147 166 211 281

periods, German money wage levels are seen to be closer to the 1913
base than those in Great Britain and the United States. This is true whether
German wage rates are compared to British rates, or German earnings to
United States earnings. The relative positions of money wages in the
three countries are plotted also in the upper portions of Charts 35 and 36.
Of particular interest is a comparison of wage changes during the last
decade and a half of the Reich's existence. Between 1929 and 1944,
German hourly wage rates and earnings show a net decrease while British
wages show an increase of 50 percent and United States wages an even
larger growth (80 percent in hourly earnings).
Because the measures presented above must have been influenced by
differences in wage concepts and coverage in. the several national indexes,
2
For Germany, the data developed in Chapter 2 were used. For Great Britain and
the United States, sources and adjustments are described in Appendix Table A-48
and notes thereto. The intention was to base comparisons on comprehensive wage
measures over long periods of time. Resort to disparate wage measures was unavoidable,
since comparable long-term series of rates and earnings were not always available.
To permit comparisons between roughly equivalent wage measures, both rate and
earnings series were reported for post-1913 Germany. Comparison between the two
series shows a steeper long-term increase in earnings than in rates. In order to exclude
the effects of the differences in wage concepts and of differences which might arise from
the varying industrial and occupational coverage of the national indexes, direct com-
parisons of wages for building workers will also be carried through. For 1913 and 1924-
44, some measures of wage rates proper can be obtained for all three countries.
For the United States, these observations are largely based on earnings, but there
is no doubt that similar rises would be shown also by rates.
270 WAGES IN GERMANY

CHART 33
Hourly Money Wages in Germany, Great Britain, and the United States,
1871—1913 and 1924—1944

Index (1913100) Index (1913=100)

200

—
F-.
0
00
't) 0 0
0
to
0 0
—
it)
,-.
0 IL)
C'J
0 U)
r)
0
q
0 0 0 0— '0 Q)
— ' C)
— C)
—
C)
—
C)
.- 0)
_ 0)
—
0)
.-

Source: Appendix Table A-48.

 GERMANY, GREAT BRITAIN, AND THE U.S. 271

CHART 34
Daily or Weekly Money Wages in Germany, Great Britain, and the United
States, 187 1—1913 and 1924—1944

Index Index

I.-.
-.
P-
0 U) U) 0 U)
0 0.- I') 0 U) 0
IVI
U)
,fl
0
r r —
Q
'
U,
r r r
—
r r r r0)
— —

Source: Appendix Table A-48.

 272 WAGES IN GERMANY

CHART 35
Hourly Money Wages, Cost of Living, and Real Wages in Germany,
Great Britain, and the United States, 1924—1938

Germany Great BrLtain

Money Wages Money Wages Money Wages
300

250

0
200
N)

a) v-...-:
C

100 I LI I

Cost of Living Cost of Living Cost of Living

\
200 ......

0
0 •:•:

150
S

100 I I i i i i :L-:•:j j.•:•.. I I I i i i

Real Wages Real_Wages

200

::

0 :•: :•:•:•:•:•:•:•:
:::

:::.:.:::::::
:::
. Rates

.....-.-...:..

• •::•
::::::::::::::::
:::
C
I •:—:— — lp&

: ::::::::::: Aotio scales

i

0 N NN 0 NN) N) (OW N '.0W

N N 0 N t (0W
0)0)0)0)0)0)0)0)
N N N N) N) N)
N) N) N)
0)0)0)0)0) a N)
m
N) Ni
0)0)0)0) O Ni0)0)0)N) N) N)

Shaded areas represent business contractions; dots indicate cyclical turning points of the series.
Source: Appendix Tables A-48, A-49, and A-50.
 S

GERMANY, GREAT BRITAIN, AND THE U.S. 273

CHART 36
Weekly Money Wages, Cost of Living, and Real Wages in Germany,
Great Britain, and the United States, 1924—1938

Germany Great Britain United States

Money Wages Money Wages Money Wages
250

200 —

V
. Rates

150 Earnings

I
100 — S.

90 I t I I I I I I

Cost of Living Cost of Uving Cost of Living

— .." —

\ —/
200 .

0 : .— — :::
0
-
II

a)
150 '\
3,
0 ::
C : •:•::::•::::•:: •
I
tOO ELi :
I I I I I

Real Wages Real Wages Real Wages

200

ISO
0
0
II

a) Earnings
'C
100
C
90
80
70
WNWON N 0 WNWONr)
mm mm ma)r) m
NN N N N N
W
a) a) a) a)
r4)
a) a) mm
r()
a) a) a) a) a) a) m m

Shaded areas represent business contractions; dots indicate cyclical turning points of the series.
Source: Appendix Tables A-48, A-49, and A-50.
 274 WAGES IN GERMANY

an attempt was made to exclude the effect of such differences by comparing

hourly wage rates for skilled and for unskilled building workers in the
three countries. These measures will be referred to later, in connection
with the analysis of real wages and wage differentials. The result, simply
stated, is that the relative movements of the three series for building workers
generally confirm the major findings derived from the comprehensive
indexes (see Appendix Table A-51).4
The seventy-five years covered by our survey witnessed a drastic
reduction in the length of the workweek, amounting to roughly one-third,
in each of the three countries. Though direction and total reduction are
similar for all three, the levels and the timing differ from one country to
another. Germany's workers, in 1871, still labored six days a week and
about twelve hours a day, or approximately seventy-two hours a week.
In the United States in that year the workweek was shorter, probably
not much above sixty hours.5 About the same workweek prevailed in
Great Britain.6 Table 69 traces the approximate course of the reduction
of hours in the three countries. Note that from 1871 to 1932 hours seem
to have gone down faster in Germany than in the other two countries.
The faster reduction of hours in Germany prior to World War I is
intimately related to that country's position as an industrial late-corner,
starting out with a long workweek in 1871. After 1933, the extraordinary
conditions that prevailed in Germany make it difficult to ascertain
"trends."
In view of the general movement toward shorter working hours in all
three countries, the upward trend of weekly wages is, of course, milder
than that of hourly wages. The differences in the development of hours in
the three countries affect the relation among the national wage trends
(see Appendix Table A-48 and Chart 34). From 1871 to 1944 weekly
wages increased about threefold in Germany and Great Britain, but almost
sixfold in the United States. Again, the pre-1913 trend in German weekly
wages is steeper and the post-1913 trend flatter than the comparable
trends in the two otifer countries. The lower average level, relative to 1913,
of German wages during the selected interwar periods shows up also in
weekly wages. Thus, the major findings derived from hourly wage informa-
tion seem confirmed by the movement of weekly wages.
During the interwar period 1924-39,skilledbuildingworkersinGermanycommanded
hourly rates about 50 percent above those prevailing in 1913. Comparable British rates
were about 90 percent and United States rates almost 150 percent above pre-Worid
War I levels. Increases in the rates for unskilled building workers were steeper in all
three countries but the order in the relative rise remains the same.
See Clarence D. Long, Wages and Earnings in the United States, 1860-1890
(Princeton University Press for National Bureau of Economic Research, 1960), Table 13.
See also Cohn Clark, The Conditions of Economic Progress (London, Macmillan,
1951), p. 47; and Joseph S. Zeisel, "The Workweek in American Industry," 1850-1956,
Monthly Labor Review, January 1958.
A. L. Bowley, "Wages, Earnings and Hours of Work, 1914-1947, United Kingdom,"
London and Cambridge Economic Service, Special Memorandum No. 50, May 1947, p.1 1.
 GERMANY, GREAT BRITAIN, AND THE U. S. 275
TABLE 69
Hours Worked per Week, Germany, Great Britain, and the United States,
Selected Years, 1871-1944
(1913 = 100)

Year Germany Gr eat Britain United States

1871 120 114 113

1880 117 102 112
1890 113 102 107
1900 107 102 106
1913 100 100 100

1924 81 87 88
1929 84 86 89
1932 75 86 77
1939 88 89 76
1944 88 . 90 91

SOURCE:
Germany:
1871-1913, estimated, see Chapter 1, section on Trends in Hours of Work. For
1913-44, based on ratio of weekly earnings to hourly earnings, as given in Appendix
Table A-48.
Great Britain:
1871-1913, based on ratio of weekly rates to hourly rates, as given in Appendix
Table A-48. For 1913-42, Cohn Clark's estimate of average hours worked, including
agriculture, The Conditions of Economic Progress (London, Macmillan, 1951), p. 63.
For 1944, United Nations, Statistical Yearbook, 1949-1950, p. 89.
United States:
187 1-90, Clarence D. Long, Wages and Earnings in the United States, 1860-1890,
Table 13. Spliced to later segment in 1890. For 1890-1913, Albert Rees, in 38th
Annual Report (National Bureau of Economic Research, 1958), p. 59. For 1913-44,
Bureau of Labor Statistics, as given in Historical Statistics of the United States, 1789-
1945, p. 67, Series D 118.

WAGES AND PRICES

We may ask to what extent the marked wage increases and the differences
between the wage trends in the three countries are due merely to variations
in general price levels—for it is obvious that price levels must have had
some influence on wage trends. Table 70 shows long waves of raw material
prices in all three countries. At the foundation of the Reich in 1871, these
prices were as high or higher than they were on the eve of World War I;
in 1890 they were 15 to 20 percent lower; and in 1929 they were 30 to
40 percent higher. In 1929, price levels, as measured by the indexes, were
50 to 70 percent higher than in 1890, a fact that helps to explain the wage
rises during the same period. But between 1871 and 1913 wages rose,
although price levels were as low or lower in the later than in the earlier
year. And between 1924 and 1939 hourly wages showed a net rise and
prices a net drop. It is true that the price indexes cover only a small
part of the multitude of goods sold, and also that they are not comparable
276 WAGES IN GERMANY

TABLE 70
Wholesale Prices in Germany, Great Britain, aM the United States,
Selected Years, 1871-1944
(1913 = 100)

Great United
Year Germany Britain States

1871 100 118 119

1890 86 85 81
1900 90 88 80
1913 100 100 100

1924 136 165 141

1929 131 134 137
1932 86 95 93
1939 96 113 110
1944 110 186 149

Averages
1924-32 123 133 130
1924-39 109 121 122
1924-44 108 133 125

SOURCE:
Germany: Appendix Table A-I.
Great Britain: For 1871-19 13, Sauerbeck-Statist Index, as published by U. S. Bureau
of Labor Statistics, Bul. No. 284, p. 280. For 1913-44, League of Nations, Statistical
Year-book, 1932-33, p. 268, and 1942-44, p. 195.
United States: Historical Statistics of the United States, 1 789-1945, pp. 233-234,
Series L 15.

between countries. However, there is satisfactory evidence that neither

the growth nor the differential behavior of wage trends can be explained
primarily by variations in wholesale price levels.
Appendix Table A-49 and Chart 37 present data on retail price changes,
in the form Of indexes of living costs. These prices show a closer corre-
•spondence to wages than do wholesale prices. Marked long-term growth
trends are to be noted both in hourly wages and in living costs through the
last fifty years of the period under investigation. The faster net rise of
German wages before World War I—relative to that in the other two
countries—is paralleled by a steeper increase in living costs. Similarly,
the slower rise of German wages between 1913 and 1939, for instance,
finds a parallel in a milder increase in consumers' retail prices. However,
the long-term increases in wages are greater than in living costs, and the
differences in wage trends are not simply related to differences in retail
prices. For instance, the rise of money wages in the United States, from
1913 to 1939, exceeds substantially that in Great Britain, whereas the net
advance of living costs appears to have been greater in Great Britain than
in the United States during those years.
 GERMANY, GREAT BRITAIN, AND THE U. 5.. 277

CHART 37
Cost of Living in Germany, Great and the United States, 1871—19 13
and 1924—1944

index (1913=100) Index (1913=100)

Source: Appendix Table A-49.

REAL WAGES
It was found that hourly real earnings just about doubled in Germany
between 1871 and 1944. This increase is of course considerably less than
that of the comparable money wages, which quadrupled. As indicated by
the indexes presented in Appendix Table A-50 and Chart 38, the increase
in German hourly real wages is a little below that in hourly real rates in
Great Britain; it is substantially less than the rise of hourly real earnings
in the United States, which was fivefold. Before the outbreak of World
War I, German hourly real wages rose faster than British but more slowly
than those in the United States. After 1913, German hourly real wages
lagged. The low standing—in relation to 1913—of German hourly real
wage levels during the interwar period is brought out again in the lower
panel of Chart 35 and the three sets of averages found in the following
278 WAGES IN GERMANY

CHART 38
Hourly Real Wages in Germany, Great Britain, and the United States,
1871—19 13 and 1924—1944

Index (1913=100) Index (1913= 100)

0 0 IC) 0 0 it) 0 It) 0 It) 0 q

1- N 0) 0) 0
m
0 —
0)
— C'.' C'J
o
N) N)
0)
— _ -. — •1-. — w_ •- — ,_ — — —

Source: Appendix Table A.50.

tabulation. In order to permit comparisons unaffected by differences in

Germany Great Britain United States

Period Rates Earnings Rates Earnings
1924-32 108 117 119 145
1924-39 111 121 124 162
1924-44 109 124 127 179

SOURCE: Appendix Table A-50.

wage concept and coverage, hourly real wage rates of building workers in
the three countries have been analyzed, too. They show basically similar
 S

GERMANY, GREAT BRiTAIN, AND THE U. S. 279

CHART 39
Daily or Weekly Real Wages in Germany, Great Britain, and the United
States, 1871—1913 and 1924—1944

Index (1913=100) Index (1913= 100)

50

150

100
90
80

70

Source: Appendix Table A.50.

behavior.7 The data and measures underlying these results can be found
in Appendix Table A-52.
Appendix Table A-50 and Chart 39 contain measures of weekly real
rates and earnings. The rise of weekly real wages is of course affected by
the reduction of working hours which occurred at somewhat different rates
in the three countries. The net rise of weekly real wages between 1871 and
1944 amounted to about 55 percent in Germany, in Great Britain to about
80 percent, and in the United States to about 300 percent. Between 1871
and 1913 the weekly real wage rises in Germany and Great Britain are very
similar. However, they were concentrated in different subperiods. While
Similarity is to be noted in the low interwar position of German hourly real wage
rates, the intermediate position of British, and the high levels of United States wages,
all measured relative to 1913. This order—though not the extent of differentiation—is
the same as that observed in money wage trends.
280 WAGES IN GERMANY
enormous gains seem to have occurred in Great Britain during the decades
of the 1870's and 1880's German real wages rose considerably faster than
British toward the end of the century. As pointed out in Chapter 2,
around 1900 weekly real earnings in Germany had virtually reached their
1913 level. Comparison with real wage behavior in England suggests
that this was not necessarily a specifically German phenomenon. Indeed
the indexes seem to indicate that 1913 levels in Great Britain were reached
as early as 1890 and that they were actually exceeded around 1900. It is
necessary, however, to make liberal allowances for margins of error.8
The evidence available at the time of this writing suggests, in any case,
that the decade or decades immediately preceding World War I did not
witness substantial increases of weekly real wages in the two large
European industrial countries. Economic progress for wage earners
during these years seems to have been concentrated in other directions.
In Germany it was expressed in a decline of average hours worked and an
increase of industrial employment opportunities at relatively high wages.
In Great Britain the available data on the length of the workweek show
only a 2 percent decrease between 1890 and 1913. However, industrializa-
tion in Britain made rapid progress during these years, and the additional
industrial employment opportunities, at wages in excess of those paid
in agriculture or handicrafts, for example, may have contributed to the
well-being of British wage earners.
For the years following 1913, both Appendix Table A-50 and Chart
39 point up the generally high levels—relative to that base year—of
weekly real earnings in the United States and the low levels in Germany.
Note that in 1929 and 1939 the relatives for Germany are higher than those
for Great Britain. However, the British index numbers are based on wage
rates for a normal workweek, and therefore do not reflect the wage
increases accruing from overtime, other work at premium pay, and perhaps
changes in the composition of the work force. These elements may have
been important factors in the wage increases between 1913 and the two
prosperous years under consideration. Thus, for these two years, the
increase of weekly earnings probably exceeded the reported increase of
weekly rates. The real wage position of the three countries during the
interwar years can be judged on the basis of the period averages relative
8
Phelps Brown and Handfield-Jones, in their article, "The Climacteric of the 1890's:
A Study in the Expanding Economy," Oxford Economic Papers, October 1952, called
attention to the stagnation of real wages in several industrial countries during the
pre-1913 decades. They advanced the idea that the "climacteric" was an international
phenomenon, brought about by basically similar circumstances. But note that, pending
further studies, the similarity of the "leveling-out" should not necessarily be regarded
as a proof of the phenomenon. The limited coverage of the cost-of-living data in parti-
cular may have brought about a similar bias in several countries. For the United States,
furthermore, recent recomputations of real wages show little of the leveling-out that
characterized earlier estimates. The more recent studies, by Albert Rees and by Clarence
D. Long, were used throughout to describe wage behavior in the United States up to
1913.
 GERMANY, GREAT BRITAIN, AND THE U. s. 281
to 1913, given below. During 1924-32 the average level of German weekly

Germany Great Britain United States

Earnings Rates Earnings
1924-32 96 104 125
1924-39 101 108 130
1924-44 105 110 147

SOURCE: Appendix Table A-50.

real earnings was 4 percent below 1913 levels and substantially lower than
the comparable position of British and United States wages. The ranking
of the three countries is, with respect to wage increases, the same if
the period is extended forward to 1939 or to 1944.
An important finding of the study of German wages was the close
resemblance of trends in weekly real earnings to corresponding trends
in per capita income. That such resemblance exists for all three countries
is suggested by the following tabulation of interwar levels relative to
1913.
Weekly Real Wages and Per Capita Real Income, 1
(1913 = 100)

Great United
Germany Britain States
Weekly real wagesb 99 105 125
Per capita real national income 93 107 126
a The period 1925 through 1932 was selected so that comparison between wages and

income could be made for the same years. German real Income data are not available
for the year 1924.
b Earnings for Germany and the United States, rates for Great Britain.
souRcE: See Appendix Table A-SO (weekly real wages), and Table 5 (per capita real
income).

Note that the relative position of weekly real wages is extremely close
to that of per capita real national income, particularly in Britain and the
United States. However, a good deal of this surprisingly high correspond-
ence must be attributed to coincidence. There is, for example, considerable
difference in the coverage of wage and income indexes. Moreover, index
levels in individual years vary far more than do the averages. Nevertheless,
comparison of real wage indexes and per capita real income data shows
the following points of correspondence:
I. Weekly real wages reflect the strong net rise in per capita real income
in the years 1871 to 1939.
2. Between these end years the approximately equal rise of per capita
income in Germany and Great Britain, as well as the substantially
greater rise in the United States, is reflected in real wage behavior.
3. A similar correspondence is notable also for the interwar periods
1925-32 or 1925-33, relative to 1913.
282 WAGES IN GERMANY

4. The rank of the countries and the basic correspondence of real

wages and real income are maintained also if the period of comparison
is expanded to 1939. That is, during the decade and one-half between
the stabilization of the German currency and the onset of World
War II, German weekly real earnings and per capita real income
averaged approximately their 1913 levels; British real rates and per
capita income were roughly 10 percent above 1913; United States
earnings were about 20 percent and per capita income 30 percent
above their levels on the eve of World War I.
It is not necessary to analyze such correspondence in every detail.
Note, however, a striking lack of agreement with reference to the United
States, where per capita real income is reported to have almost trebled
during the years 1871-1913, while weekly real earnings advanced by only
70 percent. Part of this difference can be readily explained as a consequence
of the industrialization process, during which the employment composition
shifted from agriculture to manufacturing. But there may be other causes
such as the use of different deflators, and perhaps inadequacies in the
basic information. Apart from this major exception, a basically close
relation between per capita national income and average weekly real
earnings in the three countries prevails.

Wage Differentials
GENERAL
Trends of German wages—according to the analysis in Chapter 2—tended
to show considerably less dispersion than, for instance, trends of different
groups of wholesale prices. Similar conclusions follow from a cursory
examination of data for Britain and the United States. Again, it is not so
much the absence of consonant changes in the price structure that dis-
tinguishes wage changes from price changes, but rather the grossly in-
congruous behavior of a few classes—which can be found usually in
wholesale prices but rarely in wages. To take an example from the United
States: while the index of building material prices computed by the
United States Bureau of Labor Statistics increased two and one-half times
between 1890 and 1944, the index of metal and metal product prices
decreased slightly.9 Such divergences in trends cannot be found in wage
movements. The reason is that technological progress and consequent
decreases in production costs may affect industrial products in widely.
different ways, though they tend to have a uniform influence upon the
price of labor.
The relative uniformity of wage changes does not, of course, preclude
gradual changes in the wage structure. For Germany these changes have
been measured by wage differentials and described in Chapter 3. We have
observed a long-term tendency toward decreasing differentials, that is,
See Historical Statistics of the United States, 1789-1945, pp. 23 3-34.
 GERMANY, GREAT BRITAIN, AND THE U. S. 283
toward a greater equality in the wage structure. However, the equalizing
trends showed differing strengths in the several segments of the wage
structure and during different periods of German wage history.

SKILL DIFFERENTIALS
The analysis of German skill differentials showed a long-term tendency
toward narrowing, and a fairly close empirical relationship between
changes in skill differentials and changes in living costs. In keeping with
the latter relationship we found a tendency toward first a widening, and
then a narrowing, of skill differentials before 1913. During World War I
and continuing through the inflation, the skill gap declined sharply, and
by 1923 had virtually disappeared. Stabilization brought a re-establish-
ment of substantial differences between the wages of skilled and unskilled,
but never to the extent obtaining before the war. In the early post-
stabilization years there was a moderate tendency toward a further narrow-
ing of the skill gap. However, from the end of the 1920's the relation
between wage rates of unskilled and skilled workers remained fairly stable.
This was the case not only during the last years of collective bargaining,
but also during the wage administration of the National Socialists. For
all of this period wage rates of unskilled male workers, in an average of
seventeen industries, amounted to about 80 percent of those for
skilled.
Neither the long-term tendency toward a narrowing of skill differentials
nor the close relation between these differentials and living costs was a
peculiarly German phenomenon, as we may observe when we compare
skill differentials in the building industries of Germany, Great Britain,
and the United States. Chart 40 and Appendix Table A-53 show important
elements of correspondence among the major changes in skill differentials.
Noteworthy in this connection decline of these differentials between
1913 and the early 1920's, the subsequent moderate increase to less than
prewar scope, and the more recent tendency toward a further reduction of
the skill gap. Illustrations are taken from the building industry because
long-term data on that industry are available for all three countries.
There is much evidence that basically similar trends characterized wage
behavior in the manufacturing, mining, and transportation industries.'0 It
10 For the United States, see Leo Wolman, in 32nd Annual Report, National Bureau

of Economic Research, 1952, P. 43; Harry Ober, "Occupational Wage Differentials

1907-1947," Monthly Labor Review, August 1948, pp. 127-34; Philip W. Bell, "Cyclical
Variations and Trend in Occupational Wage Differentials in American Industry since
1914," Review of Economics and Statistics, 1951, pp. 328 if.; Edwin Mansfield, "Wage
Differentials in the Cotton Textile Industry, 1933-1952," in Review of Economics and
Statistics, February 1955, p. 80. For Great Britain, see the data for industries other
than building presented by K. G. J. C. Knowles and D. J. Robertson, "Differences
between the Wages of Skilled and Unskilled Workers, 1880-1950," Bulletin of the Oxford
University Institute of Statistics, April 1951, p. 111; also Bowley, "Wages, Earnings and
Hours of Work, 1914-1947, United Kingdom," p. 6. For evidence on further contraction
of British skill differentials during recent years, see Edwin Mansfield, "A Note on
Skill Wage Differentials in Britain, 1948-54," Review of Economics and Statistics,
284 WAGES IN GERMANY

CHART 40
Skill Differentials in the Building Industry and Cost of Living, Germany,
Great Britain, and the United States, 1904—1950
Index (1913=100) Germany Per cent
250

200

150

100

50

Great Britain
300

250-

200 -
I
' Cost of living,'
I I
I
150 -

100 - -——a———
.—'I 40

30

20

HIll 11111111 I I iii ii iiil iiuli iiil i I tili iiili iiil to

United States
250
,
I
..'/
200- s.

/ \•____ Cost of living

150- I
I /
II
100 -

50-
40

30
i i Ii I I I I i i i I i I I

o ui 0 0 In 0 ifl In
o 0 . —
r
C\J C"J
0'
— —
01
—
0} '01 01
— r —
0)

The differentials are the difference between wage rates skilled and unskilled workers, expressed as percent
or the former.
Source: Table 74, Appendix Tables A.49, A-53, and their sources.
 GERMANY, GREAT BRITAIN, AND THE u. s. 285
isworth noting that the magnitudes of the skill differentials, despite their
dependence upon the classification used and upon the occupations selected,
tended to resemble each other roughly in the three countries. To take an
illustration from the close of our period: around 1943 wage rates of
unskilled workers for the average of all manufacturing industries in
Germany were about 80 percent of those for skilled; practically the same
relation existed in Great Britain for the skill differentials of wage rates
in the building, shipbuilding, and engineering industries; in the United
States average hourly earnings for unskilled workers were about 73 or
74 percent of those for skilled and semiskilled. The comparable ratios for
the period before World War I were around 60 percent for Germany and
Great Britain, and a little below 70 percent for the United States."
Skill differentials were narrowest in all three countries at the close of
each war and during the years immediately following, largely owing to
the inflationary price rises. A relationship between skill differentials and
cost-of-living movements for all three countries appears clearly from
Chart 40. The differentials declined with the price rises from 1914 to the
early 1920's; they increased to more than prewar size with the subsequent
deflation; and in the late 1930's and the 1940's they declined in the non-
regimented countries while living costs rose.
The effect of retail price rises on skill differentials under extreme circum-
stances is illustrated by the fact that the differentials were virtually
obliterated in Germany where, during the Great Inflation, skyrocketing
of prices was practically unchecked. The reasons for the narrowing of skill
differentials in times of rapidly rising prices do not apply uniformly from
country to country or from circumstance to circumstance. In Germany
during World War I and the Great Inflation the need to protect the low-
paid unskilled worker was probably the most important cause. However,
the development may be very different in a country where inflation is not
accompanied by a rapid deterioration of living conditions. In the United
States during World War II, for example, increased living costs and the
narrowing of skill differentials appears to have resulted from the increasing
relative scarcity of goods and manpower, rather than from the social need
to protect low-paid workers. Here the scarcity of goods led to a rise in the
general price level, including living costs; and the scarcity of labor gave rise
to a need to procure labor for "less desirable," mostly unskilled, jobs.'2

August 1957, pp. 348-5 1. The tendency toward closing of skill gaps between 1913 and
1920, and the moderate widening thereafter have been observed also for other countries.
See, for instance, J. H. Richardson, "Some Aspects of Recent Wage Movements and
Tendencies in Various Countries," International Labour Review, 1928. pp. 179-203.
Caution is required in the interpretation of the similarity of these figures. While
it seems significant that the skill ratios are higher than about 50 or 60 percent, the
reported average ratios are affected by the selection of industries and occupations.
Around 1943, for instance, these ratios in Germany ranged from 58 percent (hard-coal
mining) to 89 percent (soft-coal mining).
12 Albert Rees made valuable suggestions on this point.
286 WAGES IN GERMANY

The evidence presented in Chart 40 suggests that living costs are not
the only major determinant of skill differentials. In the United States
we may note the diverse directions of the changes in these two variables
before World War I, and the stability of the differentials during much of the
interwar period—while living costs showed distinct short-term variations.
But even broad long-term tendencies show far from perfect correspondence.
Apart from the rise in living costs, a wide variety of long-term factors
tend to operate toward a narrowing of skill differentials. Among these are:
(1) the increasing use of mass-production techniques with an accompanying
breakdown of skilled operations into simpler jobs; (2) the mechanization
of some typical tasks of unskilled labor, such as handling, storing, and
transporting materials, entailing large expenditures for capital equipment;
(3) the spread of general education, democratic ideologies, and political
franchise; (4) the efforts of trade unions to reduce skill differentials, and
the increasing unionization of unskilled workers; (5) the growing role of
government in wage determination, tending to promote greater wage
equality especially in times of social stress; and possibly (6) the equaliza-
tion of efficiency fostered by generally higher levels of health and economic
well-being. There are, on the other hand,, factors that set limits to the
narrowing of skill differentials: (1) differences in aptitudes and training;
(2) the growing supply of unskilled labor; (3) greater ease of substitution
and sharper competition among unskilled workers; (4) a tendency to
"freeze" the wage structure in order to simplify negotiation of wage
contracts. Skill differentials, despite the observed historical trends toward
narrowing, must be regarded as a permanent feature of any industrial
wage structure.'3

AGE DIFFERENTIALS
Information on differentials in German wage rates is largely qualitative.
There are no series by which trends in these differentials can be measured
over long periods of time. But information available for a number of
briefer periods shows that:
1. Wage rates for younger workers during World War I tended to
increase more than those for adults, and especially more than those
for skilled adults.
2. During the 1920's and early 1930's, age differentials tended to decline,
largely as a consequence of the inclusion of wages for youths in
collective bargaining contracts.
3. Over the whole period 187 1-1945, the status of apprentices changed
from that of paying workers to that of paid workers.
4. The coverage of apprentice remuneration by collective bargaining
There might, of course, also be factors leading toward wider skill differentials.
An example may be technological developments requiring high skills and affecting the
relative scarcity of skilled workers as compared with unskilled. For a discussion of a
related subject see Richard Perlman, "Forces Widening Occupational Wage Differ-
entials," Review of Economics and Statistics, May 1958.
 -w

GERMANY, GREAT BRITAIN, AND THE U. S. 287

contracts tended to standardize and raise the level of such wages in
relation to adult workers' pay.
Age differentials, apart from their decrease, diminished in importance
in the wage structure, largely as a result of the increasingly stringent
provisions of child labor legislation.
Indications of a narrowing of age differentials and of a decline in their
importance in the wage structure can be found also in the wage histories of
Great Britain and of the United States. In Great Britain, child labor legisla-
tion as well as union campaigns tended to improve the wage levels of children
and youths relative to those of adults. A special problem, that of "dead-
end employment," was created by the fact that age differentials, for
instance in British coal mines, were large for youths up to 17 years of age
but contracted sharply at age 18 and over. This resulted in a common
practice of "sacking" most youths at the age of 18, when higher wage
levels made their employment less profitable. A gradual raising of wages for
youths and a consequent narrowing of age differentials brought the solu-
tion to this problem.'4 And reduced employment of youths generally
brought about a decreasing importance of age differentials in the British
wage structure.
There is little doubt that in the United States, as well, the general trend
in age differentials between wages of youths and adults was downward.
During the last decades of the period under review the general tendency
in both governmental wage regulations and collective bargaining contracts
in this country was to establish "entry rates" without special provisions
for the remuneration of youths.'5 As in the other two countries, a decreasing
importance of the age differential in the wage structure can be observed
as child labor legislation gained ground. In keeping with the experience
of other industrial countries, child labor, as reported by the Census,
increased up to the beginning of this century, when about one-fifth of all
children between 10 and 16 years of age were listed as gainfully occupied.
It is possible that the reported increase may reflect to some degree improved
reporting, and shifts from agricultural to industrial jobs, since earlier
reporting of gainfully employed youths in agricultural occupations might
have been unreliable. Less subject to doubt, however, is the subsequent
drastic reduction of the percentage of youths in the American labor force.
By 1930 less than one-twentieth of the 10 to 16 age group was gainfully
occupied; by 1940 this ratio must have dropped further.'6 Thus the major
14
See Paul T. David, Barriers to Youth Employment (American Council on Education,
1942), pp. 85-86.
15
It is sometimes pointed out that this does not always mean an improvement in the
relative remuneration of youths. See Ibid., p. 87.
16
Ibid., p. 50. See also Clarence D. Long, The Labor Force in Wartime America,
(National Bureau of Economic Research, Occasional Paper 14, 1944), Table 2; and, by
same author, The Labor Force under Changing Income and Employment (Princeton
University Press, for the National Bureau of Economic Research, 1958), Appendix
Table A-2.
 -w

288 WAGES IN GERMANY

trends observed in wage levels, related to age of German workers, appear
to have been experienced also in Great Britain and the United States.'7
SEX DIFFERENTIALS
Sex differentials in Germany tended to show a moderate decline during the
period under investigation. The decline was concentrated largely between
1914 and 1924, and was more clearly apparent when wages of women
(skilled or unskilled) were compared with those of skilled men. The rise
of wage rates for unskilled men, during the period noted, was steeper than
that for women. During the 1920's the gap between rates for women and
those for men closed somewhat more, but from 1933 on, the rate structure
was practically frozen, and little change can be observed in the relation
of women's wage rates to those of skilled male workers.
The above observations are based on rates for comparable occupations.'8
More important for the average wage level of employed women—but
not measurable by available statistics—is the fact that more and more
women were admitted to remunerative occupations and to industries
paying higher wages, a factor contributing heavily to the general trend
toward greater equality between women's and men's wages. Investigations
of fairly wide coverage but somewhat doubtful comparability show average
sex differentials to have been about 60 percent during the 1870's, about
55 percent before the outbreak of World War I, and somewhat below 50
percent on the eve of World War II.
The long-term decline in sex differentials has been observed also in
other industrial countries. In Great Britain the differential between men's
and women's earnings was reported as 56 percent of men's earnings in
1906 and 52 percent in 1924 and 1935. A further narrowing of the
differential is reported on the basis of a different sample for the period.
1938-45 (from 53 percent to 47 or 48 percent).19 Both the dimensions and
the trends of the British differentials are rather close to their German
counterparts. For the United States a long-term trend toward declining
sex differentials can be inferred from the statistics of average hourly
earnings published by the National Industrial Conference Board. For
the group of industries reported on, average hourly earnings by women
workers were below 60 percent of those received by male workers in 1914.
This ratio fluctuated between 60 and 70 percent during the period 1920-44,
and tended to stay above 70 percent in 1947 and 1948.20 A recent study of
For a description of recent world-wide trends toward a decline of age differentials
and toward equal pay for equal work, see "Wage Differentials Affecting Young
Workers," International Labour Review, December 1955, especially pp. 531-34.
18 Sex differentials in average hourly earnings for a combination of industries (constant

weights) can be computed from the mid-1930's on. Between 1935 and 1943, sex differ-
entials declined minutely (see Table 61).
19 Bowley, "Wages, Earnings and Hours of Work, 1914-1947, United Kingdom,"
p.6.
20 National Industrial Conference Board, The Economic Almanac, 1950, p. 343.

Publication of the data ceased after July 1948.

 r
GERMANY, GREAT BRITAiN, AND THE U.s. 289
sex differentials in the American cotton textile industry revealed a clear
tendency toward reduced differentials, between 1933 and 1952. The narrow-
ing of the gap was largely concentrated in the periods of the early New
Deal and World War 11.21
It appears that labor shortages during times of war have played an
important role, historically, in equalizing wages of men and women. The
relative gains in women's wages were sometimes reduced after the war—
but never to prewar levels. It might be mentioned in this context that the
trend toward narrowing sex differentials during and after a war has been
observed also in France and other countries.22

CITY-SIZE, REGIONAL, AND INDUSTRIAL DIFFERENTIALS

Tendencies toward a tightening of the German wage structure, measured
by narrowing differentials, have been observed as between cities of different
size, between different regions, and between different industries. The
narrowing of these differentials appears to have been less marked and less
unambiguous than, for example, the narrowing of skill differentials.
Some information on the narrowing of regional differentials is available
also for Great Britain and the United States. Although no published
report of long-term changes of these differentials in Great Britain was
found, it appears that regional differentials narrowed in the course of
time and—because of the development of national minimum rates—
possibly more for unskilled than for skilled workers.23 For the United
States the narrowing of North-South differentials—due to broadening
industrialization, unionization, and so on—is well established. "In most
industries, southern wage rates have been rising relative to comparable
northern rates during the past fifty years. The narrowing of the South-
North differential has generally been more marked during the past two
decades."24 Fr the cotton textile industry, the decline of the North-South
differential during the past twenty years has recently been analyzed.25
A narrowing of regional differentials for the United States as a whole has
also been found by several observers.26
Also industrial differentials in the United States have narrowed. This is
21 Mansfield, op. cit., p. 82.
22
See, for example, Richardson, op. cit., p. 191.
23 This evaluation was contributed by K. G. J. C. Knowles of the Institute of Statistics,
Oxford University. A suggestion of decreasing regional differentiation appears also in
the discussion of district and local wage variations in Great Britain, in Margot
Heinemann's Wages Front (London, Lawrence and Wishart, 1947), pp. 145-53. This
author, incidentally, describes also long-term trends toward narrowing of differences
between wages for time workers and for piece workers (p. 210).
24
R. A. Lester, "Southern Wage Differentials: Developments, Analysis, and Impli-
cations," Southern Economic Journal, April 1947, p. 386.
25
Mansfield, op. cit., p. 81.
For the period 1907-19, see J. W. Bloch, "Regional Wage Differentials 1907-46,"
Monthly Labor Review, April 1948, p. 371. For 1939-46, see W. Woytinsky and associates,
Employment and Wages in the United States (Twentieth Century Fund, 1953),
p. 481.
 290 WAGES IN GERMANY
evident from 1933 on,27 but was pronounced during and after
World War Woytinsky analyzed the industrial earnings structure
during the first four decades of this century, but his findings did not lead
to clear-cut conclusions.29 The decline of industrial differentials since the
Great Depression was largely brought about by the establishment of
minimum wages, and by wage increases in similar absolute amounts for
different industries •30

CONCLUSIONS
A general trend toward a tightening of the wage structure is observable
for all three countries, pronounced in skill, age, and sex differentials but
perceptible also in regional, city-size, and industrial differentials. It can
hardly have come about by chance that in all these aspects of the wage
structure the long-term trend was in the same direction—toward greater
equality. The mass production of goods, their distribution throughout
wide areas, the tendencies toward tight industrial organization, the growth
of communication, and the spread of education—all these tend to reduce
differences among groups of the labor force and thus the differences in
their wages.

Cyclical Behavior
MONEY WAGES
The discussion of German money wages in Chapter 4 dealt with the
conformity of wage cycles to general business cycles, their timing in relation
to reference turning points, their cyclical amplitudes, their numerical
contribution to total payroll changes, and their relation to labor market
conditions. The results of the investigation are summarized below and,
whenever possible, compared with related findings on wage behavior in
Great Britain and the United States.
Conformity. Perhaps the most important finding on the cyclical behavior
of German wage rates as distinguished from earnings is the rarity of
substantial cyclical declines. Between 1871 and 1944 wage rates showed
27 A. M. Ross and W. Goidner, "Forces Affecting the Interindustry Wage Structure,"

Quarterly Journal of Economics, 1950, pp. 255 and 263.

28 for instance, David R. Roberts, "The Meaning of Recent Wage Changes,"
See,
in into Labor Issues, R. A. Lester and J. Shister, eds. (Macmillan, 1948), pp.
228-29. See also Herman P. Miller, "Changes in the Industrial Distribution of Wages
in the United States, 1939-1949," in An Appraisal of the 1950 Census Income Data
(Studies in Income and Wealth, Vol. 23, Princeton University Press for National Bureau
of Economic Research, 1958).
29 W. Woytinsky, Earnings and Social Security in the United States (Social Science

Research Council, 1943), p. 202.

80 Ross and Goidner regard the closing of the wage structure as a kind of statistical

illusion produced by the unsatisfactory method of measuring fairly uniform absolute

increases in percentage terms (op. cit., pp. 263-65). It seems to this writer that increasing
equality in the wage structure remains a tangible effect of the fairly uniform absolute
increases—whatever form be used to describe the underlying wage changes.
 —

GERMANY, GREAT BRiTAIN, AND THE U. S. 291

material losses only twice: during the long and severe contraction follow-
ing the Grunderjahre boom of the early 1870's, and during the Great
Depression of At times, as during the 1925-26 contraction,
monthly wage rates leveled out or decreased minutely; declines in some
industries amounted to one and one-half percent of the peak level. In
other industries contracts were permitted to lapse but the levels of the
expired agreements continued to appear in the statistics (see Chart 15
and Chapter 4, p. 130). At other times, as during the major contractions
prior to 1913, a deceleration in the increase of wage rates has frequently
been observable, even on the basis of the crude annual data. Finally, there
were instances where no cyclical responses of wage rates to changes in
business conditions could be discovered. However, if all responses—
including deceleration of growth—are counted, money wage rates on the
whole appear to have conformed fairly closely to major cyclical changes in
general business conditions.
Both the resistance of wage rates to sizable downward adjustments,
and cyclical response in the form of deceleration or leveling out can be
observed also in British and United States wage experience. For the period
before World War I, cyclical observations are based on annual data only.
In Great Britain, wage rates reacted to all major contractions in the form
of actual, albeit mild, declines. Their reaction to the briefer contractions,
such as those occurring between 1900 and 1914, are not clear.32 For the
United States there is no doubt that during the major price and earnings
declines wage rates also receded.33 However, the available annual series
on rates, show no reaction to the three brief contractions that occurred in
the period
For the period following World War I the cyclical behavior of wage
rates can be judged on the basis of monthly information. In the course of
the five contractions between 1919 and 1945 in both Great Britain and the
United States, wage rates showed substantial declines only in connection
with the deflationary postwar contraction of 1920-21 and during the
Great Depression. Wage rates either did not react cyclically to the other
contractions or responded so inperceptibly that only special techniques
revealed the responses.35 However, the mild rate cycles thus established
The change of money wage levels brought about by the stabilization of the currency
at the close of the year 1923 are disregarded here.
For basic data see Wood, op. cit., pp. 102 if., and Bowley, Wages and Income in
the United Kingdom since 1860 (London, Cambridge University Press, 1937).
" For basic data see Appendix Table A-48, col. 7 (1871-90), and Historical Statistics
of the United States, 1 789-1945, p. 69. The latter figures refer to union rates and may not
be representative for the period.
See Wesley C. Mitchell, Gold, Prices, and Wages under the Greenback Standard
(University of California Publications in Economics, 1908), Table 37. Furthermore, wage
material published by labor departments of various states contains instances of cyclical
sensitivity of wage rates during later contractions.
See Daniel Creamer, assisted by Martin Bernstein, "Behavior of Wage Rates during
Business Cycles" (National Bureau of Economic Research, Occasional Paper 34,
1950), Charts 1 through 5.
292 WAGES IN GERMANY
could be related to reference cycles. All in all, it can be said that in Britain
and the United States, as well as in Germany, wage rates conformed fairly
well to major changes in business conditions.
The response of earnings to cyclical changes in business activity is
clearer than that of rates. In Germany quarterly shift earnings of coal
miners showed, on the whole, good conformity over the period 1889-1932
and a skipping of only a few brief, mild cycles. Comprehensive earnings
series became available in 1924, but for a decade were published in annual
form only. From these series it appears that earnings responded clearly
to the later of the two business cycles occurring between 1924 and 1932.
During the brief and mild contraction of 1925-26, the rise in hourly earn-
ings showed only a minute deceleration, and that in weekly earnings a
somewhat stronger one. A more regular cyclical response of hourly earn-
ings, compared to wage rates, emerges also from an examination of
United States data.
In general, weekly earnings show more reliable cyclical responses than
hourly earnings. There is, however, a strong difference between series
on an annual and on -a monthly basis. On an annual basis, weekly earnings
series tend to be cyclically rather insensitive to mild business cycles. This
is valid for both Germany and the United States. As pointed out pre-
viously, weekly earnings in Germany showed only a mild response to
the 1925-26 contraction. In the United States weekly earnings, on an
annual basis, declined during only three out of the five business contractions
occurring between 1920 and 1939. Monthly series of weekly earnings on
the other hand, declined in all five contractions.
Timing. The outstanding feature in the timing of turning points in
German wage rates is their substantial lag behind turns in general business
conditions. At the two times when distinct specific turns in wage rates
occurred (the Gründerjahre and the Great Depression), wage rates in
annual form show lags of one year or more. For the interwar period, the
timing of wage rates can be established on the basis of monthly data.
Only two business cycles fall into the poststabilization phase of this period.
However, at each of the four turning points involved, a substantial lag
of wage rates behind monthly reference dates appears. The lags vary
between seven and twenty months, their precise extent being dependent
partly on the rules adopted for determining specific turning points.36
The tendency of turning points in wage rates to lag behind those in
general business conditions appears clearly also in British and United
States experience (compare the annual dates for reference and specific
turning points given in Table 71; see also Chart 35). For Great Britain,
the average length of the lag during the 1920-40 period has been computed
at eleven months, for the United States over the period 1923-31, at nine
months.37 With reference to the turning point preceding the Great
36
See Chapter 4, pp. 138 if.
" See Cleamer, op. cit., pp. 17 and 30. NBER referencecycle turns are used throughout.
 GERMANY, GREAT BRITAIN, AND THE U. S. 293
TABLE 71

Percentage Change of Money Wages during the Great Depression,

Germany, Great Britain, and the United States
(change expressed in percent of peak levels)

GERMANY GREAT BRITAIN UNITED STATES

Peak Trough Change Peak Trough Change Peak Trough Change

RATES AND EARNINGS, ALL INDUSTRY

Reference Contraction .

Hourly rates 1929 1932 —19 1929 1932 —4 1929 1932 —18
Hourlyearnings 1929 1932 —24 ... ... .. 1929 1932 —21
Weekly earnings 1929 1932 —33 ... ... .. 1929 1932 —32

Specific Contraction
Hourly rates 1930 1937 —22 1927 1934 —7 1929 1933 —23
Hourly earnings 1929 1933 —27 ... ... .. 1929 1933 —22
Weekly earnings 1929 1932 —33 ... ... .. 1929 1933 —33

HOURLY WAGE RATES, BUILDING

Reference Contraction
Skilled workers 1929 1932 —25 1929 1932 —7 1929 1932 —11
Unskilled workers 1929 1932 —25 1929 1932 —8 1929 1932 —11

Specific Contraction
Skilled workers 1930 1936 —35 1929 1934 —9 1931 1933 —17
Unskilled workers 1930 1936 —34 1929 1934 —10 1930 1933 —19

souRcE: Appendix Tables A-48 and A-51. For United States rates, see Statistical Abstract of the
United States, 1940, p. 339 (entrance rates for 13 industries).

Depression, German wage rates lagged as much as nineteen months, those

in Great Britain and the United States about one year. In this instance
the German rates obviously exhibited a particularly effective resistance
to downward revision.
The fact that there were pronounced lags in the turning points of wage-
rates in all three countries supports the thesis that such lags are inherent
in the economic nature of wage rates. Some major reasons for the occur-
rence of the lags were discussed in connection with the German experience
(see Chapter 4, p. 142 if.). Among them were the difficulty of identifying
turning points at the time of their occurrence; the unpopularity of wage-
rate cuts, and the need for a sufficiently strong change in employment and
profit conditions to make moves for wage adjustments feasible; the exist-
ence of contractual obligations extending over many months. No doubt
these factors help to explain the wage-rate lags in Great Britain and the
United States as well.
At this point we should note the relation between the strong lags in the
turning points of wage rates and the skipping of brief mild cycle phases,
which we have observed earlier. Briefer contractions might run their
294 WAGES IN GERMANY

course before wage rates begin to respond, and significant declines be

prevented by the ensuing recovery. This tendency is re-enforced in
periods of long-term upward trends in wage rates, which tend to delay
the occurrence of upper turning points and thus contribute further to
the skipping.
Besides the timing of turns in wage rates, Tab1e 71 also gives some
indication of the timing of turns in earnings during the Great Depression.
The German evidence shows greater frequency of coincidences and reduced
lags of earnings compared with lags in wage rates. There is no comparable
evidence for Great Britain for the same years. In the United States, annual
series of both average hourly and weekly earnings show coincidence at the
1929 peak and a one-year lag at the 1932 trough. The generally closer
timing of turns in weekly earnings, and to a certain extent also hourly
earnings, was brought about by the lead (or, on an annual basis, the
approximate coincidence) of turns in average hours worked compared
with turns in general business conditions. It should be noted, however,
that the observed coincidences are based on annual data. On the basis
of monthly data, earnings in the United States definitely show lags,38
and a similar situation might be surmised in the case of Germany.
Amplitudes. One of the characteristics of German wage rates, mentioned
above in connection with their conformity to business cycles, is their
strong resistance to downward adjustments. Thus, cyclical responses to
contractions consisted often merely of growth deceleration or leveling-
out into plateaus. Close scrutiny of plateau periods for which monthly
data are available reveals, however, that behind the macroscopic picture
of these plateaus there may lie mild cycles with minute amplitudes.
Appendix Table A-21 and Chart 37, for instance, show that in connection
with the 1925-26 business contraction average wage rates not only leveled
out, but actually declined by about 0.5 percent of the peak—with somewhat
larger declines in certain industries. This finding checks neatly with
Creamer's measurement of similar mild wage-rate cycles in Great Britain
and the United States.39 During the two periods when German wage
rates underwent major full cycles, their fluctuations happened to be
roughly similar. During the Gründerjahre cycle their specific advance was
about 50 percent, their decline 20 percent. During the interwar period the
specific advance between 1924 and 1930 was 68 percent, and the decline
during the Great Depression was 22 percent—all measured on an annual
basis in percent of levels at initial turns.40
The annual reference turning points marking the boundaries of the
Great Depression are set by the National Bureau at 1929 and 1932 for
38
Ibid., p. 32.
Ibid., Charts 1, 2, 3, and 5.
Measured in percent of peak-trough averages, the increase during the Grunderjahre
cycle is 45 percent for building rates and 33 percent for printing rates, the decline 20
percent and 7 percent respectively. During the interwar period average hourly rates
increased by 50 percent of the peak-trough average, and decreased 25 percent.
 GERMANY, GREAT BRITAIN, AND THE u. S. 295
each of the three countries covered by the present study. The coincidence
of these reference years enables us to compare wage changes in the three
countries. The comparative data are set forth in Table 71 which also
presents information on amplitudes during the corresponding specific
cycles. As to the amplitudes of wage rates, the most conspicuous feature
of the comparison is the mild decline in British as compared with German
and United States rates, an observation which applies both to the com-
prehensive wage measures and to the building industry wage rates. The
relatively mild decline in British rates, furthermore, is apparent in both
reference and specific cycle behavior. The question arises why British
rates declined relatively little and why, for instance, wage rates in Germany
—the country that experienced the longest lag before any rate reaction
appeared at all—finally showed significantly stronger declines. In principle,
there are many factors which could be held responsible for the compara-
tively severe decline of rates in Germany, and the mild decline in Britain.
Prominent among them might be the severity of the contraction of business
activity as measured by the cutbacks in real national income or employ-
menL Also, the fall in price levels would be expected to exert an important
influence. The following tabulation describes the decline in these factors
in the three countries between 1929 and 1932. Great Britain shows the
mildest declines in all the selected indicators of contraction severity.4'
Percentage Changes in Per Capita Real Income, Employment, Wholesale Prices,
and Retail Prices: Germany, Great Britain, United States, 1929-1932
(percent of peak levels)
Germany Great Britain United States
Per capita real income —26 —4 —37
Employment, excluding agriculture —29 —9 —25
Wholesale prices —34 —29 —32
Retail prices —21 —13 —20
SOURCE:
Per capita real income, Table 5. Employment, United Nations, Statistical Yearbook
1948, p. 80. Wholesale prices, Table 70. Retail prices, Appendix Table A-49.
For Germany and the United States, behavior of wage rates as well as
that of earnings can be compared on the basis of the data in Table 71.
Both comprehensive measures show broadly similar declines during the
Great Depression. During the reference contraction the declines in the
composite series of rates amounted to 18 to 19 percent, in hourly earnings
to a little more than 20 percent, and in weekly earnings to somewhat
more than 30 percent. For specific contractions a similar relation generally
prevailed.42 The close resemblance between wage declines in the two
countries is not found in building wage rates. As shown by the lower
Note also, however, that between 1924 and 1929 British wage rates declined, in
contrast to the increases observable in Germany and the United States.
42 The
exception is the relation between hourly rates and earnings in the United
States. However, the coverage of the hourly rates and earnings series for this country
is so different that little systematic importance can be attached to their relative ampli-
tudes in either reference or specific contractions.
296 WAGES IN GERMANY
panel of Table 71, German building rates decreased about twice as much
as United States rates.
Per capita national real income experienced a greater decline in the
United States than in Germany (see the tabulation above), and the
employment and price measures give evidence of roughly similar behavior.
Why, then, did German wages react as strongly or even more strongly to
the Great Depression than their counterparts in the United States? It
is possible, of course, that the difference is due to variations in the con-
cepts and the industrial coverage of the measures used. The steep German
wage-rate decline, on the other hand, occurred under circumstances which
could well provide an explanation for it. One of the unique features of
German wage history during the Great Depression was the "deflationary"
intervention by the government. In particular, it was Chancellor BrUning's
Fourth Emergency Decree which finally forced German wage rates down
almost to their ultimate trough levels. It may be surmised that such govern-
ment intervention brought rates to levels lower than they would have
touched if market factors alone had been the prevailing force. Bruning's
intervention may also explain the curious fact that the decline could be as
strong as it was in the very country where the delay of the wage-rate
decline was so prolonged.
Occasionally one encounters statements to the effect that wage rates
have a long-term tendency toward increasing rigidity. Since German wage
rates underwent only two substantial declines during the period 1871-
1945, broad generalizations would scarcely be acceptable on the basis of
the available evidence. Nevertheless, it is interesting to compare the two
instances of marked decline. The wage-rate declines during the post-
Gründerjahre contraction and during the Great Depression were roughly
similar in magnitude. But wholesale-price declines were somewhat milder
and production declines very much milder during the first contraction
than during the second. That is, although the limited German evidence
does not of itself indicate any "tendency" toward more pronounced
downward rigidity, it would not be incompatible with such a thesis, if
the wage decline is measured against employment and price changes.
The British experience has been examined by Dunlop,43 who finds no
evidence of long-term trends in wage-rate variability, measured either by
itself or in relation to employment and price changes. Creamer notes some
tendency toward increasing rigidity for the United States during the
period 1920-49, especially if wage reactions are compared with fluctuations
in production and employment. However, Creamer hesitates to generalize
from so brief a periodP' Our knowledge of the wage behavior in the three
countries obviously does not support any sweeping statements on long-
term trends toward increasing rigidity.
"John T. Dunlop, "Trends in the Rigidity of English Wage Rates," Review of
Economic Studies, June 1939, PP. 190 and 198.
"Creamer, op. cit., pp. 39-40.
 GERMANY, GREAT BRITAIN. AND THE U. S. 297

Wage Cycles and Labor Market Conditions

EMPLOYMENT. Since wage rates are prices prevailing in the labor market,
some major processes in this market were studied in order to determine
their effect on wage behavior. Wage-rate behavior in Germany has been
related to employment on an aggregative, as well as on an industry-by-
industry basis. As far as timing relationships are concerned, the sub-
stitution of cyclical turns in employment for those in general business
conditions does not modify the basic findings on the sluggishness in the
response of wage rates. In particular, the lag in German wage rates
behind changes in general business conditions cannot be explained by a
systematic lag of turns in employment or unemployment. Creamer has
set forth similar findings for Great Britain and the United States.45
Moreover, the differences between turning points in various industries
do not appear to be related to corresponding differences in employment
conditions in these industries, but rather to the length and expiration
dates of wage agreements.
More fruitful has been the attempt to relate the amplitude of wage rate
responses to those in employment. For Germany it has been shown that
cycles which brought only mild wage responses were those characterized
by milder declines in employment. The same situation prevailed also in
Great Britain and the United States.46 However, for none of the three
countries would the evidence support a statement to the effect that
cyclical responses of wage rates are directly proportional to declines in
employment. The declines of United States wage rates during the 1920-2 1
and 1929-32 contractions certainly bear little resemblance to employ-
ment changes during the same periods. Nor is this finding surprising;
the amplitude of wage-rate declines is obviously codetermined by other
factors, such as changes in price levels.
If comparisons between employment changes and wage-rate changes
were carried out industry-by-industry during the same cycle, the effect of
changes in the price level would be minimized. Such changes would in
fact be ruled out if the comparisons were made between the same dates.
In order to see whether, under such circumstances, wage-rate changes bear
any relation to business activity in various industries, German employ-
ment and production data were compared with wage rates, for the refer-
ence contraction 1929-32. A fair degree of correlation was established
(see Chapter 4, p. 159 and Table 39). In Great Britain, juxtaposition of the
available wage-rate and employment information by industry, for the
Great Depression, showed no significant relationship between changes
in wages and those in employment.47 Exploratory study of hourly wage
Ibid., Table 1 and Chart 5.
Ibid., pp. 12 and 26.
This observation is based on changes in wage rates, as given by E. C. Ramsbottom
in "The Course of Wage Rates in the United Kingdom, 1921-1934," Journal of the
Royal Statistical Society, 1935, pp. 665-66; and as given by Bowley, London and
298 WAGES IN GERMANY

and employment behavior in the United States suggests a low positive

correlation between the mentioned variables.48 The relationship observed
for Germany should not, of course, be interpreted as denoting a simple
causal relationship between employment and wage fluctuations. Many
of the cyclically sensitive industries, such as building and hard-coal mining,
happen also to be high-wage industries. The positive relation between
employment and wage-rate decline could thus be due, at least in part, to
the tendency of high rates to decline more than low rates.49 Similarly, the
fact that some industries with smaller wage-rate declines also experienced
relatively small employment setbacks does not necessarily support
contentions that employment changes determine wage changes, or that
wage rigidity has but slight effect upon employment. Again, it must be
understood that the low-wage consumers' goods industries may tend to
suffer less decline in wages and employment for reasons other than those
reflected in either of the two variables.
LABOR STRIFE AND GOVERNMENT ACTIVITY. In studying the German
experience, no way was found to isolate the effect of union or employer
activity on wage cycles. It was possible, however, to relate the occurrence
of labor strife—strikes and lockouts—to the lag of wage rates after lower
turning points. Strikes tended to reach their heights (in terms of man-
days lost) close to mid-expansion, that is, after the rise of living costs
began to depress the purchasing power of the hourly rate, after increased
employment began to ease competitive pressures in the labor market, and
after increased sales provided some leeway for businesses to grant wage
rises. Evidence in the United States confirms the occurrence of peaks in
labor strife well within the expansion phase. Albert Rees5° finds, on
examination of the period 1915 to 1940, that strike peaks character-
istically precede peaks in general business activity by about five months.
He reports also a lag of strike troughs behind reference troughs. Rees's
findings are in keeping with the observation in Germany of low-strike
activity during contractions, though the German evidence does not show
sufficient regularity to permit a generalization on the timing of troughs
in labor strife. For Great Britain, it is possible to establish a basically
positive general conformity of strike activity to business cycles. However,
Cambridge Economic Service, May 1947, p. 12. For employment, see Agatha L.
Chapman, Wages and Salaries in the United Kingdom, 1920-1938 (London, Cambridge
University Press, 1953), pp. 98-100. The wage-employment comparisons were hampered
by the fact that information for these two measures is rarely available for comparable
industrial classifications. Furthermore, wage rates showed relatively small declines during
the Great Depression.
"This study was based on (1) average hourly earnings and employment in twenty-one
manufacturing industries, as reported by the National Industrial Conference Board,
and (2) entrance rates for common labor and employment as reported by the Bureau
of Labor Statistics.
"This is not always true, however. Brewing, a high-wage industry, showed a small
decline in wage rates.
"Industrial Conflict and Business Fluctuations," Journal of Political Economy,
October 1952.
 GERMANY, GREAT BRITAIN, AND THE u. s. 299
the annual information examined does not show any evidence of peaks
in labor strife at or close to the mid-expansion phase of the cycle, as was
observable for Germany.51
German wage history offers many examples of effective government
intervention in wage setting. Throughout the Weimar Republic, the
settling of wage disputes by compulsory arbitration provided a tool for
government influence. Up to the onset of the Great Depression, this tool
was used largely to promote "social progress"—to iron out wage in-
equities and to support, in a moderate fashion, wage earners' demands
that they participate in the fruits of economic recovery. After the
depression was under way, however, the arbitration boards changed their
goals, attempting instead to bring about a moderate decline of wage rates.
More drastic acts of intervention by the government were embodied in
the deflationary emergency decrees of 1930 and 1931. The second effected
a reduction of wage rates to 1927 levels—leading to a total wage-rate
cut of about 10 percent. After the National Socialists took power, wages
became wholly subject to government administration, and wage rates
were stabilized at or close to depression levels. In Great Britain there was
no comparable government initiative aimed at wage setting, but the
United States government did intervene in the process. Its intervention
was initiated only after the trough of the Great Depression was reached,
and had as its objective the raising and not the lowering or stabilization
of wage levels. These differences in wage policy are reflected clearly in
the differentiated short-term trends of German and United States wage
levels (see Charts 35 and 36).
WAGES AND OTHER PAYROLL COMPONENTS. Chapter 4 sets forth the
attempt to establish for Germany the separate contributions to changes
in the total industrial payroll, of wage rates, excess of hourly earnings
over rates, average hours worked, and employment. Tabulated below
are the percentage changes in each of the above variables and the per-
centage contribution made by each factor to the decline of the total payroll
for the years 1929-32 in Germany and in the United States. Unfortunately,
comparable information is not available for Great Britain.
A striking feature of this evidence is the close resemblance of German
and United States experience in both the extent and structure of payroll
declines. During a contraction of roughly similar magnitude—as measured
by employment—the changes in the payroll and its components, and the
contribution of the various components to the total decline were very
similar in the two countries. For both countries about 60 percent of the
total payroll decline is attributable to the drop in employment and only
18-20 percent to the change in wage rates.52
51
See data on industrial disputes, International Labour Office, Yearbook of Labour
Statistics, 1937, and 1951-52.
52 This
statement refers, of course, to the numerical contribution of these factors
only, not to their causal importance.
300 WAGES IN GERMANY

Germany United States

Percentage Changes, 1929-32
1. Employment —41 —38
2. Average hours worked —9 —13
3. Hourly earnings —25 —21
4. Excess of earnings over rates —8 —4
5. Wage rates per hour —18 —18
Payroll 59

Contributiona to Payroll Decline, 1929-32

(percent)
1. Employment 59.7 57.2
2. Average hours worked 11.9 17.4
3. (Hourly earnings)b (28.4) (25.4)
4. Excess of earnings over rates 10.8 5.3
5. Wage rates per hour 17.6 20.1
100.0 100.0
a The method used in the derivation of these measures is briefly explained in Chapter
4, footnote 59.
b Hourly earnings (line 3) are subtotals of lines 4 + 5.

WAGES AND PRICES

German wage rates were found to be a rather insensitive type of price,
with regard both to timing and amplitudes. They turned later and declined
less than most major categories of wholesale prices—in fact later and less
than living costs.
The later and lesser cyclical reactions of wage rates as compared with
wholesale prices can be said to have occurred in all three countries. On
an annual basis, the specific peaks of wage rates, closest to the 1929 turn
in general business conditions, were reached in the following years:
Hourly Wage Rates Wholesale Prices
Germany 1930 1928
Great Britain 1927 1924
United States 1929 1928

As for amplitudes, a comparison of percentage declines, measured from

peak levels, between the reference turning points of 1929 and 1932 stands
as follows:
Hourly Wage Rates Wholesale Prices
Germany —19 —34
Great Britain —4 —29
United States —18 —32

The lesser variability of wage rates can be demonstrated also by a com-

parison of wage and price behavior during periods when wage rates under-
went the minute declines observable in monthly records, described
previously. While German wage rates declined by about 0.5 percent in
connection with the 1925-26 contraction in general business conditions,
 GERMANY, GREAT BRITAIN, AND THE U. S. 301
wholesale prices dropped by 10 percent or more. Similarly, in the two
mild contractions of the 1920's (1924-26 and 1927-28 in Great Britain;
1923-24 and 1926-27 in the United States) both countries experienced only
minute indications of cyclicity in wage rates but clear reactions in whole-
sale prices.53 Thus, although the differing compositions of the national
indexes preclude a quantitative comparison of, for instance, the change
in wage-price ratios in the three countries, the finding that there is relatively
slight variability in wage rates, as compared with wholesale prices, is
firmly established.
The cyclical insensitivity of wage rates to downward pressures appears
less pronounced in comparison with retail prices for goods and services.
The German experience during the years 1924-32 shows that in specific
and reference expansions wages went up more, and in contractions declined
less, than living costs. Somewhat analogous relations prevail also, for the
same period, in Great Britain and the United States. The relation of money
wages and living costs, and the consequent changes in real wages, can be
observed in Charts 35 and In all three countries the prevailing
tendency during the period 1924-32 for wage levels was to increase or hold
their own in the face of mild decreases in living costs and, when living
costs declined radically, to show smaller declines than the latter. The
relations of living costs and wage rates in the three countries for the period
1924-38 can be followed in Chart 35. They are, of course, reflected in the
movements of real wages, to which we shall now turn.

REAL WAGES
Conformity. Real wage rates in Germany before World War I showed
occasional evidence of inverse conformity to business cycles, particularly
when wage rates were fixed for a number of years—as in the printing
industry—and when real wage fluctuations were thus due entirely to
cost-of-living changes. On the whole, however, and particularly if the later
and more sensitive monthly information of the interwar period is used
as a guide, positive conformity of real wage rates with changes in business
conditions must be regarded as the norm. Appraisal of conformity is
The price indexes used in these comparisons are the Sauerbeck Index for Great
Britain, and the Bureau of Labor Statistics Index of Wholesale Prices for the United
States. Note that the declines of British wholesale prices in the contractions of 1924-26
and 1927-28 are not separated by any significant specific expansion.
"In Great Britain wage rates changed but little between 1924 and 1932. But the
slight increase, and the leveling-out through 1927, were accompanied by a living-cost
decline of about 8 percent. And during the remainder of the period both wage rates
and living costs declined, the latter exhibiting significantly larger movements. In 1932
wage rates were about 5 percent below their 1925 levels, living costs about 19 percent
below. Also in the United States, wage rates, hourly earnings, and living costs showed
relatively shallow movements up to 1930. In 1924 and 1925 living costs rose, and during
1926 through 1930 they decreased, in the face of practically stable wage rates. From
1930 to 1932 or 1933, the decrease in living costs exceeded the declines in wage rates.
(Based on wage rate and earnings data as reported by Creamer, op. cit., Table A. The
cost-of-living data are those compiled and published by the Bureau of Labor Statistics.)
 --

302 WAGES IN GERMANY

complicated by long lags in real wage rates, but if allowance is made for
timing relations, the preponderance of positive conformity is clear enough.
In the case of real earnings, even the pre-1913 evidence tends to show
positive conformity, an impression that is confirmed by the behavior of
earnings during the interwar period.
The positive conformity of German money and real wage rates to
business cycles lends support to the critics of J. M. Keynes' thesis that
money and real rates tend to move, cyclically, in opposite directions.
As Chart 14 shows, the behavior of money and real wage rates was
characterized by positive co-variance rather than by inverse variation,
especially during cyclical rather than intracyclical movements.
A tendency toward positive conformity in the cyclical fluctuations of
money and real wage rates has been found also by Dunlop55 and by
Tarshis56 for Great Britain and the United States. Their findings have
been doubted by Ruggles.57 Dunlop makes the following summary state-
ment: "Statistically, real wage rates generally rise with an increase in wage
rates, rise during the first period after the peak, and then fall under the
pressure of severe wage reductions."58 This would describe quite well the
behavior of German wage rates during the 1926-32 cycle. It would describe
only approximately, however, the German wage behavior of the 1924-26
cycle, since the eventual decline of real wage rates was brought about by
rising living costs rather than by severe money wage-rate reductions, and
since the decline occurred only after the next expansion in business had
begun. Ruggles has expressed, on a priori grounds, some expectations
about the behavior of money and real wage rates, as follows "In a
mild recession or in the early phases of a major depression it would be
quite possible for the real wage rate to decline, largely because of the lag
in the response of rent to changes in income. In a deeper, more prolonged
depression, however, it seems likely that the real wage rate would rise.
Rent becomes more variable than money wage rates in the longer run,
so that the only components of expenditures whose prices remain less
flexible than money wage rates are a few public utilities." It is clearly
outside the scope of this study to evaluate the theoretical merits of Ruggles'
expectations. It appears, however, that the behavior of German wage
rates does not support his view. Typically, German real wage rates rose
during mild contractions and during the early phases of severe contractions,
but declined as the depression deepened.
John T. Dunlop, "The Movement of Real and Money Wages," Economic Journal,
June 1938, pp. 413-434.
56Lone Tarshis, "Changes in Real and Money Wages," Economic Journal, March
1939, pp. 150-154.
Richard Ruggles, "The Relative Movements of Real and Money Wage Rates,"
Quarterly Journal of Economics, November 1940, pp. 130-144.
op. cit., p. 434.
58

Richard Ruggles, "The Nature of Price Flexibility and the Determinants of

Relative Price Changes in the Economy," Business Concentration and Price Policy
(Princeton University Press for National Bureau of Economic Research, 1955), p. 495.
 GERMANY, GREAT BRITAIN, AND THE U. S. 303
Timing. The lag of turning points in German real wage rates behind
those in general business conditions is still more pronounced than the lag
in money wage rates. At the 1925 peak the lag in money wage rates was
eight months, that in real wage rates twelve months. At the 1929 peak,
a money wage lag of thirteen or twenty months occurred, as compared
with a real wage lag of about two or even two and one-half years.6°
It is not clear whether a similar extension of lags occurred also at troughs.
Following the 1925-26 contraction, money and real wage rates showed
about the same lag. After the Great Depression, on the other hand, there
occurred a slight upturn of real rates between July 1932 and February
1933, while money rates were still declining. It is difficult, however, to
distinguish cyclical and shorter-term movements at that point; and from
1933 on, the stabilization of money wage rates under the Nazis led to an
abnormally extended decline in real wage rates—with no revival up to the
end of World War II.
The lag of real wages behind money wages can be observed also in
German hourly earnings. Annual series of hourly real earnings, for
instance, show a lag of two years, while the peak of hourly money earnings
coincides with the business cycle peak. When we look at weekly earnings,
however, we find that both the money and real wage series turn together
with general business conditions—a result of the important role played
by hours in the determination of weekly earnings.
The more pronounced lag of real as compared with money wages can
be shown to prevail in all three countries at the 1929 peak. The peak in
real wage rates was reached in Germany, Great Britain, and the United
States in 1931, that is, two years after the reference turning point and one
or more years after the turn of money rates in each country. (See Table
72 and Charts 36 and 37.) Real wage rates also lagged considerably behind
the 1932 reference trough. However, while real wages lagged behind
money rates in Germany and Great Britain, they turned in the same year
in the United States. Comparison of the lower panels of Tables 71 and 72
indicates the extended lags of real wage rates as compared with money
rates for skilled and unskilled building workers in the three countries.
The findings are similar to those based on the comprehensive measures.6'
60 At the 1925 peak, the break of the sharp wage rise and the leveling-out into the

intermediate plateau in November was regarded as the upper turning point of money
wage rates. Real wage rates show a clear peak in March 1926. For the onset of the
Great Depression, the alternative figures mentioned refer to the beginning (P1) and end
(P2) of their peak plateau in the case of money rates, and to the first and second double
peak in the case of real rates (see Chart 14).
61
All series show longer lags of real rates at both the peak and the trough. With the
exception of one instance (the turn of skilled workers' rates in the United States after
the 1929 peak), all real rates had their peak in 1931, that is, two years after the reference
peak and one or two years later than money rates. At the subsequent trough the lag of
real rates was two years longer than that of money rates in the United States, three
years longer in Great Britain, and at least eight years longer under the extraordinary
conditions prevailing in Germany.
 304 WAGES IN GERMANY

TABLE 72
Percentage Change of Real Wages during the Great Depression,
Germany, Great Britain, and the United States

GERMANY GREAT BRITAIN UNITED STATES

Percent Percent Percent

Peak Trough Change Peak Trough Change Peak Trough Change

RATES AND EARNINGS, ALL INDUSTRY

Reference Contraction
Hourly rates 1929 1932 +4 1929 1932 +10 1929 1932 +3
Hourly earnings 1929 1932 —4 ... ... ... 1929 1932 —1
WeekJy earnings 1929 1932 —15 ... ... ... 1929 1932 —15

Spec(flc Contraction
Hourly rates 1931 1944 —18 1931 1937 —4 1931 1933 —3
Hourly earnings 1931 1936 —6 ... ... ... 1931 1932 3
Weekly earnings 1929 1932 —15 ... ... ... 1929 1932 —15

HOURLY WAGE RATES, BUILDING

Reference Contraction .

Skilled workers 1929 1932 —4 1929 1932 +7 1929 1932 +10

Unskilled workers 1929 1932 —5 ... 1932 +5 1929 1932 +11

Specific Contraction
Skilled workers 1931 1944 —30 1931 1937 —6 1933 1935 —7
Unskilled workers 1931 1944 —29 1931 1937 —5 1931 1935 —6

sOURCE: Appendix Tables A-50 and A-52.

Also in hourly earnings, the stronger lag of real wages as compared with
money wages is in evidence. The peak of hourly money earnings in 1929
coincided with that of general business conditions, both for German and
for United States wage earners; but hourly real earnings did not turn
until 1931. The longer delay in the turn of hourly real earnings appeared
in Germany also at the end of the Great Depression. However, in the
United States hourly real earnings hit their trough in 1932—one year
before money earnings. For Great Britain, there are no earnings data for
the period under review.
The cyclical timing of weekly real earnings, finally, reflects the responsive
behavior of average hours worked, of the output of piece work, and of the
incidence of premium payments, rather than changes in basic rates.
In Germany and the United States weekly earnings, both money and real,
turned at the reference peak year of 1929. Weekly real earnings also turned
in the reference trough year of 1932. However, in the United States the
real earnings trough occurred one year earlier than that of money earnings,
reflecting the anomaly previously reported for average hourly earnings.
Amplitudes. Even a casual comparison of the record of money wages
 GERMANY, GREAT BRITAIN, AND THE U. s. 305
(Charts 33 and 34) with that of real wages (Charts 38 and 39) reveals
clearly the smaller cyclical amplitudes of the latter. Let us look more
closely at real-wage behavior during the Great Depression. In all three
countries we find the smaller amplitudes of real wages (see Tables 71
and 72 and Charts 35 and 36). There are, however, a number of special
situations which modify the generalization, to be considered below.
Between the terminal years of the 1929-32 slump, money wage rates in
all three countries show uniquely steep declines. Yet living costs declined
still faster, so that the indexes of hourly real rates show actual increases
in the three countries (and building rates, increases in Great Britain and
the United States). These increases indicate, of course, a low degree of
responsiveness of real wage rates to the decline in business conditions.
The described increases of real wage rates, between the reference turns,
do not imply that real rates were depression-proof. Declines in real rates
did occur, albeit with delay. However, the comprehensive rate indexes,
as well as the series describing changes in building rates, all show specific
real rate decreases, which are smaller than the comparable decreases in
money rates.
The general observation that real wages have smaller amplitudes than
money wages holds also for average hourly and weekly earnings. Com-
parative information is available for Germany and the United States
only. In these two countries real wages as well as money wages experienced
declines of very similar magnitudes; the weekly real earnings declines
in both countries amounted to 15 percent.

Wages during War and Inflation

GENERAL
Chapter 5 of this study dealt with the behavior of German wages under
extraordinary circumstances, such as the two world wars and the Great
Inflation. The purpose of that chapter was to describe the unique deter-
minants of wage behavior in Germany during those episodes. Here, the
task is to compare the major findings with corresponding findings for
Great Britain and the United States. Apart from the difficulty of sum-
marizing the detailed description presented for Germany, it might seem
altogether futile to undertake a comparison of individual national situa-
tions. Nevertheless, despite the unique character of special events in each
of the countries, their national histories have some features in common.
The two world wars involved all three countries. The Great Inflation
and subsequent stabilization were but the specific ways in which Germany
underwent its postwar adjustment; and the development of National
Socialism up to World War II was, to be sure, the peculiar course Germany
followed in moving out of the Great Depression and into rearmament.
Thus the course of events in Germany had historic and economic counter-
parts in British and American experience, although, of course, with
306 WAGES IN GERMANY

marked differences from country to country. Germany lost the two world
wars; Great Britain and the United States were victors. Germany was
unable to control her currency depreciation after World War I; the other
two countries managed to do so. Germany's economic expansion after
1932 was dominated by a rearmament drive, while the expansions in
Great Britain and the United States, at least for the greater part of the
period before 1939, were predominantly in the civilian sector of the
economy.
The broad similarities as well as the broad differences in the three
countries' experiences during these unusual periods must be reflected in
their wage histories. This, at least, is a surmise worth testing. For this
purpose, the major similarities and differences in the economic fortunes
and in the wage behavior of the three countries will be analyzed below.
WORLD WAR I
Money wages during World War I increased each year in Germany,
Great Britain, and the United States. In all three countries wage levels
went up more sharply in the second half of the war than in the first. And
in all three countries earnings tended to rise more than wage rates. How-
ever, there were marked differences in the extent of wage increases, the
rise being sharpest in Germany and mildest in the United States. The
differences, already discernible between 1914 and 1916 levels, became
more pronounced during the latter part of the war.
The information underlying the above generalizations is assembled in
Table 73. Data on wage rates of skilled building workers and earnings of
coal miners, available for all three countries, permit comparisons of
tolerably similar categories. The comprehensive wage indexes presented
in the last three columns of the table show so many differences in concepts,
coverage, and composition that they cannot be directly compared. They
do, however, shed some light on the representativeness of the more
narrowly defined measures and, at least for Germany and the United
States, provide some rough indication of comparative wage trends.
Over the years 1914 to 1918 wage rates of building workers rose by about
a quarter in the United States, by 60 percent in Britain, and by about
100 percent in Germany. Average weekly earnings rose more sharply in
all instances, and again the most marked increase occurred in Germany,
the least marked in the United States. Indications are that between the
first and the last year of the war, average earnings increases in each
country may have exceeded those in its building rates by roughly 30
percentage points.
The order of wage rises corresponds with the extent to which the
countries were affected by the war. Germany, closest to actual battle-
grounds and ultimately defeated, experienced the strongest rises. The
United States, far removed from the scenes of the conflict, entering it later
than the other participants, and being on the victorious side, showed
 TABLE 73
Money Wages in Germany, Great Britain, and the United States, 1914-1923

WAGE RATES OF S KILLED EARNINGS OF COAL MINERS COMPREHENSIVE I NDEXES

BUILDING WORKERS, PER HOUR PER WEEK OR SHIFT WAGES PER WEEK OR DAY

Great United Great United Great United

Germany Britain States Germany, Britain, States, Germany, Britain, States,
weekly shift weekly earnings weekly weekly
or rates rates earnings
Year (May) (July) (May) (Year) (Year) (Year) (July) .

1914 bOa 100 100 100 lOOb 100 100C 100 100
1915 102& 103 101 110 115b 106 121C 108 100
1916 116 108 104 130 129b 116 141C 118 111
1917 152 123 110 160 136b 144 194C 135 125
1918 20D 160 122 200 195b 178 235c 175 152

1919 ... 188 139 340 224b 207 360d 210 175
1920 703 259 186 794 260 233 820b 252 213
1921 859 252 189 1,206 246 250 920b 250 194
1922 3,171 191 179 4,IOOb 154 250 4,220h 188 186
1923 237,627 183 197 2,762,lOOb 160 257 2,781,300b 168 200
a April.
b July.
Average of March and September.
d Third quarter.
SOURCE:
Building Rates
Germany (data refer to Berlin): 1914-18: Waldemar Zimmermann, "Die Veränderung der
Einkommens- und Lebensverhältnisse der deutschen Arbeiter durch den Krieg," in Die Einwirkung
des Krieges auf Bevolkerungsbewegung, Einkommen und Lebenshaltung in Deutschland, Wirtschafts-
und Sozialgeschich:e des We It krieges (Carnegie Foundation for International Peace, Stuttgart,
Deutsche Verlags-Anstalt, 1932), p. 398. For 1920-23: Robert Kuczynski, "Postwar Labor Condi-
tions in Germany," U. S. Bureau of Labor Statistics, Bul. 380, pp. 125-27. Weekly rates given are
adjusted for change of hours. (Adjustment factor: 51 to 46, prewar to postwar hours as reported
in source.)
Great Britain: Bowley index, as given in C. E. Lyon, British Wages, U.S. Department of Com-
merce, Trade Promotion Series, No. 42, 1926. Weekly rates given are adjusted for change of hours
in London. (Adjustment factor: 50 to 44, prewar to postwar hours as reported by A. L. Bowley,
"Wages, Earnings and Hours of Work, 1914-1947, United Kingdom," London and Cambridge
Economic Service, Special Memorandum No. 50, p. 11.) The change of hours in London is fairly
representative for that in the cities included in the wage-rate index. See A. L. Bowley, Prices and
Wages in the United Kingdom, 1914-1920 (Oxford, Clarendon Press, 1921), pp. 116-20.
United States: Historical Statistics of the United States, 1889-1945, p. 69, Series D 154 and D 156.
Miners' Earnings
Germany (data refer to Ruhr): Wirtschaft und Statistik, 1925, "Zahien zur Geldentwertung in
Deutschland, 1914 bis 1924," p. 41.
Great Britain: 1914-19, A. L. Bowley, Wages and Prices in the United Kingdom, 1914-1920, p. 150.
(Figures refer to July.) For 1920-23, A. L. Bowley, Wages and Income in the United Kingdom since 1860
(Cambridge University Press, 1937), p. 22.
United States: Paul H. Douglas, Real Wages in the United States, 1890-1926 (Houghton Muffin,
1930), p. 162.
Comprehensive Indexes
Germany: 1914-18, daily earnings of male workers in twelve industries, average for March and
September; see Appendix Table A-37. For 1919-23, weekly wage rates of skilled workers in eight
industries, see Appendix Table A-42, col. 3.
Great Britain: Wages in eleven occupations, predominantly weekly rates. Bowley data as reported
in C. E. Lyon, op. cit., p. 52. Where range was given, mid-point was selected.
United States: Full-time earnings in all manufacturing industries, see Douglas, op. cit., p. 130.
307
308 WAGES IN GERMANY

smaller increases. The connection between the two sets of facts lies prob-
ably in the extent to which direct participation and all-out effort increased
inflationary tendencies—in the absence of direct price and wage controls.

TABLE 74
Cost of Living in Germany, Great Britain, and the United States, 1914-1923

Germany Gr eat United States

Ministry of
Year Labour Bowley

I II

1914 100 100 100 100

1915 125 125 120 101
1916 165 145 135 108
1917 245 180 160 128
1918 304 205 180 150

1919 403 210 ... 172

1920 988 232 ... 200
1921 1,301 219 ... 178
1922 14,602 184 ... 167
1923 15,437 bill 169 ... 170

The adjusted (Bowley) index takes into account the increasing proportion of income
spent on food and clothing during the years 1914 to 1918. It assigns progressively
larger estimated weights to these items. The unadjusted (Ministry of Labour) index
keeps weights constant throughout. See A. L. Bowley, Prices and Wages in the United
Kingdom, 1914-20 (Cambridge University Press, 1937), pp. 74-75.
SOURCE:
Germany: Appendix Table A-41, col. 2, put on base 1914 = 100.
Great Britain: For 1914-18, from Bowley, op. cit., p. 106. For 1919-23, see Charles
E. Lyon, British Wages, p. 52 (base shifted to 1914).
United States: Historical Statistics of the United 1 789-1945, p. 236. Series
L 41 (B.L.S. Index, all items), base shifted to 1914 = 100.

The available wholesale price indexes are too different in composition

and coverage to permit comparisons in the wartime changes of price
levels.62 The movements of retail prices, as represented by cost-of-living
indexes, exhibit indeed the same order, with regard to price increases, as
that observed for wages. Table 74 shows that living costs in the United
States increased by one-half, in Great Britain they doubled, and in Germany
trebled.
62
The German index comprises thirty-eight commodities—eighteen foods and
twenty industrial raw materials (see Wirtschafl and Statistik, 1925, "Zahlen zur Geldent-
wertung in Deutschland 1914 bis 1923"). For Great Britain there is the Board of Trade
Tndex covering forty-seven articles—twenty-five foods, six coal and metals, six textile
raw materials, and ten miscellaneous; the Economist Index covering forty-four raw
materials; and the Statist Index covering forty-five raw materials. (See U.S. Bureau of
Labor Statistics, Bul. 284, pp. 26 1-62, 270, and 276-78.) For the United States there are
various indexes, the most comprehensive being that of the Bureau of Labor Statistics,
which during the war years covered between three hundred and four hundred com-
modities. See Ba!. p. 284, p. 109 and Bul. 200, p. 8).
 GERMANY, GREAT BRITAIN, AND THE U. S. 309
Whatever the merits of the foregoing speculation on the connection
between money wage levels and degree of war effort, the relation between
retail price changes and wage changes is close and significant. The extent
to which retail price changes modified the comparative movement in
money wages will be examined next. The reader is asked to bear in mind
that the quotients describing the relation of money wage changes and cost-
of-living changes in comparison with prewar levels are particularly
difficult to interpret in times of rapid shifts such as wars. Real wages
under such circumstances, are indeed far from "real." It is, therefore,
the broad changes rather than the specific numerical values of the real
wage quotient with which we are here concerned.
In contrast to the direction of money wage changes, the general trend
of real wages was downward. Specifically, the data assembled in Table 75
show lower real wage levels in 1918 than in 1914 for all series except
miners' shift earnings in the United States and possibly in Great Britain—
depending upon the cost-of-living index used as deflator.63 In each country,
real-rate declines are more pronounced than earnings declines, a fact
which follows from the smaller rise of money rates as compared with
earnings. The changes in real wage rates range from —35 percent to —3
percent, in earnings from —32 percent to + 13 percent. While in money
wages Germany experienced the most pronounced and the United States
the smallest rises, in real wages the decline was largest in Germany and
smallest in the United States.64
The comparison of real wage movements in the three countries illus-
trates—in different ways than comparison of money wage movements—the
close relation between wage behavior and the political and economic
developments of each country. The extraordinary decline of real wages in
Germany during World War I was traced, in Chapter 5, to the long-
drawn-out war experience, to the proximity of military operations and,
most important, to the military disasters that befell that nation. The
somewhat smaller decline in British real wages would appear plausible
in the light of the greater protection from war damage and the victorious
outcome of the conifict. The remarkably favorable showing of the
United States, with earnings levels approximately maintained or even
increased, must be explained by the same factors.

POSTWAR ADJUSTMENTS
The behavior of money wages, in the five years following World War I,
is dominated by the fate of the respective price levels in the three
countries. The major differences are to be noted between countries with
rapidly depreciating currencies and those that underwent deflation or
63 Note, however, that the weekly earnings decline in the United States was negligible.
64 This is true for each of the selected wage types if the cost-of-living indexes pub-
lished by the respective governments are used as deflators. Real rates in building appear
to be an exception, if A. L. Bowley's adjusted living-cost index is used to deflate the
British wage data.
 TABLE 75
Real Wages in Germany, Great Britain, and the United States, 1914-1923

WAGE RATES OF SKILLED EARNINGS OF COAL MINERS COMPREHENSiVE INDEXES

BWLDING WORKERS, PER HOUR PER WEEK OR SHIFT WAGES PER WEEK OR DAY

Germany Great Britain United Germany Great Britain United Germany Great Britain United
States Shift Earnings States Weekly Rates States
Weekly Weekly Earnings Weekly
I II Earnings I 11 Earnings or Rates I II Earnings
Year (May) (July) (July) (May) (Year) (Year) (Year) (Year) (Year) (July) (July) (Year)

1914 bOa 100 100 100 100 lOOb lOOb 100 bOOC 100 100 100
1915 796 82 86 99 87 92b 96b 108 96C 86 90 102
0 1916 71 73 80 96 80 88b 96b 108 87c 80 87 104
1917 60 68 77 85 67 76b 85b 112 79c 75 84 98
65a 78 89 83 68 95b 108b 113 77c 85 97
1918 99

1919 ... 90 ... 81 88 107b ... 116 90d 100 ... 101
1920 55 112 ... 89 83 112 ... 113 74e 109 ... 103
1921 115 ... 105 95 112 ... 141 73e
87 114 ... 110
1922 92 104 ... 106 74b 84 ... 151 72e 108 ... 113
1923 56 108 ... 115 51b 95 ... 153 62e 99 ... 119

a April. (Table 73) and cost-of-living indexes (Table 74 and sources

b July. thereto), with this exception: Real earnings in the United States
Daily earnings averages for March and September. for coal miners and all manufacturing workers were taken directly
d Weekly rates, third quarter. from Paul H. Douglas, Real Wages in the United Stares, 3890-1926,
Weekly rates, July. pp. 130 and 162. For Great Britain columns (i) are deflated by
souica: The series for real wages are derived from money wages the Ministry of Labour index, columns (is) by the Bowley index.
 GERMANY, GREAT BRiTAIN, AND THE U. S. 311
managed to maintain their price levels. Germany suffered one of the most
conspicuous sieges of hyperinflation, as a comparison of its money wage
levels with those of Great Britain and the United States amply demon-
strates. Some differences between price behavior in Great Britain and in
the United States are worth mentioning. Although in comparison with
prewar conditions both British and United States retail price levels in
1923 (measured by living costs) are about equally high (roughly 70 percent
above 1913), in Great Britain the level was attained after rapid price
rises up to 1920 and subsequent declines, while the amplitudes of price
fluctuations in the United States were appreciably milder (see Table
74). These conditions are reflected in the differential movement of money
wages in the two countries. Between 1920 and 1923, for instance, a sharp
decline in British wage rates and earnings compares with an actual increase
in the corresponding measures in the United States. The decline in British
wages is obviously related to the deflationary price behavior during this
period. However, the increase in United States wage levels occurs despite
a reduction in living costs. Thus, factors other than the fate of the currency
must have affected money wage movements. These factors will be reflected
more clearly in the behavior of real wages.
Table 75 contains real wage information for the series introduced in
this section, up to the year 1923. The prewar base is maintained, since it
offers a more meaningful standard for comparison than, say, the immediate
postwar years 1919 or 1920.65 Comparisons of real wage movements
relative to prewar levels are startling. In 1923, for instance, real wages in
Germany were 44 to 49 percent below prewar levels, in Britain from 5
percent below to 8 percent above, and in the United States 15 percent to
53 percent above 1913—the position within the range depending upon the
particular wage category in each instance. Thus, in the adjustment period
after World War I the tendencies that developed during the war itself
were continued. The political and economic conditions after the war
obviously were consequences of the differentiated war experiences. They
affected real wage levels of the three countries in much the same directions
as war conditions had affected them previously. This perpetuation of
fortunes and calamities was to lead, also in subsequent years, to differ-
entiated courses in general economic trends and in wage behavior.

THE WAY OUT OF THE GREAT DEPRESSION

Wage behavior from the trough of the Great Depression to the outbreak
of World War II differed substantially among the three countries (see
Table 76). Money wage rates of building workers declined in all three
countries from 1932 to 1933, but the decrease in German building rates
The year 1914 is here used as indicating prewar levels, in order to maintain com-
parability with the measures describing wage behavior during World War I. The use of
the year 1913 as basis of comparison would not affect any conclusions, since neither
wages nor prices changed substantially between these two years.
 TABLE 76
Money Wages in Germany, Great Britain, and the United States, 1932-1945

HOURLY WAGE RATES OF SKILLED EARNINGS OF COAL MINERS PER WEEKLY RATES OR EARNINGS, COMPREHENSIVE
BUILDING WORKERS WEEK OR SHIFT WAGE INDEXES

Germany Great Britain United States Germany Great Britain United States Germany Great Britain United States
Hard Coal All Coal Bituminous Coal Average Average Average
Year Per Shift Per Shift Per Week Earnings Rates Earnings Earnings
(1932 = 100)
1932 100 100 100 100 100 100 100 100 ... 100
1933 79 98 97 101 99 104 102 99 ... 98
1934 79 98 98 102 100 130 110 99 ... 108
1935 79 100 99 103 101 141 112 101 ... 118
1936 79 103 102 104 110 163 117 103 ... 128
1937 79 105 109 105 116 169 121 107 ... 141
1938 79 108 119 106 125 150 126 110 ... 131
1939 80 108 120 118 131 172 131 112 ... 140
1940 81 116 121 123 147 178 135 125 ... 148
— 1941 82 125 126 121 168 222 144 135 ... 173
1942 82 127 133 123 196 252 145 145 ... 215
1943 82 134 133 124 215 299 146 152 ... 253
1944 82 133 134 ... 246 369 144 163 ... 270
1945 ... 143 137 ... 262 376 ... 173 ... 260
(1939 100)
1939 100 100 100 1._O 100 100 100 100 100
1940 101 108 101 104 112 104 103 112 130 106
1941 103 115 105 103 128 129 110 121 142 124
1942 103 118 111 104 150 147 110 130 160 154
1943 103 124 112 106 164 174 111 136 176 181
1944 103 123 112 ... 187 215 110 146 182 193
1945 ... 132 114 ... 200 219 ... 155 180 186

a October 1938. United Kingdom from 1860 (Cambridge University Press, 1937), p. 22.
SOURCE For 1937-45, A. L. Bowley, "Wages, Earnings and Hours of Work,
Building Rates: Appendix Table A-51. Data for Germany refer to 1914-1947, p. 10.
April, those for Great Britain to September, and those for the United United States: Bureau of Labor Statistics, Handbook of Labor
States to May. Statistics, 1947, p. 80.
Miners' Earnings: Comprehensive Indexes: See source to Appendix Table A-48. Data
Germany: Handbuch 1928-44, p. 468. for Great Britain refer to September.
Great Britain: For 1932-37, A. L. Bowley, Wages and Income in the
 GERMANY, GREAT BRITAIN, AND THE U. S. 313
was particularly steep, owing to especially unfavorable conditions in the
building industry.66 Fundamental differences in wage-rate behavior
emerge from 1933 onward. Between that year and the outbreak of World
War II, building wage rates in Germany were stabilized close to their
depression levels, while in Great Britain they increased by 10 percent, and
in the United States by more than 20 percent—all measured from their
specific troughs. Comparison with levels at the reference turn of 1932
further increases the differences between the experience of the three
countries. The reasons for these extreme differences may be traced to
government interference in wage determination. In Germany the policies
of the National Socialists aimed at wage stabilization, while in the United
States the policies of the New Deal tended to encourage wage-rate increases
both directly and indirectly by furthering collective bargaining and the
growth of unionism.
The differential behavior of wage rates is reflected in differences among
earning trends. Coal miners' earnings in 1939 were 18 percent above 1932
levels in Germany, 31 percent above in Great Britain, and 72 percent
above in the United States. The data used would, however, appear to
favor the United States experience, where earnings are measured per
week, while in Germany and Great Britain they are measured per shift.
Number of shifts worked per man and week are cyclically sensitive and
thus, in expansions of general business activity, tend to boost the rise of
average weekly earnings above that of average shift earnings. For the
comprehensive weekly earnings measures, direct comparison is possible
only between Germany and the United States. The comparison shows
stronger gains for the United States than for Germany, whether 1932 or
1933 is used as a point of departure. However, the difference between the
composite indexes (weekly earnings for both countries) is far less pro-
nounced than that between the reports on miners' earnings (shift earnings
for Germany, weekly earnings for United States). Part of the explanation
must lie in the fact that German wage policy under National Socialism
brought about a significant extension of working hours per week, while
hours in the United States actually declined.67 Moreover, in Germany the
increasing resort to payment according to results helped to raise average
weekly earnings.
The comparative behavior of money wages should be evaluated in the
light of concomitant changes in price levels, and especially in living costs.
Reference to Table 77 shows that for the period 1932-39 changes in living
costs can provide only part of the explanation of wage behavior. Living
costs declined in all three countries between 1932 and 1933, and rose
thereafter—just as wage rates did. But the rises of living costs in the three
The comprehensive index of German wage rates shows a decline of only 4 percent
between these dates.
67 In
Germany average weekly hours in manufacturing establishments increased by
17 percent between 1932 and 1939, while in the United States they declined fractionally
between the two dates (see Table 69).
314 WAGES IN GERMANY

TABLE 77

Cost of Living in Germany, Great Britain, and the United States,

1932-1945

Germany Great Britain United States

Year I II I II I II

(1932 100)
1932 100 100 100 100 100
1933 98 96 95
1934 100 98 98
1935 102 99 101
1936 103 ... 102 102
1937 104 109 107 105
1938 104 109 108 103

1939 105 110 110 111 102

1940 108 113 122 129 103
1941 110 116 138 141 108 109
1942 113 119 138 151 119 121
1943 115 120 138 155 127 130
1944 117 123 139 159 129 133
1945 1226 141 161 132 136

(1939 100)

1939 100 100 100 100 100

1940 103 103 110 117 101
1941 106 106 125 127 106 107
1942 108 108 125 136 117 118
1943 110 110 125 140 124 127
1944 112 112 126 143 126 130
1945 1166 128 145 129 133

August.
SOURCE:
Germany: See Table 68, both for the official index (i) and for the index adjusted for
admitted bias (u).
Great Britain: Official index (i) as published contemporaneously by Ministry of
Labour, and reprinted in London and Cambridge Economic Service Bul. iv, Nov. 10.,
1947, p. 129. Adjusted index (u), as computed by R. G. D. Allen, ibid., But. i, February
1949, p. 16.
United States: Official index (i) as published by the U. S. Bureau of Labor Statistics,
see Monthly Labor Review, May 1952, p. 615. Index (ii) adjusted for underestimate
reported by the Mitchell Committee in Prices and the Cost of Living in Wartime—An
Appraisal of the Bureau of Labor Statistics Index of Cost of Living 1941-44 (Report of
the technical committee of the President's Committee on the Cost of Living, Wesley
C. Mitchell, chairman, Simon Kuznets, and Margaret 0. Reid, June 15, 1944).

countries during the subsequent period bore scant relation to wage-rate

behavior. Whereas in the United States, between 1933 and 1939 building
wage rates, for instance, rose more strongly than in the other two countries,
living-cost increases in this country were comparatively low. The largest
increase of living costs occurred in Great Britain, the country that held
an intermediate rank with regard to wage-rate rises. This situation
 GERMANY, GREAT BRITAIN, AND THE U. S. 315
emphasizes the favorable wage trends in the United States, shown in the
following comparison of real wages.
Real wage rates in Germany declined between 1932 and 1939, approxi-
mately maintained their levels in Great Britain, and increased in the
United States. For building wage rates, the movements may be followed
in detail in Table 78.68 The same order is maintained in the movement of
real earnings of coal miners and—for Germany and the United States—of
comprehensive weekly earnings series. The larger increase of real earnings
in the United States and the smaller increase in Germany cannot be traced
to the developments of real per capita income in these countries. Reference
to Table 5 shows indeed that income increased considerably faster in
Germany than in the United States between 1932 and 1939. The distri-
bution of the increase of income rather than the extent of the increase
accounts for the differential development of workers' real earnings in the
two countries. It is the contrast between the guns-before-butter policy of
the German National Socialists and the social policies of the New Deal,
which is reflected in the real earnings behavior of the two countries in
the years following the Great Depression.
WORLD WAR II
Money Wages. During World War II wage levels in all three countries
rose, continuing a post-1932 trend. As can be seen in Table 76 and Chart
41, Germany, during the war, experienced very mild wage increases only,
reflecting the thoroughgoing system of controls and perhaps the high
utilization of manpower that were in effect as early as 1939. Between
1939 and the years 1943 or 1944 (the last years for which information is
available) wage rates of German building workers increased by 3 percent,
shift earnings of coal miners by 6 percent, and weekly earnings in all
German industry by about 10 percent. Much greater increases occurred
in wage levels in Great Britain and the United States than in Germany----
just as they had before the war. This is true both for wage rates and for
earnings. Weekly earnings levels in the latter countries rose by about
80 to 90 percent, compared with the 10 percent increase in German weekly
earnings reported above.
For the period of World War II wage comparisons of the three countries
may be based on several sets of fairly similar data. Specifically, hourly
wage rates of skilled building workers, average shift or weekly earnings of
coal miners, and average weekly earnings for all industry (or an approxi-
mation to such coverage) will be used for the following observations.
The most striking characteristic of wage rates for building workers is their
relatively mild rise in all three countries, amounting to only 3 percent in
The reader is reminded of the nonrepresentative character of the decline in German
building wage rates, particularly during the first two years of the comparison period.
Between 1932 and 1934, real building wage rates in that country declined by 20 percent
whereas the comprehensive index of real wage rates declined by only 3 percent. (See
also Table A- 13.)
316 WAGES iN GERMANY

TABLE 78
Real Wages in Germany, Great Britain, and the United States, 1932-1945

A. BUILDING WORKERS, COAL MINERS

HOURLY WAGE RATES OF EARNINGS OF COAL
SKILLED BUILDING WORKERS PER WEEK OR SHIFT

Great United
Germany, Britain, States,
Great United Hard Coal All Coal Bituminous
Year Germany Britain States Per Per CoalPer Week
I II I II I II I II I IL I II

1932 = 100
1932 100 100 100 100 100 100 100 100 100 100 100 100
1933 83 ... 99 ... 106 ... 103 ... 103 ... 110
1934 80 ... 99 ... 10! ... 101 ... 102 ... 133
1935 78 ... 99 ... 99 ... 101 ... 102 ... 140
1936 77 ... 100 ... 103 ... 100 ... 108 ... 161
1937 77 69 96 ... 104 ... 101 91 109 ... 160
1938 77 69 99 ... 116 ... 102 92 116 ... 147

1939 78 70 97 97 119 ... 112 101 119 119 169

1940 75 68 89 82 119 ... 114 103 121 114 173
1941 75 68 90 86 118 117 110 99 122 120 206 204
1942 74 67 91 82 111 109 108 97 142 131 211 210
1943 72 65 96 84 106 104 108 97 156 139 236 232
1944 71 64 94 82 105 102 ... ... 176 156 287 279
1945 ... ... 100 87 104 101 ... ... 186 164 285 275

1939 = 100

1939 100 100 100 100 100 100 100 100 100 100 100 100
1940 97 97 91 85 100 ... 101 101 102 96 103
1941 96 96 92 89 99 98 98 98 103 101 122 121
1942 95 95 93 85 93 92 96 96 120 110 125 124
1943 93 93 99 87 89 87 96 96 132 117 140 137
1944 91 91 97 85 88 86. ... ... 148 131 170 165
1945 ... ... 103 90 87 85 ... ... 157 138 169 163

Germany, to 32 percent in Great Britain and to 14 percent in the United

States. All these rises are appreciably below the corresponding increases
in earnings. The stronger increase of hourly rates in Great Britain than
in the United States is attributable to a virtual absence of wage controls
in Britain.69 Shift earnings of coal miners show the insignificant rise
characteristic of all wages in Germany in that period. Earnings of British
coal miners doubled, those of United States miners somewhat more than
doubled.7° A basically similar relationship is to be observed among the
69 See Jean Flexner, "Great Britain: Wage Trends and Policies, 1938-47," Monthly
Labor Review, 1947, pp. 290, if.
The difference in coverage might affect these comparisons. The British data refer
to shift earnings and include all branches of coal mining. The United States data are
average weekly earnings and cover bituminous coal mines only (see p. 313).
 GERMANY, GREAT BRITAIN, AND THE U. S. 317

Table 78, continued

B. COMPREHENSIVE INDEXES, WEEKLY RATES OR EARNINGS

WEEKLY RATES Oft EARNINGS,

COMPREHENSIVE WAGE INDEXES

Great Britain United

Germany, States,
Average Average Average
Year Earnings Rates Earnings Earnings
I II I II I 11 1 II

1932 = 100

1932 100 100 100 100 100 100

1933 105 100 104
1934 110 100 110
1935 111 100 117
1936 114 ... 99 126
1937 117 105 98 134
1938 121 110 101 127

1939 126 113 100 100 137

1940 126 113 95 89 144
1941 131 117 97 94 161 159
1942 128 115 103 94 180 178
1943 128 .114 109 96 200 199
1944 124 111 115 101 210 203
1945 121 105 198 190

1939 = 100

1939 100 100 100 100 bOa bOa 100 100

1940 100 100 95 89 110 104 105
1941 104 104 97 94 111 109 117 116
1942 102 102 103 94 124 114 131 130
1943 101 101 109 96 138 123 145 142
1944 98 98 115 101 140 124 153 148
1945 121 105 137 121 144 139

a October 1938.
SOURCE: Money wages, see Table 76. Cost of living, see source to Table 77. The cost-of-
living index numbers used for deflation were not always those given in Table 77. In
cases where the wage quotation referred to a specific month, the cost-of-living index
for the same month was used. For the adjusted cost-of-living indexes the monthly
levels were approximated by raising the official index for the month by an adjustment
factor derived from annual data. Real wages in columns (i) are derived by use of the
official cost-of-living indexes as published contemporaneously. Real wages in columns
(ii) take account of adjustments as presented in Table 77. Building rates for Germany
refer to April, for Great Britain to September, for the United States to May. Also the
comprehensive wage rates for Great Britain refer to September.

more comprehensive weekly earnings measures (available for all three

countries from 1938 on, the year Great Britain started to report average
weekly earnings in time series form). The increases during the war amounted
 318 WAGES IN GERMANY

CHART 41
Coal Miners' Earnings in Two Wars, Germany, Great Britain, and the
United States
Money earnings
Real earnings

Index (1914 100)

World War I
Index (1914 100>
220 220

200 aoo

180 180

160 160

140 140

120 120

100 100

80 80

60 60
1914 '16 '18 1914 '16 '18 1914 '16 '18

Index (1939=100)
World War II
Index (1939=100)
220 220
Germany Great Britain
200 -

-7
180 -

160 - I 60

140 - 140

120 -
JI ,_/ 120

100 - — —— -- 100

80 I I I I I I
80
1939 '41 '43 '45 1939 '41 '43 '45 1939 '41 '43 '45
Source: Table 79.

to 10 percent in Germany, 80 percent in Great Britain, and 86 percent in

the United States.7'
The German data cover the period up to 1944 only. The British and United States
earnings indexes differ with regard to industrial composition and other elements of
construction. For instance, the British treat two persons at half-time work as one fully
employed worker, while in the United States half-time workers are fully counted in the
employment indexes. It is believed, however, that for the broad comparisons here
 GERMANY, GREAT BRITAIN, AND THE U. S. 319
Living Costs. It is particularly necessary, when one seeks to appraise
wage behavior during wartime, to juxtapose wage developments with
the concomitant changes in the retail price levels of goods typically
consumed by working-class families (see Table 77). The problems created
by rapidly changing consumption patterns have already been pointed out.
The need to maintain the character of one's price measure over time con-
ificts increasingly with the need to keep the measure representative of
current patterns of expenditure. How did the statistical agencies of the three
nations react to these difficulties? The Germans approached the problem
by gradual substitution of available goods of similar function and foods
of similar caloric value—a procedure which, in view of the grave shortages,
led to basic changes in the quality and composition of the goods priced.
In the United States also, the composition of the index underwent changes;
scarce goods were dropped, and available goods or grades of goods were
linked to the established index. But these substitutions were not nearly so
radical as those in the German index, and the retail price measure could
thus maintain a higher degree of comparability over time without becom-
ing obsolete. In Great Britain, at the outbreak of World War II, the govern-
ment decided to postpone a long-overdue revision of its retail price
measure. Thus Great Britain continued during that period to employ an
index established in 1904 (and revised only slightly thereafter), geared to
time-honored consumption patterns. Furthermore, subsidies to stabilize
food prices were granted predominantly for the goods represented, or
even overrepresented, in the British measure. The resulting "stabilization
of the index" was presumably intended to limit the inflationary con-
sequences of wage-rate changes arising out of contracts with escalator
clauses. While some downward bias in the measure of living costs and some
upward bias in real wages must be expected in the contemporaneous
measures of all three countries, the bias is apt to be least serious in the
United States index, and considerably more serious in those of Germany
and Great Britain. In Germany, the bias is created mainly by lack of goods
and deterioration of quality—elements whose quantitative impact on the
index is difficult to measure. In Great Britain, the bias is caused mainly
by the limitation of the index to a number of simple price-supported
staples. In all three countries the defects of the indexes were recognized,
and attempts were made to gauge the extent of bias and possibly to revise
the cost-of-living measure. In Germany the inadequacy of the index had

pursued The indexes offer an adequate guide. The U.S. Bureau of Labor Statistics in
1944 computed weekly earnings changes in Great Britain and in the United States
between October 1938 and July 1943, using comparable industrial groups and the same
(United States) employment composition as weights. The results were in line with those
shown by the unadjusted data, in that they indicated similar weekly earnings trends in
both countries. The similarity was produced by a smaller increase of hours but a larger
increase of hourly earnings in Great Britain as compared with the United States. See
"Wartime Hours and Earnings in the United States and Great Britain," Monthly
Labor Review, July 1944, especially pp. 153-54 and 156.
320 WAGES iN GERMANY
already become evident during the preparedness economy preceding the
actual launching of the war. 72 In the United States the debate raged during
the war years and led to re-evaluations of living-cost changes.73 In Great
Britain a major revision of the index was undertaken in 1947, when attempts
were made to recompute all changes in living costs on the basis of the
new index structure.74 In the present analysis of living costs and in the
comparison of real wages, allowance, in the form of alternative indexes,
has been made for revisions.
Real Wages. We now turn to the quotients which result when we divide
consumers' goods price indexes into money wage measures—"real wages."
For the years under discussion these measures can at best indicate broad
tendencies. They are presented in Table
Real wage rates per hour, as represented by time rates for skilled build-
ing workers, decreased during World War II in all three countries. The
lowest relative level, 15 percent below 1939, occurred in 1942 in Great
Britain (deflation by Allen index). These rates, because of their minimum
character, can scarcely be regarded as describing properly the effectively
paid real hourly rates in any of the three countries—not to speak of the
many other important elements that shape the total wage picture in time of
war. It is more instructive, therefore, to turn to measures of real earnings.
Looking at the real earnings of coal miners, we find the following order
in the extent of war changes: in Germany real earnings fell, in Britain
they rose, and in the United States they rose still more. Specifically, real
shift earnings of German coal miners were 4 percent below 1939 levels
in 1943. Real shift earnings of British coal miners rose by 57 percent
between the beginning and the last year of the war, according to official
figures. If deflation is carried through by the adjusted living-cost measure,
as computed by Allen, the increase amounts to only 38 percent. The
United States figures show a weekly earnings increase for bituminous
coal miners of close to 70 percent (or 63 percent after adjustment) between
the years 1939 and 1945. Coal mining was of course an important industry
during the war, and earnings in that industry are not necessarily indicative
of earnings behavior in general. Weekly real earnings changes for all
industry are in fact somewhat more moderate, Germany registering a
small decline, Great Britain an increase of 20 percent (deflated by the
revised retail price measure), and the United States a rise of about 40
percent. In all three countries there was a decline of weekly real earnings
in the last year reported. In Germany the decline appears in the figures
See Chapter 5, section on Wages under National Socialism.
See Report of the President's Committee on the Cost of Living, Office of Economic
Stabilization, 1945, as quoted in Table 77.
See R. G. D. Mien, "Prices," London and Cambridge Economic Service, February
1949.
The revised living-cost measures, discussed above, have also been included in
the tabulations. For Germany the revisions related to increases during the period
1933-37. Thus only the level but not the movement of living costs and real wages during
World War Ii are affected by these adjustments.
 GERMANY, GREAT BRITAIN, AND THE u. s. 321
for 1944 (and must be assumed to have continued through the remainder
of the war), in Great Britain and the United States it appears only in the
data for 1945. Both the decline during the late war years and the differential
for Germany and the other two countries reflect again the major economic,
political, and military circumstances of the three powers at the end of the
war.
Comparison of the Two War Periods. The availability of information on
miners' earnings, in both world wars and for all three countries, permits
some comparisons of wage behavior during the two wars. The course of
money earnings and real earnings of miners is illustrated in Table 79
and Chart 41. In these comparisons it is important to consider that the
first war lasted about four years and the second about six. For money
wages, we find the greatest contrast in wage behavior in Germany for the
two war periods. In that country coal miners' earnings doubled during
the first war, but increased by only a very few percent during the second.
The increases in miners' money earnings in England and America were
TABLE 79
Coal Miners' Earnings in Two World Wars: Germany, Great Britain, and
the United States
MONEY EARNINGS REAL EARNINGS

Great United Great United

Germany Britain, States, Germany, Britain, States,
Hard Coal, All Coal, Bit. Coal, Hard Coal, All Coal, Bit. Coal,
Year per Shift per Shift per Week per per per Week
WORLD WAR I
(1914 = 100)
1914 100 100 100 100 100 100
1915 110 115 106 88 96 108
1916 131 129 116 79 96 108
1917 165 136 144 67 85 112
1918 208 195 178 68 108 113

WORLD WAR II
(1939 = 100)
1939 100 100 100 100 100 100
1940 105 112 103 101 96 103
1941 103 128 129 98 101 121
1942 104 150 147 96 110 124
1943 106 164 174 96 117 137
1944 ... 187 215 ... 131 165
1945 ... 200 219 ... 138 163

SOURCE:
For 1914-18: Germany, fur das Berg-, Hlltten- und Salinenwesen, passtm.
Deflation by cost-of-living index as derived from data published by the Statistische
Reichsamt.
Great Britain, A. L. Bowley, Prices and Wages in the United Kingdom, 1914-20,
pp. 106 and 150. Deflation by "modified mdcx."
United States: Douglas, Real Wages in the United States, p. 162.
For 1939-45: Money wages, Table 76; real wages, Table 78.
322 WAGES iN GERMANY
somewhat milder during the first four years of World War II than during
World War I. However, in the course of the last two years of the more
recent conflict, the cumulative increase in war wages clearly surpassed
that experienced from 1914 to 1918. The totalitarian approach to the
control of money wages obviously was more effective than the less incisive
measures adopted in Great Britain and the United States.
For real wages, we observe a marked similarity in the comparative
behavior of the three countries between the two wars. In both, Germany
occupied the least favorable and the United States the most favorable
position. The most conspicuous contrast between the two wars is the
apparently more favorable real earnings record in all three countries
during World War II. During World War I real earnings of German coal
miners were cut drastically, and earnings of their British counterparts
were moderately reduced, whereas earnings of United States miners
increased by about 10 percent. These movements are to be compared with
the insignificant decline of miners' real earnings in Germany (at least
during the first four war years), with the substantial increase of miners'
earnings in Britain, and the still more pronounced gains of miners'
earnings in the United States during World War II. The findings may
appear surprising, in view of the greater scope, longer duration, and
greater destructiveness of the more recent conflict. Yet there are plausible
explanations for the reported developments. For Germany, our informa-
tion reaches only to 1943. Up to that year that country was able to avoid
the worst consequences of war conditions. It was militarily successful,
could base its war production not only on the efforts of Germans but also
on the exploitation of foreign workers, and consistently ransacked the
economies of conquered areas by sequestering their production, wearing
out their equipment, and so on.76 For the British experience the greater
effectiveness of the German blockade during first war and the more
substantial aid from abroad during the second war may provide some
explanation. For both Germany and Great Britain the higher productivity
of labor in the second as compared with the first war forms an important
condition for the more favorable showing of real earnings in World War
II. As for the United States, World War II brought about such an un-
paralleled expansion of industrial activity that new records were set
in output of war and war-related goods, and at the same time weekly real
earnings could rise more than they did during the earlier war.
The experiences of the three countries during World War I had pro-
found effects upon their respective economic conditions and upon their
wage histories during the decadesL following the Armistice of 1918.
Similarly sweeping effects were also to follow from the varying experiences
of these countries during World War II. But this is a new story that will
require the perspective of future students for its presentation and
appraisal.
See JUrgen Kuczynski and M. Witt, The Economics of Barbarism (London, 1942).