'■ 3<o//C5^
JOBS
MACHINES AND UNEMPLOYMENT
' I
/
\_z( _S \ e cl, Waving
Reprinted from REPORT ON PROGRESS
OF THE WPA PROGRAM, June 30, 1938
WORKS PROGRESS ADMINISTRATION
Harry L. Hopkins, Administrator Corrington Gill, Assistant Administrator
October 29, 1938.
Hon. Harry L. Hopkins,
Works Progress Administrator.
Sir: I am transmitting to you a reprint of a chapter in the
Report on Progress of the 11'orks Program for June 1938. The
material in this report is based on Summary of Findings to Date,
March 1938, by David Weintraub and Irving Kaplan (Works
Progress Administration, National Research Project, Philadelphia).
The National Research Project on Reemployment Oppor¬
tunities and Recent Changes in Industrial Techniques is one
of the major aspects of the research program organized by the
Works Progress Administration for the purpose of collecting
and analyzing data bearing on problems of employment, unem¬
ployment, and relief. The Project was organized in December
1935 to inquire into the extent of recent changes in industrial
techniques and to evaluate the effects of these changes on the
volume of employment and unemployment. The Project has
enjoyed the valuable cooperation of various agencies of the
United States Government as well as private agencies which are
collectors and repositories of pertinent information.
A steadily increasing amount of data on unemployment
collected under our research program is becoming available in
final form. This reprint is intended to make available to a
wider audience a significant part of that data.
Respectfully yours,
A ssista nt A dm in istrator.
NATIONAL RESEARCH PROJECT
on Reemployment Opportunities and Recent Changes in Industrial Techniques
David Weintraub, Director Irving Kaplan, Associate Director
TECHNOLOGY AND UNEMPLOYMENT'
The Problem of Technological Unemployment
The relationship of technological advances to unem-
■ ployment has been vigorously debated since the turn
of the nineteenth century, when machines began to be
used on a large scale to displace hand-tool and hand¬
craft methods. The "industrial revolution" was ex¬
perienced first in England, and was accompanied by
the loss of employment by many skilled workers. En¬
raged jobless workers sometimes smashed the machines
which had displaced them. Despite this resistance,
machinery was slowly installed in all the chief branches
of manufacturing.
Europe and the United States soon followed England
in introducing labor-saving machinery on a large scale.
It was observed that, while the immediate effect of the
machines often was to create unemployment, the new
machine era also ushered in greatly increased produc¬
tion and employment opportunities in many new fields.
The "Machine Age" was therefore hailed for a consider¬
able period in some quarters as a day of opportunity for
labor. In recent years, however, widespread unemploy¬
ment has once again focused attention on the relation¬
ship between technology and joblessness.
In the modern industrial era, unemployment is a
problem of much more importance than it was a century
ago. At that time the major part of economic activity
in the United States was carried on by self-employed
persons; today only about one-fourth of the persons
who have a gainful occupation are self-employed while
three-fourths are subject to hiring and firing. The
economic security of the great majority of the popula¬
tion therefore depends today on the ability of industry
to employ their services continuously. Yet it is an
outstanding characteristic of private industry not to be
able to employ workers continuously.
* This article is based on Summary of Findings to Date, March 1938, by David
Weintraub and Irving Kaplan (Works Progress Administration, National Research
Project. Philadelphia).
Some industries require a large number of workers
for a period of months or weeks and few or none for the
rest of the year. Also, although some industries need
workers and many workers need jobs, the workers and
the jobs are rarely so located as to result in the fullest
possible employment. Above all, modern industry is
characterized by periodic booms and by depressions
which result in large-scale unemployment for all classes
of workers. It is during these periodic depressions that
the modern worker's insecurity is particularly em¬
phasized because it is then that his primary needs re¬
main unsatisfied while all around him lie unused goods,
idle factories, and other imused productive resources.
This insecurity notwithstanding, each depression
until the one of the early 1930's saw the United States
emerge on a higher level of production and employment.
Not only were there employment for the workers who
were here and investment opportunity for American
capital, but the country provided jobs for millions of
immigrants from all parts of the world and an outlet for
billions of foreign capital. At the same time, rapid
improvements were being made in production methods
A. O. Smith Corporation
AUTOMATIC MACHINES RIVET AUTOMOBILE FRAMES AS THEY
PASS BY ON CONVEYOR BELTS
3
Mechanization of production and high labor efficiency
were things to which Americans pointed with pride.
Of course, many craftsmen lost their jobs, the value of
their skills, and their chances for reemployment as a
result of mechanization and technological change.
The immediate effects of technological change involved
displacement of labor, changes in the type of labor
needed, and occupational obsolescence. These, in
turn, presented problems of destitution and relief.
During periods of widespread unemployment especially,
there was serious and organized opposition to the intro¬
duction of machines. However, in the upward sweep
of the country's economic development, the problems
which accompanied the displacement of workers were re¬
garded as merely shortlived consequences of temporary
dislocations associated with progressive mechanization.
Indeed, machines and the productive apparatus in
general are only the inanimate means used by the
economic organization to produce for the market.
Technological improvements are introduced in order to
produce more goods at lower costs and with less labor
than would otherwise be required. Whether these im¬
provements result in a sufficient increase in production
to require the employment of an increased number of
workers, depends on the way in which the improved
technology is used. Certainly, if the needs of the
population and the mechanical capacity of the available
machines were the only criteria, no man would ever need
to go unemployed. Yet such data as are available
suggest that successive major depressions have seen
increases in the number of unemployed workers, and
the years since 1929 have been marked by unemploy¬
ment which is greater that at any other time in the his¬
tory of the country.
Even before 1929 many economists, labor groups, and
some governmental bodies began to draw attention to
the increasing volume of unemployment and to the
emergence of certain new phenomena in the industrial
development of the country. The manufacturing in¬
dustries which had hitherto provided work for increas-
THE CATERPILLAR TRACTOR IS A FLEXII1LE TYPE OF LOGGING
EQUIPMENT
4
ing proportions of the country's growing labor supply
had ceased lo absorb workers although their production
continued to grow. The same was true of the mining
industries. The railroads were handling a larger
volume of traffic with a decreasing number of workers
and the relative declines in the population working on
farms turned into an absolute decline during the second
decade of this century.
Thus even before 1929 there was a new solicitude
concerning the relationship between technological
change and unemployment. The number of jobs
available at any time is a result of the volume of produc¬
tion and the average amount one man can produce.
If the average productivity keeps going up, the number
of available jobs must go down unless production rises
faster than productivity. With the unparalleled busi¬
ness depression that developed after 1929—creating at
its low point an unemployment total estimated variously
from 14 million to 17 million—a number of questions
became more and more insistent: Is productivity con¬
tinuing to go up while production is lagging behind
earlier levels? What new levels of production are
required to employ the increasing population depend¬
ent on jobs for a livelihood? What is the role of chang¬
ing technology in modem industrial society? How
does it affect employment and unemployment ?
Studies of the National Research Project
Because the existing data could not answer these
questions satisfactorily, the Works Progress Adminis¬
tration, in December 1935, organized the National
Research Project on Reemployment Opportunities and
Recent Changes in Industrial Techniques in order "to
inquire, with the cooperation of industry, labor, and
governmental and private agencies, into the extent of
recent changes in industrial techniques and to evaluate
the effects of these changes on the volume of employ¬
ment and unemployment." Many of the studies
undertaken on this project are now approaching com¬
pletion ami a large body of data has already been pub¬
lished.2 These studies present a mass of detailed infor¬
mation covering changes in technology and productiv¬
ity during the last two or three decades. The Summary
of Findings as of March 1938 brings together the avail¬
able findings to date and presents them in convenient
form.3
These studies have revealed that although there are
many forces in operation which tend to reduce produc¬
tivity when the level of industrial activity is low, wide¬
spread increases in productivity have taken place since
1929 in almost all industries. Since the labor supply
lias increased by about 4,000,000 workers during the
last eight years, these increases indicate that a national
2 See list of reports published to date at the end of this article.
8 See footnote 1.
Jeffery Manufacturing Co.
UNDERCUTTING MACHINE ELIMINATES THE LABORIOUS TASK
OF UNDERCUTTING COAL WITH A PICK
output of about 20 percent beyond that of 1929 must be
attained before the number of jobs will reach a level
that would reduce unemployment to the 1929 level.
Skills Are Being Leveled Off
The studies indicate that technological changes have
tended toward the leveling of the skill requirements of
industry. The mechanical innovations introduced
generally substituted a small number of semiskilled
machine operators for a larger number of unskilled
laborers or skilled artisans. Automatically operated
conveying systems for the handling of materials have
been adapted to a wide variety of uses. These devices
eliminate the need for much backbreaking work in
lifting, carrying, and placing material. For instance,
during 1936 and 1937 more than 1,500 conveyors were
sold to coal mines, as compared with a total of 849
in use in 1935; 641 mobile coal loaders were sold during
the same two years, as compared with 657 in use in
1935. Aside from the effect that these machines may
have on the total number of persons employed in coal
mining, the workers required for the operation of these
machines need no longer have the highly diversified
skills and experience of the pick-and-shovel coal miner.
The substitution of machine methods for hand proc¬
esses substitutes the semiskilled machine operator for
the old skilled artisan. For example, when a cigar
factory is mechanized, the older, skilled cigar makers
are almost always laid off and young girls, fewer than
a third in number, are hired to tend the nearly automatic
cigar-making machines.
Automatically operated machines, macliines which
may be pre-set for a variety of standard operations, also
eliminate need for special skills. All these processes
tend to create reservoirs of the unemployed among the
groups displaced, particularly when production is
lagging.
Workers with specialized skills in a particular indus¬
try are especially limited in their opportunities for
adjustment even when jobs are on the increase. Some
of these skills may be usable elsewhere, but others,
like those of coal miners, cigar makers, weavers, or
railroad workers, are not easily adaptable to other
work. If these displaced workers are young they may
l>e able to start over again, but if they are past middle
age they find it very difficult if not impossible to gain a
now foothold. Then they become dependent on public
aid. For, the most general effect of prevailing types of
changes in industrial processes is to place a premium
on speed, dexterity, and adaptability rather than on
experience and specialized skill. The basis is thus pro¬
vided for preference for younger workers to the disad¬
vantage of the more experienced and the older workers
with family responsibilities.
Effect of Technological Change Different
in Prosperity and Depression
When the country was moving forward to higher
levels of production and employment, the effect of
technological change on selected groups of workers was
often different from what it has been during the last
eight years. In the past, when industrial activity in
certain communities collapsed under the competitive
pressure of newly discovered resources elsewhere or of
new or improved processes, the communities became
"ghost towns." Their populations moved to other
parts of the country where jobs were available. Under
conditions of widespread unemployment, however,
such places become "stranded communities." The
unemployed population has no place to go. During the
past eight years this has been an important factor in the
cumulation of dependents in the poorer agricultural
areas, in the cut-over timber areas, in many mining
communities, and in many one-industry towns.
Production per Unit of Labor Rising
Despite Declining Output
Notwithstanding the operation of factors which tend
to decrease labor productivity when production is at a
LL
*
International Harvester Co.
THIS TRACTOR EQUIPPED WITH PNEUMATIC TIRES REPLACES
THREE DRIVERS AND TEAMS AND DOES THE WORK MORE
QUICKLY
5
Reading Railroad
THE BALLAST PLOW IS DESIGNED TO DO THE WORK OF MANY
TRACK MAINTENANCE WORKERS
low level, notable increases in productivity have been
achieved since 1929 in practically all industries. Thus
in 1335 output per hour of work was substantially higher
than in 1929 in most of the mining industries, in the
telephone and electric-power industries, in railroad
transportation, and in almost all the manufacturing
industries.
Even before 1929 the number employed had begun
to decline in the railroad and coal-mining industries,
and in spite of large increases in production the number
of jobs in the manufacturing industries was about the
same in 1929 as in 1920. In agriculture also, the num¬
ber of people working had been declining with the intro¬
duction of tractor power and power-driven equipment.
Many industries have achieved productivity increases
at a faster rate since 1929 than during the preceding
decade. Thus productivity in the telephone industry
increased by 25 percent between 1919 and 1929, and by
another 35 percent between 1929 and 1935. Railroad
traffic units per hour of employment increased by 28
percent during the decade of the 1920's and by another
37 percent during the first six years of the 1930's. Many
manufacturing industries, particularly in the food and
textile group, experienced at least as great an annual
average rate of increase in productivity after 1929 as
in the earlier decade.
Many industries increased their productivity even
during the years of sharply falling production between
1929 and 1933. This was true of the railroads, the
6
operating departments of the electric light and power
industry, the telephone industry and a number of manu¬
facturing industries. For example, 15 percent more
tobacco products were obtained per hour of work in
1933 than in 1929: 13 percent more cigars, 23 percent
more cigarettes, and 18 percent more chewing and
smoking tobacco and snuff. For each of these products
the output had been declining after 1929.
In almost every case, industries which suffered a
decline in productivity during the years of sharply
declining production achieved anew high level of produc¬
tivity once there was an upturn in production. Accord¬
ing to the Project's findings, manufacturing industries
employing nine-tenths of the total labor employed in
manufacturing had a higher productivity in 1935 than
in 1929. In the automobile industry, for example, when
during the period 192!) to 1932 production fell to 30 per¬
cent of the 1929 peak, a given number of workers pro¬
duced only 82 percent as many finished vehicles and
chassis as in 1929. But by 1933 and 1934 they were
producing at the same rate as in 1929, and by 1936,
though production was still one-sixth below the peak,
the output per man-hour was almost one-sixth higher
than in 1929.
Again, when iron and steel production declined by
almost three-quarters between 1929 and 1932, output
per hour of employment declined somewhat, but in
1933 and 1934 when production had increased, more
tonnage was obtained per hour of work than in 1929
I nternational Harvester Co.
DISK HARROW AND ROLLER DRAWN IN TANDEM COMBINE
TWO OPERATIONS IN ONE
By 1935 the output per man-hour was 9 percent above
the 1929 level, and by 1936 it was 13 percent higher.
Increasing productivity in this industry is accounted
for in part by the growing installation of the hot-strip
mills for the production of sheet steel. Of tho 8,5()0,()<)0
ton capacity of these continuous strip mills in 1936,
5,750,000 had been installed after 1929. The capacity
either available today or under construction is now
reported to be at 14 million tons. The amount of labor
saved by the substitution of these continuous hot-strip
mills for the older type can bo gauged by the example of
one mill in which 375 men were employed at producing
an output which required 4,512 men using the old pro¬
duction process.
The most striking increases in productivity occurred
in industries whose production rose through the last
eight years. The output of rayon yarn manufacturers,
for instance, has been increasing steadily and rapidly.
The increase in production was accompanied by a rapid
increase in the output per unit of labor time. By 1933
productivity was twice the 1929 level, and in 1937 it
was 2.4 times that of 1929.
Typical Current Technological Changes Require
Relatively Small Capital Expenditures
During periods of economic expansion, technological
changes are introduced primarily as part of the process
of expanding productive facilities. This process has in
the past involved an increasing demand for the produc¬
tion of capital goods and has provided increasing
employment and income to workers employed in these
industries and, through them, increasing demand for
consumption goods. During the past eight years, the
emphasis has, however, been on saving labor rather
than on expanding production facilities.
The low level of production in the capital goods pro¬
ducing industries and of investment in capital goods
since the end of 1929 is characteristic of depression
periods. And yet productivity has risen since 1929.
The ability of so many of our industries to increase pro¬
ductivity with a small expenditure of capital is due in
part to the types of technological changes which are
currently available to industries. Existing plant facili¬
ties and processes have been improved by the addition
of auxiliary equipment designed to improve the opera¬
ting efficiency of machines already installed, to control
the production process, frequently automatically, or to
substitute automatically operated equipment for labor.
These innovations involve relatively small capital costs
because they are directed primarily toward economies
in operation rather than toward plant expansion. In
addition, the development of chemical processes yielded
great economies in labor while requiring relatively little
capital. For example, in the rubber tire and tube in¬
dustry, the number of pounds of product manufactured
per hour of labor has doubled during the past decade.
The increased productivity was attained primarily by
the reduction in the time required for rubber processing
through the use of new organic accelerators and other
chemicals.
Many of the improvements made during recent years
have served to expand the productive capacity of in¬
dustry. Increases in productivity, realized by reducing
California Fruil Growers Exchange
SORTING AND GRADING OF FRUIT HAS BEEN EXTENSIVELY MECHANIZED
the labor time per worker rather than the number of
workers needed for the production process, also increase
production capacity. Other expansions of capacity
have been attained by substituting larger units of
equipment because of their greater operating efficiency.
For instance, in cement manufacture, only 56 of the 810
kilns in'the industry in 1925 were over 200 feet long.
By 1935 there were 119 such kilns out of a total of 823.
Return of Unemployment to Predepression Levels
Requires Expansion of Purchasing Power and of
Capital Investment
The Project's studies show that with the same amount
of labor and without any additional investment in plant
expansion many industries can produce much more
today than in 1929. Furthermore, the increases in
productivity which characteristically accompany every
gain in production reflect the current ability of industries
to realize further labor economies with the next increase
in production.
These findings have strengthened the conviction that
the number of unemployed cannot be brought down to the
levels of the late 1920's without very much heavier cap¬
ital investment than has been taking place during recent
years. Government has always played an important
role in the economic expansion of the country. For
example, during the years 1919-27, public expenditures
for construction purposes amounted to a sum equal to
one-eighth of the total national expenditures for pro¬
ducers' durable goods and residential construction,
during 1927-35 they amounted to one-fourth, and dur¬
ing the entire period 1919-35, to one-sixth of the total.
Many of these Government expenditures can take
such forms as will encourage private capital outlays,
e. g., the construction of roads, the building of streets
and sewers, the construction of airports and airway
facilities, rural electrification, irrigation, and flood
control work. Others, such as low-cost housing, can
result directly in large capital outlays.
In this direction the activities of the WPA may
be regarded not oidy as a means for relieving the needs
of the unemployed but also as part of a program which,
through expenditures for airports, roads, sewer systems,
and other construction work, tends to stim"'°+"
outlays by private business.
8
Reports
The following reports prepared on the WPA National
Research Project have been issued or are in press:
Studies in Changing Technology and Labor Productivity
Agriculture
Changes in Technology and Labor Requirements in Crop Pro¬
duction:
Corn
Hugar Beets
Potatoes
Changes in Farm Power and Equipment: Mechanical Cotton
Picker
Trends in Size and Production of the Aggregate I arm Enterprise,
1909-36
Selected References on Practices and Use of Labor on Farms
Mining
Technology and the Mineral Industries
Mechanization Trends in Metal and Nonmetal Mining as Indi¬
cated by Sales of Underground Loading Equipment
The Decline in Grade of Ore
Fuel Efficiency in Cement Manufacture, 1909-35
Small-Scale Placer Mines as a Source of Gold, Employment,
and Livelihood in 1935
Employment and Related Statistics of Mines and Quarries,
1935: Coal
Manufacture
Employment and Productivity in the Beet Sugar Industry
Labor Productivity in the Leather Industry
Effects of Mechanization in Cigar Manufacture
Mechanical Changes in the Cotton-Textile Industry. 1910 to
1936
Mechanical Changes in the Woolen and Worsted Industries,
1910 to 1936
Systems of Shop Management in the Cotton Garment Industry
Studies of the Effects of Industrial Change on Labor Markets
Recent Trends in Employment and Unemployment in Phila¬
delphia
Employment and Unemployment in Philadelphia in May 1936
The Labor Force of the Philadelphia Radio Industry in 1936
Philadelphia Weavers and Loom Fixers
Cigar Makers—After the Lay-Off
General
The Research Program of the National Research Project
TT o-nH Tncrruming Productivity
A 1938
GOVERNMENT PRINTING OFFICE: 193C