ArticlePDF Available

Industrial Labor Mobility in Historical Perspective

SANFORD
M.
JACOBY"
Industrial Labor Mobility
in
Historical Perspective
THE
LABOR
MARKET
TODAY
is less fluid than it was in the nine-
teenth and early twentieth centuries. Job and geographic mobility rates declined
after
1920
as a result of complex changes in industrial employment policies
and the character of the labor force. Although mobility rates in the United
States currently
are
higher than in other industrialized nations (OECD,
1965;
Cole,
1979),
they appear relatively low when viewed in historical perspective.
Many believe that previous generations of American workers resided in
stable communities and had a strong commitment to work. But high rates of
job and spatial mobility before
1920
show that this is an idealized image of
our
past bearing little relationship to the historical record. In fact, employment
stability as a feature of working-class life has a rather short history,
as
do
modern personnel practices which encourage long-term employment rela-
tionships.
Attempts to explain the decline in mobility since
1920
typically emphasize
the progressive adoption
by
industry of a number of new personnel practices
(associated with the development of internal labor markets) which increased
employee attachment to
the
firm
(Ross,
1958).
The rise of internal labor
markets, in turn, is now generally attributed to industry's desire to reduce
turnover and related costs (Doeringer and Piore,
1971).
In this paper,
I
will
argue that the decline in labor mobility had sources much broader than
employer personnel policies, and that the rise of internal labor markets was
due at least as much to the threats posed
by
unionization as to a pervasive
desire
by
management to obtain perceived economic benefits.
The first part of the paper explains the high rates of mobility existing before
1920
in terms of the immigrant and pre-industrial origins of the labor force
and prevailing industrial employment policies. The second part discusses the
reasons for
the
fall
in mobility after
1920,
including changes in personnel
policies, the spread of unionization, and changes in social and demographic
*Assistant Professor, Graduate
School
of Management, University
of
California,
Los
Angeles.
INDUSTRIAL
RELATIONS,
Vol.
22,
No.
2
(Spring 1983). 01983
by
the Regents
of
the University
of
California.
0019/8676/83/525/261/$10.00
261
262
/
SANFORD
M.
JACOBY
characteristics of the labor force.
I
will conclude
by
evaluating alternative
theories of
the
long-run determinants of industrial labor mobility.
Labor
Mobility
Before
1920
The
employment relationship during the nineteenth and early
twentieth centuries was one of weak attachment. Both parties took
full
advantage
of their legal rights, acquired in the early nineteenth century, to quit or
dismiss at will. Consequently,
the
labor market was a market of movement,
characterized
by
high
rates of mobility.
The earliest data on labor turnover in manufacturing come from the textile
industry of New England.
The
young Yankee women who worked in the mills
were an unstable labor force. Most of the women were unmarried and could
return to their parents’ farms
if
they were dissatisfied or work was scarce
(Ware,
1931).
But
the immigrants who began to replace native workers in
the eighteen fifties had high turnover rates too.
A
study of
151
Scottish
weavers recruited
by
Lyman Mills in
1853
found that nearly
80
per cent
of
the
women had left the firm within three years (Ginger,
1954).
A
similar picture emerges in other industries, especially those employing
relatively more men.
A
Massachusetts firm that manufactured textile machinery
recruited large numbers of French-Canadians between
1860
and
1890.
But
“so
rapid was the turnover” that only one of every three workers hired stayed
with the firm (Navin,
1950,
pp.
160-161).
There
also
were very low rates of
persistence at the Boston Manufacturing Company in Waltham. For quin-
quennial periods between
1850
and
1865,
only
10
to
12
per cent of
the
male
workers present at the beginning of a period still were working for the firm
five years later (Gitelman,
1974,
p.
71).
Nineteenth-century employers occasionally complained about what one of
them termed “the nomadic system of employing men.
A
common response
to high quit rates was to withhold the wages of those who left without giving
prior notice, a practice that
also
deterred strikes. When a Massachusetts mill
agent was asked in the eighteen seventies to explain these wage forfeitures,
he insisted that “if a mill did not keep back workers’ wages, it would simply
awake to find
all
its hands gone by the morning” (Rodgers,
1978,
p.
164).
Better data available for the first two decades of the twentieth century
show continuing high levels of turnover. During the nineteen tens, it was
quite common for firms to experience monthly separation rates in excess of
10
per cent. Thus, a survey of
14
Detroit manufacturing establishments found
an average monthly separation rate of
15.3
per cent in
1913-1914
(Slichter,
1921a;
Brissenden and Frankel,
1922).
,,
‘To
put this in perspective, the highest average monthly separation rate in manufacturing during
the
sixties
was
4.9
per cent in
1969
(U.S.
Bureau of
Labor
Statistics,
1975,
p.
130).
Industrial Labor Mobility
I
263
Wartime turnover rates exceeded those of the preceding period, although
the rise was not unprecedented. Turnover in a relatively “good” prewar year
like
1913
approached wartime levels: the unemployment rate was about
35
per cent higher, and the separation rate was
33
per cent lower, in
1913
as
compared
to
1917-1918
(Brissenden and Frankel,
1922;
Lebergott,
1964).
Thus, the fragmentary data on labor turnover indicate a pattern of continuously
high separation rates before
1920.
Geographic mobility.
One indirect way of measuring turnover is to examine
geographic mobility rates, the amount of movement into and out of a place
over time.
This
measure
is
imperfect, however, since some separations involve
changes between jobs in the same area
and
some moves are
by
persons outside
the
labor force. Correct estimates would demand knowledge of
the
proportion
of separations that involve a move (currently about
40
per cent) and the
proportion of moves that entail a separation (currently over
60
per cent)
(Bartel,
1979).
Although there is
no
way of predicting how these probabilities
may have changed since the nineteenth century,2 the relationship between
geographic and job mobility suggests that it is valid to use the former to track
major shifts in the latter.
For many years the standard measure of geographic mobility was the level
of net migration within a city, state, or region. Net migration represents the
intersection of two population flows, but it gives no indication of their size.
Urban historians have begun to estimate gross migration flows for various
cities by using city directories, voter registration rolls, and property tax files.
The results have been startling.
For example, net migration into and out of Boston during the ten years
after
1880
totaled
65,179
individuals, whereas the gross flows involved nearly
one and one-half million people. Moreover, rather than being a decade of
particularly high rates of movement,
the
eighteen eighties saw Boston’s lowest
decennial mobility rates between
1830
and
1920.
Sixty-four per cent of its
adult male population in
1880
still was residing there ten years later, versus
an average decennial persistence rate of about
40
per cent for other decades
(Thernstrom,
1973).
The Boston experience was not unique. Persistence rates throughout urban
America were as low as Boston’s during the
1830-1920
period (Thernstrom,
1973).
There is evidence that males were less mobile during the late eighteenth
and
early nineteenth centuries. But after
1830,
geographic mobility increased
aPrecliction
is
difticult
because
we
don’t
know
the
net
effect
of opposing forces. For example,
the
spread
of
home
ownership
and
the
equalization
of
economic opportunity between regions have acted to depress
the
proportion
of
separations that entail a move. On
the
other hand, cheaper transportation and better
information on distant labor markets have had
an
opposite effect.
264
/
SANFORD
M.
JACOBY
throughout
the
United States and remained at high levels at least until
1920
(Allen,
1977).
This ceaseless movement into and out of American cities during
this period supports the presumption that industrial turnover. rates were
continuously high before
1920.
That presumption is buttressed by the finding that geographic mobility
varied inversely with occupational status between
1830
and
1920.
Skilled and
less skilled manual workers had lower persistence rates than white-collar
workers (Thernstrom,
1973).
Whatever stability existed in American cities
during this period largely was the product of the relative immobility of such
persons as professionals, merchants, teachers, and clerks. This may have been
why the middle class considered stability to
be
a
virtue.
Sources
of
Early
Mobility
The
skiled.
During
the
nineteenth and early twentieth centuries,
skilled workers
had
what Ulman
(1955,
p.
57)
has termed a “tradition of
itinerancy” that formed at the intersection of artisanal work habits and the
requisites of learning a trade. It was supported by craft institutions, especially
the trade union, as well as
by
successive waves of immigrant artisans carrying
similar traditions to the United States.
The
footloose craftsman was a familiar figure in industrializing societies,
increasing his knowledge and versatility
by
moving from shop to shop while
acquiring the secrets of his trade. Employers often approved of this type of
mobility. The labor supervisor at National Cash Register wrote in
1907
that
for
a
skilled worker, “it is of value, not a detriment,
if
he
has
had
several
employers-he is learning the trade” (Worman,
1907,
p.
57).
Also
reinforcing
movement was a work ethic that valued manliness and independence which
would
be
unacceptably compromised by demeaning working conditions or
merely
by
long tenure itseK3
The
craft union facilitated the skilled worker’s propensity to move.
The
constitutions of the early national unions required local secretaries to furnish
reports on the conditions of the trade in their area and assist travelling
members in finding jobs. Some unions had travelling loan systems, which
gave members money to finance a search for work
if
none was to
be
found
near home. However, the system was on the decline by the beginning of
this century (Ulman,
1955),
partly “because, as in the case of the iron molders,
it was made use of to secure a
free
holiday” (Davidson,
1898,
p.
178).
Qne trade union official noted that for many skilled workers
he
knew,
“A
job
may be satisfactory in
every respect, quite
as
good
as
they
are
likely to find anywhere, yet they will leave because they
do
not
want to remain in one shop too long
. .
.it rests upon a fear of losing their independence, of getting into
a frame of mind wherein they will
come
to attach disproportionate importance to
the
retention
of
a
certain
job
(Portenar,
1917,
p.
193).
Industrial
Labor
Mobility
I
265
The
less
skilled.
About two-thirds of immigrants arriving between
1860
and
1910
were unskilled. The bulk entered occupations in mining, manufacturing,
and construction (Hounvich,
1912).
While it is well-known that immigration
flows were large and cyclically sensitive, it is less well-known that emigration
flows followed the same pattern.
On average between
1870
and
1914,
one person emigrated from the United
States for every
three
that arrived.
While
emigration decreased and immigration
increased during good years,
as
might
be
expected, the annual proportion of
emigrants to immigrants never
fell
below
20
per cent (U.S. Bureau of the
Census,
1960;
Ulman,
1955;
Hourwich,
1912).
Emigration increased during
depressed years,
as
recent immigrants-the first to lose their jobs-decided
to return home. Ninety per cent
of
the Bulgarians who made up most
of
the
unskilled labor force in an Illinois steel mill had left town
by
the end of the
1908
depression. That year, immigration fell and
the
national proportion of
emigrants to immigrants rose to
75
per cent (Brody,
1969).
While more immigrants stayed than
left,
the large backflow contributed
to the instability of the unskilled labor force and high rates of turnover. Many
immigrants were single men or
married
men with families back home. Some
came to make their “stake” and then go back to Europe to buy land or pay
off
debts. Others returned to the United States when conditions improved.
The
fact that many immigrant workers wished to minimize their time in
the
U.S.
made union organization more difficult.
A
strike just lengthened
the time that a man was away from home and family. While there were some
spectacular strikes during periods of recovery, in most instances the unskilled
immigrant worker saw no alternative in seeking higher wages or better working
conditions other than to quit.
The
number
of
strikes may have increased
during a recovery, but
so
did quits4
The
instability of the unskilled workforce also reflected the absence of
conditions that would breed loyalty to the employer. There were few definite
prospects for upward mobility within
a
firm. An economist noted in
1898
that
the
daculty
with
which occupational (“trade”) mobility could
be
achieved
made quits and geographic (“place”) mobility more likely (Davidson,
1898).
There was a continuity of experience among the Lowell factory hands, their
French-Canadian and Irish replacements, and the southeastern Europeans
who
filled
the factories after
1880.
Each group brought to the factories a pre-
industrial work ethic that was more in tune with the seasons, more migratory,
and less habituated to industrial discipline. The high level
of
quits by the
‘At a large metal-working plant, the number
of
quits rose
from
581 in 1914,
a
depressed year, to
3,035
in
1916.
The
plant’s proportion
of
quits due to “dissatisfaction” rose from
27
per cent in 1914, to
34
per
cent
in
1915,
the
beginning
of
the
recovery;
by
1916 these accounted
for
64
per
cent
of
all
quits (Slichter,
1921a, p. 180).
266
/
SANFORD
M.
JACOBY
Lowell women and their successors was a form of resistance to the rigors of
factory life. Ellen Collins quit the mill at Lowell in the eighteen forties
complaining about her “obedience to the ding-dong of the bell-just
as
though
we were
so
many living machines” (Gutman, 1976, p. 28). Thirty years later,
managers of New England textile mills complained that absenteeism and
quits made it difficult to run their machines on the hottest summer days.
One manufacturer said in 1878 that, “our mill operatives are much like other
people and take their frequent holidays for pleasure and visiting” (Rodgers,
1978, p. 162). Forty years later, the quit rate at a Connecticut silk mill
quadrupled during the hot summer months of 1915 (Slichter, 1921a). Thus,
each group successively went through
the
process of internalizing factory
discipline; this was one of the props to high turnover before 1920.
Znuoluntay
mobility.
Of course, separations were not always voluntary.
Factory discipline was maintained by foremen who were free to fire at will.
Slichter (1921a, p. 202) termed the foreman’s harsh disciplinary regimen “the
drive system” and noted that it was intended to “inspire the worker with
awe and fear of
the
management, and having developed fear among them,
to take advantage of it.” Fear of unemployment made workers more submissive
when jobs were scarce.
A
relatively tight market tended to undermine the
foreman’s authority, forcing him to rely more heavily on dismissals to preserve
discipline.
Involuntary mobility had sources wider than
high
rates of dismissal, however.
In its cyclical and seasonal forms, unemployment regularly touched a large
portion of the labor force. In Massachusetts, the incidence of unemployment
was high even during relatively prosperous years like
1900-1906,
when at
least 20 per cent of the state’s industrial workers experienced some unem-
ployment each year. Even Massachusetts’ skilled trade union members were
not immune-an average of 29 per cent experienced some unemployment
each year between 1908 and 1916 (Keyssar, 1977).
Unemployment filtered widely through the labor force during both good
and bad times because of dismissals and seasonal instability. The most stable
industry in 1WQ-bread and bakery products-had monthly employment
levels that varied by
7
per cent. That same year, the industrial average
fluctuated by 14 per cent over the year, rising to 45 per cent in the automobile
industry
(U.
S.
Bureau of the Census, 1912). Seasonal employment instability
perpetuated the drive system. Activity became frenzied during the busy
season as firms rushed to
fill
orders. Capacity utilization and employment
levels rose
by
magnitudes unknown today.
Employers rarely sought to sustain the employment relationship during or
after downturns. Employment guarantee plans were in operation at only a
Industrial Labor Mobility
I
267
handful of firms
by
1920. Work-sharing was more prevalent, although it
usually was initiated by trade unions in cooperation
with
unionized employers.
Few firms made systematic efforts to rehire their workers after layoffs, even
in seasonal industries (Jacoby,
1981).
Thus, extensive mobility before 1920 was an indicator of a tenuous em-
ployment relationship and the fluidity
of
the labor market. But it is uncertain
whether this ceaseless movement improved a worker’s economic
status.
Slichter
(1941, p. 100) thought
so,
describing
the
pre-1920 labor market as one in
which “the behavior of men was dominated by the idea of opportunity-a
market in which there was a considerable number of men ready to give up
one
job
in the hope of getting a better one.” Slichter’s remark reflected his
concern that workers in 1940 were being driven more by a search for security
than opportunity, to the detriment of labor market efficiency. Thernstrom
(1973, p. 42), coming at this issue from a different body
of
research, was more
pessimistic about the economic benefits that mobility
had
brought to workers
before 1920.
He
asked whether
“the
exceptional earlier volatility of
the
American
working class, especially its least skilled workers, does not point to the existence
of a permanent floating proletariat made up of men ever on
the
move but
rarely winning economic gains as a result of spatial mobility.”
Labor
Mobility
After
1920
Between 1920 and 1950, labor markets grew increasingly rigid.
Workers became more attached to their employers and the ‘communities in
which they resided; employers became more committed to their workforce.
This shift occurred gradually and was
the
result of
a
set of changes that
reinforced each other over time.
While available turnover data show a downward trend in separations, they
must be used with caution. Three different series exist, covering the nineteen
tens, nineteen twenties, and post-nineteen thirties (see Table
1).
They are
based on widely differing samples and turnover calculation methods. Also,
the pre-1930 data were not based on a representative sample but came from
only those
firms
that collected and were willing to give out their turnover
figures. This was a select group, presumably with turnover rates below the
“true” mean6 Finally, the data from
the
nineteen tens do not cover the same
group of establishments each
year.
The
data from
the
twenties are median rates.
The
BLS
data from
the
other periods are weighted
averages, although
the
weighting factors differ between the nineteen tens and post-1930 periods.
The
BLS
introduced its data for
the
nineteen tens with this proviso:
“The
establishments from which
the
Bureau
of
Labor Statistics
has
secured labor mobility figures have necessarily been concerns which
had
the
figures
to
give, that is to say, concerns which had given more attention
than
most firms to their
force-maintenance problems
.
.
.In such establishments
the
instability is not likely to
be
so serious
as
in
the
general run of American concerns,
which
as
a rule pay little or no attention to
the
flow
of
labor in
and out and give little attention to its control” (Brissenden and Frankel,
1920,
p.
40).
268
/
SANFORD
M.
JACOBY
Geographic mobility rates for the years following
1920
show a similar down-
ward trend but only for manual workers. That is, persistence rates of manual
workers increased, while those of white-collar workers decreased, sometime
between
1920
and
1950.
As
Thernstrom
(1973,
p.
229)
notes, the evidence
necessary to precisely date this shift is lacking, “but all of the evidence points
to a reversal [after World War
I3
of the long-established pattern in which the
tendency of men to remain rooted in the community varied directly rather
than inversely with socioeconomic rank.”
Sources
of
Labor
Immobility
A
substantial minority of industrial firms began to experiment
with new employment methods after
1915.
The most striking development
of this period was
the
creation of centralized personnel departments responsible
for recruiting and selecting workers, duties formerly undertaken
by
foremen.
In other areas, such as wage determination, promotions, and dismissals,
personnel departments introduced uniform rules and procedures that curbed
the foreman’s discretion. Between
1915
and
1920,
the proportion of industrial
establishments
with
more than
250
employees which had personnel departments
rose from less than
7
per cent to
25
per cent (Jacoby,
1981).
The
establishment of personnel departments was in large part a response
to the labor relations problems produced by the war.
A
wave of labor unrest
swept the country between
1916
and
1920.
In fact, the ratio of strikers to all
workers during this period achieved the same level as that reached during
the famous strike years of
1934
and
1937.
Trade union membership nearly
doubled between
1915
and
1920
(Bing,
1921;
Wolman,
1936).
Labor unrest,
as
Slichter
(1919b,
pp.
83&339)
observed, was “causing employers to perceive
more and more clearly the necessity for a substantially liberal labor policy
in order to avoid the organization of their plants.”
TABLE
1
(PER
100
EMPLOYEES)
AVERAGE
MONTHLY
TURNOVER
RATES IN MANUFACTURING,
191&1959
Separations
Quits
Dismissals
plus
layoffs
1910-1918
1920-1929
1930-1939
1940-1949
1950-1959
10.1
4.9
4.7
6.6
4.4
7.4
3.7
n.a.
4.1
2.1
2.7
1.2
n.a.
2.1
2.1
Sources: 1910-1918. excluding 1916: Brissenden and Frankel, (1920,
p. 48);
1920-1929: Berridge
(1929,
pp.
64-65)
and
Berridge
(1930.
p.
125); 1930-1959:
U.S.
Bureau
of
Labor Statistics (1975,
p.
130). Data
fmm
the thirties lump together
quits and dismissals. Data
from
1940-1959 calculated on the assumption that dismissals accounted for one-half
of
all
“other” separations (MedoB. 1979, p. 389).
lndustrial Labor Mobility
I
269
Personnel management placed sharp limits on the drive system and thus
functioned as a substitute for unionism. Foremen no longer were
free
to
dismiss workers at will, strike rules replaced haphazard allocative practices,
and efforts were made to stabilize employment and provide a modicum of
employment security. Most unions had evolved procedures for the equitable
allocation and payment of their members, and for protection against layoff
and unfair dismissal. Personnel management provided similar benefits and
extended them to
all
of a firm’s workers, thereby undercutting the urgency
for organization.
This
led
Slichter (1921b, pp.
261-263)
to
ask
whether personnel
management wasn’t “a backfire against unionism,
an
attempt to forestall changes
in industrial government
by
changes in managerial methods.”
Focus
on
turnover. Many reforms were explicitly intended to reduce labor
turnover. In fact, no issue received more attention in the wartime personnel
management literature than turnover and the importance of reducing quit
and dismissal rates.
The
wartime jump in separation rates undoubtedly pre-
cipitated some of this concern,
bit
such rates had also been high in the past.
The
unprecedented concern with turnover had to involve more than an
increase in its level.
One reason for the concern with turnover was the growing belief that there
was
a
clear relation between quits and worker dissatisfaction, which many
were beginning to refer to by the military term “low morale.” In modern
parlance, “exit” was increasingly seen as
a
worker alternative to “voice.” As
Lescohier (1919, p. 89) put it, trade unions “voice their protest against bad
conditions by sending a committee to see the firm.
The
unorganized voice
their protest
by
‘asking for their time’.”
Personnel managers began to accept the notion that the same conditions
that give rise to low morale and quits in the present might lead to strikes in
the
future.
They
reasoned
if
such conditions could
be
ascertained and remedied,
morale would improve, quits would
fall,
and the probability of a successful
future organizing drive would
be
lessened (Colvin,
1919).
Exit interviews
became a reglar personnel practice
in
more and more firms. At one Philadelphia
firm, “Complaints
are
heard of nagging foremen, lost time in waiting for
work, and other complaints .
.
. .These
are
investigated, and if the fault is
with us, it is remedied’ (Williams,
1917,
p.
188).
The idea soon developed
of using turnover data as a “morale index, the precursor to modern attitude
surveys. Slichter (1919a, p.
336)
noted that firms collected such data because
“management wishes its turnover figures to
be
an indication
of
the character
of its industrial relations, of the degree of discontent within the plant.”
9,
270
I
SANFORD
M.
JACOBY
Personnel departments also developed rudimentary grievance procedures
to “give every man a chance to
knock
(Fisher,
1917a,
p.
26)
rather than turn
to a union or quit. Employee counseling, welfare work, and suggestion boxes
were prescribed, ostensibly to reduce turnover, although their latent function
lay more in the realm of union avoidance (Jacoby,
1981;
NACT,
1921).
Thus, a low level of quits became
the
desideratum of a well-managed or
“high trust” firm
(Fox,
1974).
It was an indication that the firm was sensitive
to its workers’ problems and provided outlets for the voicing of dissatisfaction.
Some personnel managers even went
so
far as to claim that
the
personnel
department’s effectiveness could be measured by the trend in quit rates
(Clothier,
1916).
This response
to
quit rates also had its coercive side. Various pecuniary
welfare programs (e.g.
,
pension, profit-sharing, and stock ownership plans)
were widely espoused as a way to strengthen links between a firm and its
workforce. But by tying
hefty
benefits to an unbroken service record, these
programs also increased the cost to
the
worker
of
a dismissal or strike. The
industries where these programs were most prevalent tended to
be
those
with little or no union organization. Employers said that the programs had
a quieting effect on labor disturbances and contributed to “loyalty” and
“co-
operation” (NICB,
1925).
When a strike occurred at Studebaker in
1919,
the
firm’s body painters and trimmers refused to participate for fear they would
lose their welfare benefits (Adams,
1921).
An economics professor at North-
western (Heilman,
1918,
p.
540)
warned in
1918
that tenure-based rewards
were “likely to arouse the suspicion of the employees that the employer is
endeavoring to shackle them.
They
are
likely
to
feel that he simply wishes
to protect himself against any possibility of group action on their part.”
Turnover cost and employment stabilization.
There is little evidence that
concern with turnover was due to an increase in separation costs associated
with
technological change of a firm-specific character. In fact, manufacturing
technology was becoming less firm-specific and idiosyncratic than it had been
in the nineteenth century.
Firms
were introducing on a wide scale a technology
that de-emphasized skill and lowered training costs. This reduced the cost
of each separation for the employer
by
making workers more interchangeable.
And it made it easier, or less costly, for workers to change jobs. Discussions
of turnover ofken noted that the opportunity for movement from plant to
plant was facilitated
by
the
standardization of equipment and the ease with
which workers could learn a new job. Douglas
(1921,
p.
74)
observed that
“The very process of machinery which made work more specialized, made
the
worker less specialized. He was now transferable
. .
.
.A
machine tender
lndustrial Labor Mobility
1
271
who
has
learned the general principle of caring for a machine can tend ribbon-
weaving machinery as well as shoe-making. He really is an interchangeable
part in the industrial mechanism.”
Of course, even though there was a downward trend in training costs, there
still was a fixed cost attached to breaking in a new worker. Personnel managers
were well aware that turnover entailed a replacement cost and repeatedly
cited a 1912 study (Alexander, 1915) that had estimated these costs at several
industrial firms. But, these costs, which were usually not large, were not the
real source of personnel managers’ new concern over turnover. They cited
these costs in order to legitimize reforms- principally
the
restriction of the
foreman’s discharge prerogatives and the provision of employment security-
that were thought to
be
necessary for substantially different reasons.
To
understand the need for this strategy, one must appreciate the diversity
of views within
the
firm over the desirability of reform. On
the
one hand
were managers, principally in the personnel department, who were deeply
concerned about the long-term effects of the drive system. Unrestricted
dismissals were said to be the cause of “a surprisingly large percentage of
labor dficulties” (Bloomfield, 1915, p. 22) that could easily be avoided by
restraining foremen and adopting less punitive forms of discipline. Employment
insecurity and a failure to recognize a worker’s “sense of property rights in
jobs,” said Tead (1918, p.
70),
“will
give rise to a stronger and stronger
consciousness of injustice or to unconscious suppressed desires.”
On the other hand, production managers and line officials were satisfied
with the drive system. It allowed for a flexible adjustment to shifts in demand
and, until the war, had been effective at holding worker expectations and
unit costs in line. Slichter (1921a, p.
427)
thought that the drive system
was
profitable, given a short-run perspective on what
he
called “the interests of
employers as a class.
Production managers often had this limited perspective.
They were men, who, as one personnel manager put it, “consider labor a
commodity, to be bought and
sold
on the open market when trade is brisk
and disposed with at will when demand subsides” (Kennedy, 1920, p. 228).
They were skeptical of
the
wisdom of protecting workers
horn
dismissals,
even when labor relations problems threatened, believing that “labor will
simply take advantage of liberal treatment” (Slichter, 1919b, p. 827).
Faced with this resistance, personnel managers tried to ‘‘sell’’ their reforms
by appealing to other managers’ costconsciousness rather
than
to their political
acumen. Replacement estimates served as useful propaganda, allowing per-
sonnel managers to portray themselves
as
hard-headed and efficiency-minded.
Such estimates provided “figures that we can wave in front
of
the faces
of
factory managers and say, ‘it costs you
so
much for turnover and we can prove
272
I
SANFORD
M.
JACOBY
it.’ It seems that there
are
a large number of managers who can
be
convinced
only by combining an appeal to their altruism with an appeal to their pock-
etbooks’’ (Fisher, 1917b, pp.
65-66).
Of course, there also was an eagerness to implement these reforms because
they effectively increased the status and power
of
the personnel department.
Potential savings in turnover costs were used to justify
a
transfer of authority
from foreman to personnel manager. Also, personnel managers recommended
that these savings should
be
a criterion for “determining the amount of money
that shall
be
expended on the personnel department” (NACS,
1918,
p. 328).
Some managers were clearly skeptical of
the
cost figures thrown about
by
personnel managers.
The
superintendent of a large textile mill said that
he
considered
the
cost of replacing
a
skilled worker “inconsequential” and reported
that the departure of less skilled workers resulted in “no appreciable loss”
(Emmet, 1919, p. 14).
A
prominent industrial engineer complained that
turnover had become a “fetish” among personnel managers and that if the
man to
be
replaced “is an unskilled or semi-skilled laborer,
as
so
many of
today’s employees are, the cost of replacement is small” (Kimball, 1919,
p. 324). Moreover, as another engineer pointed out,
a
low rate of turnover
sometimes raised labor costs because “in such cases it becomes necessary to
increase the daily
or
hourly rates occasionally, if not regularly, which is not
very pleasing to some types
of
managers” (Colvin, 1919, p.
37).
Overall, however, it is difficult to say whether wartime efforts to stabilize
employment were more strongly motivated
by
economic or other considerations.
Writing in 1919, Slichter (1921a) was pessimistic about the liberalizing effect
that strictly pecuniary considerations would exert on employers. He argued
that ultimately the decision to reform employment practices did not turn on
cost considerations but on a recognition of the long-term labor relations
threats posed
by
the drive system. Slichter’s remark was prescient. Devel-
opments after
1920
showed that replacement cost arguments carried little
weight when labor relations were less problematic.
The Twenties: Labor Markets
The pace of employment reform slackened during the twenties
despite the best efforts of the personnel managers.
The
proportion of large
industrial firms with personnel departments grew more slowly than before,
rising from 25 per cent in 1920 to
34
per cent in 1929 (Jacoby,
1981).
Some
firms cut back or completely eliminated their personnel departments and at
these firms “there was a reversion to older methods” (Wolman and Peck,
1933, p.
830).
Foremen again were given considerable discretion over dismissals.
Other wartime innovations, such as internal promotion plans and the payment
Industrial Labor Mobility
I
273
of layoff compensation, did not spread beyond the minority of firms that had
adopted them during the war (Jacoby,
1981).
This evidence conflicts with the
conventional wisdom that the twenties was a “boom” period for personnel
management (e.g.
,
Wood,
1960).
Another piece of conventional wisdom-that this was a decade of widespread
prosperity-is unsupported
by
evidence from the manufacturing sector.
The
deceleration of employment reform owed much to the fact that the manu-
hturing labor market was a buyer’s market during the twenties. Manufacturing
employment did not grow at all between
1919
and
1929;
average annual
unemployment rates in manufacturing were higher between
1923
and
1927
than for any other five-year period since
1900,
excluding the depression years
of
1908, 1914-1915,
and
1921-1922
(Lebergott,
1964;
Douglas,
1930).
In
20
out of
26
manufacturing industries, average hourly rates declined between
1920
and
1929.
Partly
as
a result of the weak labor market, union membership
in manufacturing fell by
58
per cent over the decade and the threat of strikes
receded.
Labor market slack, as well as
a
widespread introduction of labor-
saving technology, obviated the need for sophisticated effort controls (Bernstein,
1960;
Wolman,
1936;
Jerome,
1934).
Despite these developments, some firms maintained or expanded their
personnel programs during the twenties. They were a visible minority that
employed, according to one survey (NICB,
1929),
about
20
per cent of all
manufacturing workers in
1929.
Here efforts were made to stabilize employment,
encourage attachment to the firm, restrict dismissals, and cement ties to the
incumbent workforce. Consequently, some of the decline in turnover during
the twenties may
be
attributed to
the
more liberal employment policies that
these firms had adopted during the war.7
But it would
be
incorrect to argue, as did
Ross
(1958)
in his seminal article
on industrial feudalism, that the overall decline in quit rates largely was due
to a “flourishing”
of
modern personnel administration after
1920.
Personnel
management hardly flourished during this period. Instead,
the
decade’s
downward trend in mobility was caused by the depressed state of the man-
ufacturing labor market-nominal wage stability, stagnant employment, tech-
nological unemployment, and slow absorption of the unemployed (Lubin,
1929)-as
well as
post-1920
changes in the labor force, discussed below.8
‘A
survey of 17 “pioneer” firms
that
utilized advanced personnel management techniques found that
they had an average monthly separation rate of 3.8
per
cent between 1924-1929, which was considerably
below the level prevailing during the preceding
decade
(Balderston, 1935, p. 387).
Wichter (1929, p. 429), who was overly optimistic about the extent to which personnel management
techniques and principles had spread during
the
twenties, nevertheless thought that “the most important
single influence” on quit rates was
“the
relative abundance of men and scarcity of
jobs
which have
existed
, ,
.almost continously since 1920.”
274
I
SANFORD
M.
JACOBY
Workers in industries with stable or declining employment levels were
reluctant to quit their jobs out of a fear of prolonged unemployment. This
further depressed quits by raising the average age of the manufacturing labor
force; older workers were less likely to quit.e Workers were timorous about
trying their luck even in expanding industries. This was demonstrated by
Woytinsky‘s
(1942)
comparison of
the
relationship between accession and
separation rates in the pre- and
post-1921
periods. Before
1921,
monthly
accession and separation rates moved together, with increases in separations
occurring at about the same time-as increases in accessions. After
1921,
upward changes in separation rates lagged behind upward changes in accession
rates
by
several months in firms that were expanding employment during
the twenties. Woytinsky
(p.
16)
speculated that this meant that “workers were
already scared by the threat of unemployment and no longer changed their
jobs as freely as before the war.”’O
Employers had little reason to be concerned with turnover, given a labor
surplus,
high
labor productivity, and low unionization and quit rates.
As
a
result, the management literature during the twenties was nearly devoid of
articles on labor turnover. One researcher found firms still collecting turnover
data in
1927,
but he found few that attached any importance to
the
figures
(Houser,
1927).
The
Twenties: Labor Force Characteristics
Labor mobility during the twenties was further depressed by
the enactment of immigration restriction legislation at
the
beginning
of
the
decade. First, the law checked the huge backflow of emigrants by making it
more dificult for an emigrating alien to re-enter the United States at some
future time. Emigration fell during the twenties to less than half of its prewar
level
(U.S.
Bureau of the Census,
1960).
Workers who might have emigrated
before
the
change in
the
law instead sank down roots and committed themselves
to particular employers and communities.
Second, the law caused a sharp decline in immigration levels after
1920.
Before
1920,
the manufacturing labor force had included a high proportion
of unmarried or unattached immigrant men, with families or wives in Europe.
me
proportion of
the
manufacturing labor force over
45
years of age increased from
25
per cent in
1920
to
30
per cent in
1930
(U.S.
Bureau
of
the Census,
1923;
1933).
‘“The
only aggregate data on turnover during
the
twenties are based on an unrepresentative sample of
firms that more than doubled their employment between
1920
and
1929,
a time when the rest of the
manufacturing sector was nearly stagnant.
We
do not know what happened to quit rates in the rest of
the manufacturing sector, a problem that
Ross
was unaware of, but it seems reasonable to infer that they
also
declined. Also, because the sample was unrepresentative and because it overweighted large
firms,
the data for the twenties were biased downward relative to the pre- and post-sample years.
Thus,
the
decline in turnover rates after
1920
was less dramatic than the data in Table
1
indicate uacoby,
1981).
lndustrial Labor Mobility
I
275
But now this relatively mobile group became a smaller proportion of the
labor force.” More broadly, the decline in immigration meant that fewer
newcomers were entering the labor force after
1920.
Newcomers were, by
definition, persons without strong ties to a particular place, hence potentially
mobile. Most had never before worked in a factory, which, as we have seen,
made for a propensity to quit. Conversely, the decline in immigration meant
that the children of the immigrant who flocked here between
1890
and
1915
now made up a larger proportion of
the
manufacturing labor force.I2 Presumably
they remained in the workplaces and communities where their parents had
finally settled. Thus, changes in the labor force also contributed to the post-
1920
drop in job and geographic mobility.
Employment Policy
and
the Labor Force,
1930-1950
During the first years of the Great Depression, the future of
personnel management was in doubt. It was uncertain whether the minority
of progressive
firms
that
had
eliminated some of
the
more egregious features
of the drive system would continue to do
so.
The prospects for reform were
even
bleaker
in
the
rest
of
the industrial sector. Here, employers worried
less about morale and workers enjoyed little in
the
way of dismissal and layoff
protection, allocation by rule,
or
benign methods for eliciting effort. Early
in
1933,
Leiserson
(1933,
p.
114)
gloomily wrote that “the depression has
undone fifteen years or
so
of good personnel work,” and
he
warned that “labor
is going to look to legislation and not to personnel management for a solution
of the unemployment problem.”
But there was no cause for pessimism after
1933.
Personnel management
and employment reform prospered in the wake of the National Industrial
Recovery Act. New or beefed-up personnel departments were given the
authority to ensure compliance to
the
NRA
codes as well as subsequent New
Deal legislation (Tead and Metcalf,
1933).
The act also touched
off
intense
organizing activity; union membership more than tripled between
1933
and
1939
(Ulman,
1961).
As
during World War I, the growth of union organization
was accompanied by a dramatic increase in the number of personnel de-
partments. According to various surveys (Jacoby,
1981),
the proportion of
ILThe
proportion of gainfully employed white males in
the
U.S.
who were native-bom was:
.77
(1900),
.75
(1910),
.77
(1920),
.83
(1930), and
.90
(1950)
(U.S.
Bureau of
the
Census, 1914;
1923;
1933;
1953).
The
proportion undoubtedly was lower in manufacturing, although
the
trend probably was similar.
‘The
decennial rate of change in
the
ratio
of
native-born whites, aged 15-24, of foreign
or
mixed
parentage (“second generation”) to the total white population, aged 15-24, was 1.2 per cent (1910-1920)
and
4.9
per cent
(192&1930).
For
the
Northeast and
North
Central regions,
the
rate of change in
the
ratio
of
native-born whites,
all
ages, of foreign parentage only,
to
the
total white population,
all
ages, was
1.8
per
cent (leoCrl9lO), 6.4 per cent (1910-1920). and 12.2
per
cent (1920-1930)
(U.S.
Bureau
of
the
Census, 1954).
276
/
SANFORD
M.
JACOBY
large firms with personnel departments rose markedly between
1919
and
1935;
the
proportion of the industrial workforce covered by a personnel
department shot up from
20
per cent in
1929
to about
70
per cent in
1935.
The
parallel expansion of unionism and personnel management was an
impetus for the rapid diffusion
of
employment policies that undermined the
drive system. At some firms, unions were directly responsible for this; at
others, threat effects and the greater power now wielded by the personnel
department accelerated
the
implementation of rule-bound employment pro-
cedures.
The
outlines of the modem internal labor market, which
had
first
appeared during World War
I,
took shape throughout the industrial sector
between
1933
and
1950.
The trend during the twenties toward the decentralization of allocative
authority
was
reversed as foremen lost much of their power. Although over
half
of
all
industrial firms gave
their
foremen sole authority to dismiss employees
in
1936,
this
had dwindled
to
less than
25
per cent
by
1948
(NICB,
1937;
NICB,
1948).
Promotion methods changed
as
job ladders and internal promotion
plans became more widespread.I3
The
fact that promotion plans and guarantees
of internal hiring were not implemented until there was
a
union threat suggests
that
their
value to the firm as a control device (Stone,
1975)
was less important
than their value to employees as a form of security and protection against
arbitrary assignment.
An important change in employment policy was the emphasis on seniority
as a key factor in age, promotion, layoff, and rehiring decisions. Seniority
had been used as an allocative criterion before the depression but few firms
had
adopted the rigid and all-encompassing approach that unions now
de-
manded.
The
depression intensified long-standing resentment of
the
foreman's
allocative powers and heightened workers' concern with security. As unions
consolidated
their
power during the war, they sought to apply seniority
criteria to wider spheres of employment decisions (Golden and Ruttenberg,
1942;
Slichter
et
al.,
1960).
These developments solidified the downward trend in job mobility that
had appeared during the twenties. First, voluntary mobility was reduced by
the
wider
use of such practices as seniority-based employment policies and
internal hiring and rehiring commitments. These raised the cost of mobility
by linking a worker's security and opportunity more closely to his tenure
with a particular employer (Ulman,
1965).
Second, the discharge rate was
lowered by the proliferation
of
dismissal rules and protections, although
a
'The
proportion
of
firms
with
internal promotion plans rose
from
14
per
cent in 1929 to
28
per cent
in
1935;
this proportion rose steadily during the war.
By
1950,
80 per cent of a surveyed group
of
firms
reported
that
"all
or most" of their manual jobs were filled from within oacoby, 1981).
Industrial Labor Mobility
I
271
heavier use of layoffs as an adjustment mechanism (Medoff, 1979) may have
counterbalanced the decline in dismissals (see Table
1).
However, workers
on layoff now were less likely to change jobs. This was due to the provision
of explicit rehiring commitments, a lengthening of the period during which
seniority was retained after and the availability of unemployment
insurance after
1935.
Unionization
had
an additional, negative effect on mobility. Grievance
systems, bargaining, and strikes provided a mechanism for expressing dis-
satisfaction and influencing working conditions that was an alternative to the
quit (Shister,
1950).
Personnel departments in nonunion firms replicated this
effect through sophisticated attitude surveys, employee counseling, and “open
door” grievance policies, much as their forebears had done during World
War
I.
Changes in the labor force also contributed to
the
downward trend in
mobility. Immigration continued to
be
a declining source of new labor force
entrants. Also, the average age of the manufacturing labor force increased
between 1940 and
1960;15
this was both an effect and cause (Pencavel, 1970)
of the decline in job mobility. Finally,
the
drop in job mobility reinforced
the post-1920 propensity
of
manual workers to remain rooted in their com-
munities. Alternatively,
the
decline in geographic mobility, by creating stable,
working-class communities centered around the workplace, may indirectly
have fostered lower levels of job mobility by making it easier to organize
unions.l6 Moreover, it is possible that workers’ overwhelming concern with
job security during this period partly stemmed fi-om a desire to protect new
forms of communal stability from the disruption caused by unemployment.
Conclusion
Postwar labor economists were fearful that the development of
job security measures and other internal labor market arrangements was
hindering the flexibility of a market-based system of labor allocation and
incentives. Fears that then were being expressed about the growing power
of the labor movement were merged with a concern. that this power was
being used to promote security and status at the expense of worker’s liberty
to move freely in the labor market (see, e.g.,
Kerr,
1954).
I‘The
proportion of manufacturing
firms
where seniority
was
lost after layoff
fell
from 14
per
cent in
The
proportion
of
manufacturing workers over
45
years
of
age rose from
26
per cent in 1940, to 31
‘This is
an
interesting hypothesis that deserves further study. Labor historians now are beginning to
1927 to
3
per cent in
1936
to less than 1 per cent in 1964 (Jamby, 1981).
per cent in
1850,
to
36
per cent in
1960
(U.S.
Bureau
of
the Census, 1943; 1955; 1967).
explore
the
role
of
communal organizations in the upsurge
of
thirties unionism.
278
/
SANFORD
M.
JACOBY
These fears were not particularly well grounded. First, the decline in
mobility had sources wider than the new employment policies stimulated
by
the spread of unionism.
The
relatively slow growth of manufacturing em-
ployment after
1920
and the concomitant aging of the labor force were a
powerful deterrent to voluntary job mobility. Also, a part of the decline can
be
attributed to
an
historic change in the character of the labor force. After
1920,
the waves of immigration and emigration subsided. Working-class life
was stabilized
as
the
first generation sank down roots, formed families, and
developed social networks. This encouraged the second generation to remain
in their parents’ communities and develop steady working habits (Piore,
1975).
Second, labor economists after
1960
developed turnover-based theories of
labor retention which implied that the long-run decline in labor market
fluidity was not inherently inefficient. The immobilization produced
by
internal
labor markets now was shown to
be
allocatively efficient because it reduced
an employer’s separation costs. However, these theories overemphasize the
extent to which internal labor markets were formed in response to economic
constraints.
The historical record suggests that competitive forces did not spontaneously
generate a new set of personnel policies. The rapidity with which employment
reforms were introduced after
1933
largely was due to the growing power of
the unions and the ascendance of the personnel department over the more
production-oriented branches of management. Changes in what Solow
(1980,
p.
3)
has termed “social conventions or principles of appropriate behavior”
pushed firms to adopt policies that decasualized employment and curbed
the
excesses of the drive system.
With the benefit of hindsight, we may observe that these policies often
enhanced efficiency through their effect on turnover and morale, or
by
stim-
ulating programs to upgrade the workforce. However, this result was not
always apparent to those who managed firms in transition, as there was
continual skepticism among managers that internal labor market arrangements
would lower costs. Efficiency incentives were neither strong nor obvious
enough to produce the modern internal labor market.
Thus,
postwar
labor economists were correct in recognizing that institutional
factors, rather than turnover costs, were a primary spur to the adoption
of
internal labor markets and to lower levels of mobility, although they failed
to realize that these outcomes might be allocatively efficient. Turnover-based
”Doeringer and Piore
(1971,
p.
28)
noted that
it
was
possible to describe internal labor market formation
as
a
process
“initiated solely at
the
discretion
of
management
. . .
by
managers seeking to minimize their
labor costs.”
Industrial
Labor
Mobility
I
279
theories of labor retention provide a rationale for the persistence
of
these
outcomes, although they overemphasize the causal significance
of
employers’
economic rationality. Neither theory is a sufficient explanation of the long-
run drop in mobility because each ignores changes in the demographic and
social characteristics of the industrial labor force over time.
There is little reason to expect a reversal of the decline in labor mobility.
Previous high levels of mobility were linked to labor force characteristics that
are unlikely to reappear, including mass immigration and a pre-industrial
work ethic. Moreover,’ Solow’s principles
of
appropriate behavior have become
embedded in a structure of law, employee expectations of fair treatment, and
norms of professional management that recognize the economic benefits of
employment stability and equity.
References
Adams, James P.
“A
Common Sense Attack on Labor Turnover,”
Zndustrial Management,
Alexander, Magnus. “Cost of Hiring and Firing Men,”
Engineering Magazine,
XLVIII
(1915),
Allen, James P. “Changes in the American Propensity to Migrate,”
Annals
of
the Association
Balderston, C. Canby.
Executive Guidance
of
Zndustrial Relations.
Philadelphia, PA: University
Bartel, Ann
P.
“The Migration Decision: What Role Does Job Mobility Play?,”
American
Bernstein, Irving.
The Lean Years.
Boston,
MA:
Houghton Main,
1960.
Benidge, W.A. “Labor ’hrnover,”
Monthly
Labor
Reuiew,
XXIX (July,
1929),
64-65.
Bing, Alexander
M.
Wartime Strikes and Their Adjustment.
New York: E.P. Dutton,
1921.
Bloomfield, Meyer. “The New Profession of Handling Men,”
The Annals,
LXI (September,
Brissenden, Paul and Emil Frankel. “Mobility of Labor in American Industries,”
Monthly
LXII
(November,
1921), 298-304.
733-736.
of
American Geographers,
LXVII (December,
1977), 580-590.
of Pennsylvania Press,
1935.
Economic Reuiew,
WIII (December,
1979), 775-786.
.
“Labor Turnover,”
Monthly
Labor
Re&,
XXX
(February,
1930), 124-126.
1915), 12&126.
Labor
Reuieu,
X (June,
1920), 3&-52.
.
Labor
Turnouer in Industry.
New York: Macmillan,
1922.
Brody, David.
Steelworkers in America: The Nonunion Era.
New York: Harper
&
Row,
1969.
Clothier, Robert. “The Function of the Employment Department.” In U.S. Bureau of Labor
Statistics
Bulletin
No.
196.
Washington, D.C.:
U.S.
Government Printing Office,
1916,
pp.
Cole, Robert
E.
Work, Mobility and Partidpation.
Berkeley, CA: University of California
Colvin, Fred H.
Labor
Turnouer, Loyalty and Output.
New York: McGraw-Hill,
1919.
Davidson, John.
The
Bargain Theory ofwages.
New York: G.P. Putnam’s Sons,
1898.
Doeringer, Peter and Michael Piore.
Znternal
Labor
Markets and Manpower Analysis.
Lexington,
Douglas, Paul H.
American Apprenticeship and Industrial Education,
New York: Columbia
7-14.
Press,
1979.
MA:
Heath,
1971.
University Press,
1921.
.Real Wages in the United States,
1890-1926.
New York: Macmillan,
1930.
Emmet, Boris. “The lbrnover of Labor.” In Federal Board for Vocational Education,
Bulletin
No.
46.
Washington, D.C.:
U.S.
Government Printing Office,
1919,
pp.
1-60.
280
I
SANFORD
M.
JACOBY
Fisher, Boyd. “How to Reduce Labor Turnover,”
The Annub,
LXXI
(May,
1917a), 10-32.
.
“Determining Cost of ’hrnover of
Labor.”
In U.S. Bureau of Labor Statistics,
Bulletin
No.
227.
Washington, D.C.: U.S. Government Printing Office,
1917b,
pp.
60-66.
Fox, Alan.
Beyond Contract: Work,
Power
and Trust Relations.
London: Faber
,&
Faber,
1974.
Ginger, Ray.
“Labor
in a Massachusetts Cotton Mill:
1853-1860,”
Business History Review,
Gitelman, Howard
M.
Workingmen
of
Waltham: Mobility in American Urban lndrtstrial De-
oelopment.
Baltimore, MD: Johns Hopkins University Press,
1974.
Golden, Clint and Harold J, Ruttenberg.
The Dynamks
of
lndustrial Democracy.
New York:
Harper and Brothers,
1942.
Gutman, Herbert.
Work, Culture and Society in
industrializing
America.
New York: Knopf,
1976.
Heilman, Ralph. “Do
You
Keep
Your
Men Too Long?,”
System,
XXXIII (April,
1918),
539-
543
*
Hopkins, E.M. “Advantages of Centralized Employment,”
The Annab,
w(I (May,
1917),
1-9.
Hourwich, Issac
A.
lmmigration and Labor: The Economic Aspects
of
European lmmigration
to
the
U
.S
.
New York: G.P. Putnam’s Sons,
1912.
Houser,
J.
David.
What the Employer Thinks.
Cambridge,
MA:
Harvard University Press,
1927.
Jamby, Sanford
M.
The Origins
of
lnternal Labor Markets in American Manufacturing Firms,
191&1940.
Unpublished Ph. D. dissertation, University of California, Berkeley,
1981.
Jerome, Harry.
Mechanization in Industry.
New York: National Bureau of Economic Research,
1934.
Kennedy, Dudley
R.
“The Future of Industrial Relations,”
Industrial Management,
LJX
(March,
1920). 226-229.
Kerr, Clark. ‘The Balkanization of Labor Markets.” In E. Wight Bakke
et al., Labor Mobility
and Economic Opportunity.
Cambridge,
MA:
Technology Press
of
MIT,
1954,
pp.
92110.
Keyssar, Alexander.
Men Out
of
Work:
A
Sock1
History
of
Unemployment
in
Massachusetts,
187&1916.
Unpublished Ph. D. dissertation, Harvard University, Cambridge,
1977.
Kimball, Harry
W.
“Some Reasons for
Labor
Turnover,”
lndustrial Management,
LVII (April,
1919), 322-326.
Lebergott, Stanley.
Manpower in Economic Growth.
New York: McGraw-Hill,
1964.
Leiserson, William
M.
“Personnel Problems Raised by the Current Economic Crisis,”
The
Lescohier, Don D.
The Labor Market.
New York: Macmillan,
1919.
Lubin, Isador.
The Absorption
of
the Unemployed by American Industry.
Washington, D.C.:
Medoff, James. “Layoffs and Alternatives Under Trade Unionism in
U.
S.
Manufacturing,”
National Association
of
Corporation Schools (NACS).
Seventh Annual Proceedings.
New
York:
XXVIII (March,
1954), 67-91.
Management Review,
XXII
(April,
1933), 114115.
Brookings Institution,
1929.
American Economic Review,
LXIX
(June,
1979), 380-395.
1919.
NACT (National Association
of
Corporation Training).
Ninth Annual Proceedings.
New York:
1921.
Navin, Thomas
R.
The Whitin Machine Works: A Textile Machine
ry
Company in an Industrial
NICB (National Industrial Conference Board).
lndustrid Pensions
in
the
U.S
.
New York:
1925.
.
Industrial Relations Programs in Small Plants.
New York:
1929.
.
Personnel Practices Governing Factory and Ofice.
New York:
1937.
.
“Personnel Practices in Factory and Office,”
Studies in Personnel Policy No.
88.
Organization
for
Emnomic Co-operation and Development (OECD).
Wages and
Labour
Mobility.
Pencavel, John.
An Analysis
of
the Quit Rate in American Manufacturing Industry.
Princeton,
Village.
Cambridge,
MA:
Harvard University Press,
1950.
New York:
1948.
Paris:
1965.
N
J:
Princeton University, Industrial Relations Section,
1970.
lndustrial
Labor
Mobility
I
281
Piore, Michael. “Notes for. a Theory of Labor Market Stratification.” In Richard C. Edwards,
Michael Reich, and David
M.
Gordon, eds.,
Labor Market Segmentation.
Lexington,
MA:
Heath,
1975,
pp.
125-150.
Portenar,
A.
J.
“Centralized Labor Responsibility from a Labor Union Standpoint,”
The Annals,
LXXI
(May,
1917), 191-201.
Rodgers, Daniel T.
The Work Ethic in Industrializing America, 18S1920.
Chicago, IL:
University of Chicago Press,
1978.
Ross, Arthur
M.
“Do We Have a New Industrial Feudalism?,”
American Economic Review,
XLVIII (December,
1958), 903-920.
Shister, Joseph. “Labor Mobility: Some Institutional Aspects,”
Proceedings,
Third Annual
Meeting of
the
Industrial Relations Research Association. Madison, WI:
1950, 42-59.
Slichter, Sumner H. “The Scope and Nature
of
the Labor Turnover Problem,”
Quarterly
Journal
of
Economics,
XXXIV
(November,
1919a), 329-345.
.
“The
Management
of
Labor,”
Journal
of
Political Economy,
XXVII (December,
1919b), 813-839.
.
The Turnmer ofFactory Labor.
New York: Appleton,
1921a.
. “Review of Personnel Relations in Industry,”
Administration,
11
(August,
1921b),
.
“The Current Labor Policies of American Industries,”
Quarterly Journal
of
.
Union
Policies
and Industrial Management.
Washington, D.C.: Brookings In-
,
James J. Hedey, and E. Robert Livemash.
The lmpact
of
Collectioe Bargaining
Solow, Robert
M.
“On Theories of Unemployment,”
American Economic Reuiew,
LXXXIII
Stone, Katherine. “The Origins of Job Structures in the Steel Industry.” In
R.
C. Edwards
Thernstrom, Stephen.
The Other Bostonians: Pmerty and Progress in the American Metropolis,
Tead, Ordway.
Instincts in Industry.
Boston,
MA:
Houghton Main,
1918.
261-263.
Economics,
XLIII (May,
1929), 393-435.
stitution,
1941.
on Management.
Washington, D.C.: Brookings Institution,
1960.
(March,
1980), 1-11.
et al.,
eds.,
Labor Market Segmentation.
Lexington,
MA:
Heath,
1975,
pp.
27-84.
1880-1970.
Cambridge,
MA:
Harvard University Press,
1973.
and Henry
C.
Metcalf.
Labor Relations and the Recooery Act.
New York: McGraw-
Ulman, Lloyd.
The Rise
of
the National Trade Union.
Cambridge,
MA:
Harvard University
.
“The Development ofTrades and Labor Unions.” In Seymour Harris,
ed.,
American
. “Labor Mobility and the Industrial Wage Structure in the Postwar United States,”
U.S.
Bureau of Labor Statistics.
Handbook
of
Labor Statistice1975 Reference Edition.
U.S.
Bureau
of
the Census.
Census
of
Manufactures:
1909.
Part I. Washington, D.C.: U.S.
.
Thirteenth Census
ofthe
US
.:
1910.
Volume
4,
Population-Occupational Statistics.
.
Fourteenth Census
of
the
US.:
1920.
Vol.
4,
Occupations. Washington, D.C.:
.
Fijteenth Census ofthe
US.:
1930.
Vol.
5,
General Report on Occupations.
.
Sixteenth Census ofthe
US.:
1940.
Vol.
3,
Pt.
1,
The Labor Force. Washington,
.
U.S.
Census ofl‘opulation: 1950.
Vol.
4,
Special Rpts., Part
1,
Employment and
.
US.
Census ofl‘opulation: 1950.
Vol.
4,
Special Rpts., Part
3,
Nativity and
Hill,
1933.
Press,
1955.
Economic History.
New York: McGraw-Hill,
1961,
pp.
366-419.
Quarterly Journal
of
Economics,
LXXIX
(February,
1965), 73-97.
Washington, D.C.:
U.S.
Government Printing Office,
1975.
Government Printing Office,
1912.
Washington, D.C.: U.S. Government Printing Office,
1914.
U.S. Government Printing Office,
1923.
Washington, D.C.: U.S. Government Printing OEice,
1933.
D.C.:
U.S.
Government Printing OEice,
1943.
Personal Characteristics. Washington, D.C.: U.S. Government Printing Office,
1953.
Parentage. Washington, D.C.: U.S. Government Printing Office,
1954.
282
I
SANFORD
M.
JACOBY
,
U.S.
Census
of
Population:
1950.
Vol.
4,
Special Rpts.,
Part
4,
Industrial Char-
acteristics. Washington D.C.
:
U.
S.
Government Printing Office,
1955.
.
Hbtorical Statistics ofthe
US.
Washington, D.C.:
U.S.
Government Printing
office,
1960.
.
US.
Census
of
Population:
1960.
Subject Reports, Industrial Characteristics,
Final Report
PC(2)-7F.
Washington, D.C.:
U.S.
Government Printing Office, 1967.
Ware, Caroline
F.
The Early
New
England Cotton Manufacture.
Boston,
MA:
Houghton
Main,
1931.
Williams, John
M.
“An
Actual Account
of
What We Have Done to
Reduce
Our
Labor Turnover.”
In
U.S.
Bureau
of
Labor Statistics,
BuUetin
No.
227. Washington, D.C.:
U.S.
Government
Printing
office,
1917, pp. 173-190.
Wolman,
Leo.
Ebb
and
Flow
in Trade Unionism.
New York: National Bureau
of
Economic
Research,
1936.
and Gustave Peck. “Labor Groups in the Social Structure.” In Wesley C. Mitchell,
ed.,
Recent
Social
Trends in
the
United States.
New York: McGraw-Hill,
1933,
pp.
801-
856.
Wood,
Norman
J.
“Industrial Relations Policies
of
American Management,
1900-1933,”
Business
History
Reotew,
XXXIV
(Winter,
1960),
403420.
Woman, H.A. “How to Secure Factory Workers.” In Clarence M. Woolley, ed.,
Employer
and Employee.
Chicago, IL: The System Co.,
1907,
pp,
51-58.
Woytinsky, W.
S.
Three
Aspects
of
Labor Dynamtcs.
Washington,
D.
C.
:
Social
Science Research
Council, 1942.
... Aan het begin van de industrialisatie hadden de meeste werkenden nog een voorkeur voor een bestaan als zelfstandige, men wilde voor zich zelf blijven werken en niet ondergeschikt worden aan een werkgever (Hoogenboom & Knegt, 2017). Uit Amerikaans onderzoek blijkt bijvoorbeeld dat aan het begin van de twintigste eeuw toen het personeelsverloop veel hoger was dan momenteel het geval is, het merendeel van de arbeidsrelaties beëindigd werd door de werkende en niet door de werkgever (Carter, 1988;Jacoby, 1983;Owen, 2004). ...
... Da sie häufig auch einen relativ hohen Anteil an betriebsspezifischem Humankapital aufweisen, wird die berufliche Mobilität 52 gering sein. Zudem suchen sie auch Arbeitsplätze in Unternehmen, die dem Ausbildungsbetrieb gleichen, um den Produktivitätsverlust zu verringern (Elliot /Lindley 2006;Jacoby 1983). ...
... 7 The 1930s saw a surge in the creation of personnel departments, often evolving out of welfare departments (Kaufman, 2008). They deployed personnel counseling alongside a broader set of concepts required to sustain what later came to be called "internal labor markets" (Jacoby, 1983b), such as centralized personal administration, job analysis, and promotion ladders. ...
Article
Full-text available
Over the last century and half, US industry has seen the emergence of several different management models, but we still understand little about the factors that drove their evolution. We propose a theory of this evolution based on three nested and interacting processes. First, we identify several successive waves of technological revolution, each of which prompted a corresponding wave of change in the dominant organizational paradigm. Second, nested within these waves, each of these organizational paradigms emerged through two successive cycles—a primary cycle which generated a new management model that obsoleted the prior organizational paradigm, and a secondary cycle which generated another model that mitigated the dysfunctions of the primary cycle's model. Third, nested within each of these cycles, we identify a problem-solving process in which the development of each model passed through four main phases during which various related management concepts competed for dominance.
Article
This volume brings together leading political scientists to explore the distinctive features of the American political economy. The introductory chapter provides a comparatively informed framework for analyzing the interplay of markets and politics in the United States, focusing on three key factors: uniquely fragmented and decentralized political institutions; an interest group landscape characterized by weak labor organizations and powerful, parochial business groups; and an entrenched legacy of ethno-racial divisions embedded in both government and markets. Subsequent chapters look at the fundamental dynamics that result, including the place of the courts in multi-venue politics, the political economy of labor, sectional conflict within and across cities and regions, the consolidation of financial markets and corporate monopoly and monopsony power, and the ongoing rise of the knowledge economy. Together, the chapters provide a revealing new map of the politics of democratic capitalism in the United States.
Article
Résumé Les emplois à vie sont généralement présentés comme un trait dominant des marchés du travail de l'après-guerre, en recul aujourd'hui. L'auteur cherche à vérifier ces affirmations avec une nouvelle méthode, qui permet d'estimer, à partir de données transversales, la probabilité d'occuper un emploi à vie selon la cohorte de naissance. Il montre que cette situation n'a concerné en réalité qu'une faible proportion des travailleurs britanniques de l'après-guerre, avec des variations selon le groupe sociodémographique. En outre, si les emplois à vie ne disparaissent pas, les emplois de longue durée reculent chez les hommes des cohortes 1947–1966, notamment à cause d'une moindre stabilité des parcours professionnels plus tôt dans la carrière.
Article
Resumen Frente a la idea de que en la era fordista predominó el empleo de por vida, el autor utiliza un método novedoso para estimar con datos transversales del Reino Unido la probabilidad para diferentes cohortes de nacimiento de alcanzar determinados umbrales de antigüedad. Se constata que la proporción de trabajadores del Reino Unido que ocuparon puestos por 30 años o más en la era fordista fue baja, con variaciones entre los grupos sociodemográficos. La prevalencia de los empleos de larga duración (20 años o más) disminuyó entre los hombres nacidos entre 1947 y 1966, en parte por la menor estabilidad laboral durante la juventud.
Article
Lifetime jobs are generally considered a central feature of the postwar labour market and contrasted with insecure and unstable jobs predominant in the post‐Fordist economy. This paper argues that the literature would gain from more precise data on the prevalence of lifetime jobs. Using a new methodological approach, this article finds that the share of UK workers who held a lifetime job in the postwar era was low, with important variations across sociodemographic groups. The importance of long‐term jobs, but not lifetime jobs, has decreased among men born between 1947 and 1966, partly driven by decreasing job stability during youth.
Chapter
Spain became part of the guestworker system with the signing of the Agreement between Spain and West Germany in 1960 to the end of the system in 1973. The question of the integration of immigrants in the labour market is a long-standing focus of sociological and economic research on immigration to West Germany. This chapter revisits the case of Spanish workers from an historical perspective. It considers two areas of integration—job promotion and labour turnover. By combining survey based research on the immigrant stock with qualitative material originating from both the host and home country, it reveals details that are not captured in official national data. The central themes of my case study include the inter-related issues of immigrants’ previous positions in their home country, their industrial roles in West Germany, their high rate of eventual return home and the legal obstacles they encountered as immigrants and non-European Economic Community (EEC) citizens.
Book
Full-text available
In any system of innovation, the constituency of participating individuals are central to determining its strengths and weaknesses. A fundamentally mobile resource, individuals are influenced by a complex nexus of factors prompting their movements. This title is a review of the literature on the topic and is designed as a resource for anyone interested in analysing human mobility and the factors and policies through which it influences innovation. It consists of three sections. The first provides a historical and theoretical background to the study of mobility and its relationship to innovative capacity. The second looks at the empirical measurement of mobility and evidence of its relationship to innovative performance. The third reviews policies used to affect the mobility of highly skilled individuals. Throughout the study, the focus is on the South African experience and its relation to the African and broader international experience.
Article
Full-text available
Until 1877, when Horace Gray Wood’s “A Treatise on the Law of Master and Servant” was published, the rule in matter of termination of the employment relationship in the US was dismissal with notice, pursuant to the British Common Law tradition. On the contrary, Wood “reckoned” that the US rule in relation thereto was Employment-at-Will, which allowed any of the parties to immediately terminate in any case the employment relationship. Notwithstanding the ungrounded nature of Wood’s statement, since then US Courts started to adhere to Employment-at-Will, which became accordingly known as “Wood’s rule”. This constitutes a puzzle for legal theory, for the rule was “invented” but largely accepted by the legal community: it was, on the one hand, a false statement about the legal system but, on the other, a legal truth once accepted. In the present paper we try to make the puzzle explicit and to present a way-out of it, distinguishing a pre-Wood and a post-Wood context. However, such a way-out does not solve by itself the legitimation issue represented by the shift from the first to the second context.
Article
Introduction, 73. — I. Some inferences from empirical work, 74. — II. Postwar movements of wages and employment in the industrial sector, 77; changes in wages and in employment — correlations, 78; wage levels and employment changes, 79; wage levels and wage changes; wage dispersion, 80; wage rates and quit rates; related evidence, 81. — III. Some instances of major change, 85; changes in production-worker wages and employment in the period including World War II, 86; changes in wages and changes in employment — extreme values, 87; changes in compensation and employment in broad industrial categories, 1948–60, 88. — IV. Summary and policy inferences, 95.