AlbertJ.Reiss, Jr.n/an/an/a, and
EvelynM.Kitagawan/an/an/an/a
Demographic Characteristics and Job Mobility
of Migrants in Six Cities1
This paper reports the characteristics of the migrant
population and compares them with those of non-migrants in six cities:
Chicago, Philadelphia, Los Angeles, San Francisco, St. Paul, and New Haven.
The data on migrants and non-migrants are based on a sample survey of some
4,000 to 5,000 persons 14 years and older in each of the six cities in
January 1951. For the most part these persons were located in about
1,900 households in each city which
were enumerated in the 1950 Census of Population and Housing. The estimates
of total males and females 14 years and older are based on a combination of
data from the survey and the 1950 Census. All other estimates were obtained
by inflating weighted sample results to the estimates of total males and
females 14 years and older.
The six cities for which findings are reported differ
considerably in their occupational and industrial structure.
Characteristically, Chicago and Philadelphia are referred to as mature
metropolitan centers with a relatively stable population and as industrial
centers with a diversified economic base where employment in
manufacturing is predominant (but less than 50 per cent of all employed),
while employment in wholesale and retail trade, though second in
importance, is substantially below that in manufacturing. New Haven has a
similar economic base, though it is much smaller in size. By comparison,
San Francisco and Los Angeles are relatively young metropolitan cities with
a diversified economic base where employment in wholesale and retail trade
is more or less equal to employment in manufacturing and together these
industries employ about one-half of all workers. St. Paul, though an older
and smaller city, has a similar economic base. These differences in major
economic activity are reflected in the occupational structure of the
cities. Chicago, Philadelphia, and New Haven, where employment in
manufacturing predominates, show a greater concentration of blue collar
workers, while Los Angeles, San Francisco and St. Paul, where employment
in retail trade is about equal to that in manufacturing, show a greater
concentration of white collar workers. The same patterns in occupational
structure hold for both men and women. However, superimposed on this
pattern is the expected pattern of substantially higher proportions of
white collar workers among women than among men. These structural
differences among the cities help to explain certain of the migrant
differentials which follow.
The rapid growth of urban centers in the United States during
the nineteenth and early twentieth centuries was made possible by
immigration, a high birth rate, and internal migration. Since the changes
in immigration policy during the 1920’s, cities have increasingly relied
upon internal migration and natural increase for their growth. Students of
urban life have therefore turned their attention to a study of the
selectivity of migration and its contribution to urban
growth.
While metropolitan centers usually require some migrants in
order to maintain their size, migration is greatest for rapidly growing
cities where the industrial base is undergoing considerable expansion.
Opportunities for such expansion occur not only under conditions of general
economic growth but under specialized conditions
such as wartime or emergency mobilization of the economy. The size and
composition of the migrant population of cities, therefore, may be
expected to vary, depending upon the size and kind of economic base, the
relative age and rate of growth of the cities, the age and sex structure of
their populations, and the relative mobility prevailing in their
occupational and industrial structures.…
STABILITY OF POPULATION
We observe considerable variation in the residential stability of
persons in the six cities. The proportion of migrants—i.e., persons who
have resided in the Standard Metropolitan Area of a city less than 12
years—in the total population 14 and older varies from a high of 46 per
cent in Los Angeles to a low of 13 per cent in Philadelphia.2
There is, moreover, a marked stability of residence for persons in the
older industrial cities, Chicago, Philadelphia, and New Haven. Over
two-thirds of the persons 14 and older in these cities have resided in the
Standard Metropolitan Area of the city 21 years or more. St. Paul also
shows rather marked stability of residence as 59 per cent of the persons 14
and older are long-time area residents. The rapidly growing metropolitan
cities of the West Coast show the least residential stability. This is more
marked in Los Angeles where only 33 per cent of all persons 14 and older
have resided in the S.M.A. 21 years or more than in San Francisco where 45
per cent are long-time residents. There were no significant differences in
duration of residence for men and women in each city, except in San
Francisco and St. Paul where there seem to be a significantly higher per
cent of women than men who are migrants.
SELECTED CHARACTERISTICS OF MIGRANTS
Studies of migrations to metropolitan cities have generally emphasized
that migrants possess those characteristics generally assumed to be
associated with mobility. In particular they find that (1) a larger
proportion of the migrants than non-migrants are in the younger age groups;
(2) a larger proportion of migrants than non-migrants are in the labor
force; (3) migrants more often than non-migrants tend to belong to families
or households whose size indicates a relative freedom from family controls
or responsibilities. Furthermore, it is usually found that cityward
migration is selective of women, i.e., a lower sex ratio for migrants than
non-migrants, and that the sex ratio of cityward migrants is lowest in the
young adult age groups. These findings generally hold for migrants in each
of the six cities.
The participation of migrants in the work force of a city is both an
indication of the factor of economic opportunity in impelling migration and
the adjustment of migrants to the requirements of a labor market as well as
an index of their availability for recruitment into a work force. It
should be remembered that migrant persons compete against non-migrants who
have various claims to employment preference (e.g., seniority, experience
in a local labor market, status requirements, etc.). Therefore, we would
not necessarily expect high participation of migrants in the labor force
relative to comparable non-migrant groups.
Examination of work status differentials of migrants and non-migrants in
the six cities shows that among both men and women, migrants were more
often at work than were non-migrants.… However, among men the higher
work participation rates for all migrants than all non-migrants were almost
entirely due to the more "favorable" age distribution of migrants insofar
as providing workers was concerned. The major reason, therefore, for
the relatively higher participation of migrant men in the work force as
compared with that of non-migrant men is that male migrants were more
heavily concentrated in age groups where work participation rates were
high. The fact that differences in the age distribution of migrants and
non-migrants explain these differences can be seen in a comparison of
age-specific work-rates for male migrants and non-migrants.… The
age-specific work-rates for male migrants are significantly higher in only
one age group. In general, in all six cities, male migrants 14–24 years old
were more often at work than male non-migrants of this age. This suggests
that young male migrants as compared with non-migrants include a relatively
larger proportion of persons who seek jobs rather than attend school. Of
course, male migrants 14–24 usually include a lower proportion of 14–17
year olds who would be attending high school than do male non-migrants
14–24 years old. It is also possible that, in cities, a larger proportion
of non-migrant men than migrant men 18–24 years old lead a "marginal
existence" for which employment in the work force is not a requisite.
However, the higher work rates for migrant women were not entirely due
to their "favorable" age distribution. Examination of the age-specific
work-rates of women supports the conclusion that migrant women in each
age group were more often at work than non-migrant women of the same age.
This probably explains why in each city the work participation rate for
all migrant women 14 and older exceeds that of all non-migrant women 14 and
older by more than the work participation rate of all migrant men 14 and
older exceeds that of all non-migrant men 14 and older. The marital status
and family responsibilities of non-migrant women as compared with those
of migrant women probably account for this somewhat higher participation in
the work force of migrant women.
While migrants apparently are attracted to cities by economic
opportunities, these opportunities seem to be disproportionately those
offered by private employment rather than those afforded by
self-employment or employment in government. In all six cities we found
that private employment has the largest proportion of migrants and "own
business" in general the lowest. This is probably so for a number of
reasons. Migrants are somewhat younger in age and often may not possess the
requisite capital to establish a private business or engage in private
practice (self-employment); also, they less frequently may be aware of such
opportunities as compared with local residents. Furthermore, male
migrants are on the average less skilled than are male non-migrants. The
findings were less clear for women.
During the 1940–49 decade there was considerable variation in job
mobility of workers in the six cities and of workers in various age and
migration status subgroups.3 The crude mobility rates for
workers in Los Angeles and San Francisco—where workers averaged about
three jobs per person during the 1940–49 decade—were higher than the crude
rates in the other four cities, where workers averaged about 2.5 jobs per
person.…
Also the average number of jobs held between 1940 and 1949 was
considerably higher for migrants than non-migrants and considerably higher
for young persons than for older persons. These age and migrant status
differentials in mobility remain clearly defined even when age-migrant
status-specific mobility rates are computed. That is, rates for migrants
are higher than those for non-migrants even when age is held constant, and
rates for young persons are higher than those for older persons even when
migrant status is held constant, except for male migrants in Los Angeles,
St. Paul, and New Haven where mobility varies very little by age.
Thus, it is clear that age and migrant status are independently
important factors influencing the average number of jobs held during the
1940–49 decade by workers in the six cities. Since this is the case and
since the proportion of migrants in the work history samples of the six
cities varied considerably—for example, from highs of 47 and 35 per cent
for men in Los Angeles and San Francisco respectively to lows of 15 and 13
per cent for men in Chicago and Philadelphia respectively—at least part of
the differences in the crude mobility rates among the six cities may well
be accounted for by differences in the migrant composition of the work
history samples in the six cities. Since there are also some variations in
the age composition of migrants and non-migrants in the cities, it is
desirable to take the influence of age composition into account at the same
time.
An "expected cases" analysis of the extent to which the largest city
differences in
TABLE 1 PER CENT COMPONENTS (BY AGE AND MIGRANT STATUS) OF CITY DIFFERENCES IN
MOBILITY, BY SEX: CHICAGO AND LOS ANGELES, CHICAGO AND SAN FRANCISCO, PHILADELPHIA AND LOS ANGELES,
PHILADELPHIA AND SAN FRANCISCO
crude mobility rates may be accounted for by differences in their age
and migrant composition has been made for selected pairs of cities—namely,
Chicago and Los Angeles, Chicago and San Francisco, Philadelphia and Los
Angeles, and Philadelphia and San Francisco. The chief reason for
selecting these four pairs of cities for the present analysis was that each
pair includes two extremes with respect to crude mobility rates and also
with respect to the importance of migrants in the population. That is,
Chicago and Philadelphia are cities with relatively low mobility and fewer
migrants, while Los Angeles and San Francisco are cities with relatively
high mobility and much higher proportions of migrants.
The expected cases analysis shows that about half of the differences
in 1940–49 crude mobility rates between cities with the lowest crude rates
and cities with the highest crude rates can be accounted for by differences
in their age and migrant composition combined (see Table 1). Among men,
the per cent of the difference in crude mobility rates accounted for by age
and migrant status combined varies between 41 and 65 per cent for three of
the four pairs of cities included in this report—that is, 65 per cent of
the difference between Chicago and Los Angeles crude mobility rates, 41 per
cent of the difference between Chicago and San Francisco crude rates, and
44 per cent of the difference between Philadelphia and Los Angeles crude
rates can be accounted for by differences in their age and migrant
composition.
Among women, the proportion of the difference in crude mobility rates
accounted for by age and migrant status combined varies from 40 per cent of
the difference between Philadelphia and Los Angeles crude rates to 51 per
cent of the difference between Chicago and San Francisco crude rates.
It should also be noted that it is the higher proportion of migrants in
the high mobility cities which is the more important factor in accounting
for differences in crude mobility rates of high and low mobility cities.
Migrant status alone accounts for a substantial part of the differences in
crude mobility rates—see column (3) of Table 1. However, differences in
age composition alone (without holding migrant status constant), instead of
being partly responsible for differences in crude rates, tend to act in the
opposite direction. In six out of eight pairs of cities, the difference in
mobility is slightly increased when age composition is held constant—this
is the interpretation of the negative components in column (4) of Table 1.
These findings, therefore, suggest that about half of the differences
between crude mobility rates of the high and low mobility cities included
in the Occupational Mobility Survey can be accounted for by their age and
migrant composition. That is to say, if the high and low mobility cities
had the same proportions of migrants and the same age composition, then
their differences in over-all mobility rates would be cut about in half. Of
course, this still leaves the West Coast cities, Los Angeles and San
Francisco, with higher mobility rates than Chicago and Philadelphia, though
to be sure the size of the difference is considerably reduced under this
assumption.
It is possible that differences in the occupational and industrial
composition of these cities might account for part of these remaining
differences. However, it is not unlikely that other factors are more
important in this respect. For example, it may well be that the much
greater relative increase in population and expansion of employment in the
two West Coast cities during the 1940–49 decade as a result of the war
industries located there (particularly shipbuilding industries in San
Francisco and aircraft industries in Los Angeles) generally facilitated job
changes and also necessitated relatively more post-war conversion in these
two cities than in
Chicago and Philadelphia. In this case, we would expect higher job
mobility rates in these cities (since mobility for this study was defined
as mean number of jobs held 1940–49) even after the influence of other
population and labor force characteristics is taken into account.
1 From ,
1953, 32:70–75. By permission.
2 In this study, persons who had resided in the Standard Metropolitan
Area (S.M.A.) of a city less than 12 years, i.e., persons who arrived after
the middle of 1939, were classified as migrants. All persons who had
resided in the S.M.A. 21 years or more are called long-time residents.
Although length of residence in the S.M.A. of a city was used to define
migrants status, survey coverage was limited to the city boundaries so that
migrants refer to 1951 city residents who moved into the S.M.A. less than
12 years previously, while non-migrants refer to 1951 city residents who
moved into the S.M.A. (though not necessarily the city itself) 12 or more
years previously.
3 Job mobility for this paper is defined as the mean number of jobs held
by persons in a given subgroup during the 1940–49 decade. A job is defined
as a continuous period of work for one employer.…