CLASS, HOUSEHOLD STRUCTURE, AND MIGRATION:
A CASE STUDY FROM RURAL MEXICO
Marila de los Angeles Crummett
Kellogg Institute for International Studies
University of Notre Dame
Notre Dame, Indiana 46556
Working Paper #92
MICHIGAN STATE UNIVERSITY
Copyright 1985, MSU Board of Trustees
CLASS, HOUSEHOLD STRUCTURE, AND MIGRATION:
A CASE STUDY FROM RURAL MEXICO
Maria de los Angeles Crummett
Kellogg Institute for International Studies
University of Notre Dame
Notre Dame, Indiana 46556
Working Paper #92
Abstract: Studies of migration in Mexico have proliferated over the years
capturing detailed socioeconomic aspects of migrants such as sex, age,
education, occupational status, income, place of origin, and destination.
These descriptive studies of migration have fallen short, however, in that
they fail to explain why men are the migrants in some regions and women in
others, or why some migrants choose destinations within Mexico and others head
toward the United States. This paper argues that an examination of class and
household characteristics offers significant insights into these questions.
The household's class position explains not only what groups have the greatest
propensity to migrate but also where they tend to migrate. An analysis of
household structure, including the sex and age division of labor, sheds lights
on who within the household is most likely to migrate.
About the Author: Maria de los Angeles Crummett received her Ph.D. in
Economics from the New School for Social Research and is currently a Faculty
Fellow at the Kellogg Institute for International Studies, University of Notre
Dame. Her field research in Peru, Chile, and Mexico has resulted in
publications on agrarian economic development, peasant differentiation and
migration, the role of households in migration, and the political mobilization
CLASS, HOUSEHOLD STRUCTURE AND MIGRATION:
A CASE STUDY FROM RURAL MEXICO
Studies of migration in Mexico have proliferated over the years capturing
important socioeconomic aspects of migrants such as sex, age, education,
occupational status, income, place of origin, and destination.1 These
studies have fallen short in some respects, however. For example, they fail
to explain why men are the migrants in some regions and women in others, or
why some migrants choose destinations within Mexico and others head toward the
United States. The aim of this study is to address these questions by
examining the characteristics of individual migrants within a more
comprehensive analysis, specifically from the perspective of class position
and household structure.
This study is based on a sample survey of 59 rural households conducted by
the author in 1982 in Calvillo, Mexico,2 and argues that examining class and
household characteristics of rural units greatly enhances our understanding of
migration. The household's class position within the agrarian sector in
particular explains not only what groups have the greatest propensity to
migrate but also where they tend to migrate, that is, whether within Mexico or
to the United States. Moreover, an analysis of household structure and
composition, including the gender and age division of labor, offers valuable
insights into who within the household is most likely to be selected for
Calvillo, located in the southwest corner of the state of Aguascalientes,
offers a rich case study for an examination of class stratification, household
structure, and migration in the rural sector. Roughly 75% of the region's
inhabitants live in rural communities and almost 60% of the labor force earn a
living from agricultural activities (Censo General de Poblacion 1980). In
addition, the region contains a high percentage of landless agricultural wage
workers, and it is the only major agricultural region in the state in which
the peque a propiedad smallholdingg) private unit of production dominates the
landholding structure. Private owners control 73.4% of the total arable land
while ejidos3 account for the rest (Censo Agricola, Ganadero y Ejidal,
Calvillo's agrarian sector is characterized by a highly polarized
production structure. Production of the region's commercial crop, guava,
completely overshadows the production of staple crops, corn and beans in terms
of area cultivated, output, and yield. The production of commercial and
staple crops is also divided along land tenure lines. While both large and
small private owners (landholding units with greater than five hectares [one
hectare equals 2.47 acres] and five or fewer hectares, respectively),
concentrate their resources in the production of high value fruit crops,
ejidos devote most of their land to basic crops for home consumption.
An examination of expenditures on wage labor and means of production by
landholding unit highlights the economically and socially differentiated
character of Calvillo's agrarian sector. In -1970, wage payments represented
over half of total farm costs for private units of production. Labor costs
for ejidos, on the other hand, amounted to about 7% of total production
expenses. Large private holdings are also the most capitalized units in the
region accounting for 70% of total expenditures on raw materials and 76% of
investments in capital equipment (Censo Agricola, Ganadero y Ejidal,
In addition to the uneven character of agriculture in the region, Calvillo
is notable for high rates of out-migration. Temporary cyclical migration to
the United States has been particularly acute since the mid-1960s and
constitutes the most outstanding migratory trend in the region and the state.
The paper is comprised ofj three major sections. The first section
presents a brief historical overview of the process of capitalist development
in agriculture and its effects on Mexico's rural population. Two key
processes are highlighted for the region and for the nation as a whole:
increasing socioeconomic differentiation within the rural sector; and the
emergence of widespread internal and international migration.
The second section offers both a theoretical and an empirical treatment of
class structure in the rural sector. The analytical framework goes beyond
traditional classifications of the peasantry by attempting to account for the
structure of the production process as well as the social relations that
emerge from it. Thus, the rural household's ownership of the means of
production is considered within the context of the way in which households
participate in the relations of production. !This theoretical foundation is
then employed to develop empirical indicators of social classes. Three
particularly salient characteristics of the region's agrarian structure--the
buying and selling of household labor, crop and land type, and access to means
of production other than land--are used to determine divisions among rural
The third part of the paper uses the analysis of class structure developed
in section two to examine major patterns and rates of migration by class. It
explores the way in which class characteristics of rural units, in conjunction
with the gender and age division of labor and household size and composition,
shape and reinforce specific migration patterns. In addition, the impact of
migration on the household division of labor is considered.
Agrarian Structure and Migration: A Brief History
Calvillo's agrarian development differed fundamentally from that of the
rest of Aguascalientes. The hacienda, an integral part of pre-revolutionary
history in the rest of the state, did not play a major role in Calvillo. (In
1910 Calvillo contained three haciendas that occupied less than 15% of the
region's land.) Rather, the pequera propiedad smallholdingg) dominated the
rural sector since the founding of the community in the early 1700s.
Throughout most of the 19th century, these private units of production grew
and prospered. By 1925, however, primarily through property sales and
inheritance, the majority of parcels had -been reduced to minifundios,
landholding units less than a hectare in size (Rojas Nieto 1981).
After the Mexican Revolution, the agrarian reform land distribution
program of the 1930s created ejido units from the expropriation of one of the
region's three haciendas. The remaining haciendas were either subdivided
among estate owners' relatives or sold as ranchos (large units of mixed
farming/cattle) to individual buyers to avoid expropriation. Consequently,
reform beneficiaries accounted for only a fraction of the landless peasantry.
From 1930 to 1945, Calvillo's present agrarian structure began to take shape.
The rural population consisted of owners of a few extensive ranchos, numerous
penque'os propietarios, a handful of ejidatarios, and a relatively large
contingent of landless agricultural wage workers.4
The smallholding class of the peasantry cultivated corn, beans, chili
peppers, and some fruit trees, principally guava and peach, for personal
consumption and sale at the local market. Cattle raising was a viable
activity on the larger ranchos. Households without land rented or
sharecropped small parcels; the majority, however, survived through a
combination of artisan production, petty trading, and wage work, primarily in
With the exception of occasional, temporary, seasonal migration to
undertake wage work in neighboring areas, out-migration from Calvillo was
minimal prior to 1940.5 From 1942 to the mid-1960s, temporary migration to
the United States dominated the region's as well as the nation's migration
patterns. The Bracero Program, a contract labor agreement set up between the
United States and Mexico at the start of the Second World War, drew thousands
of workers from the community to short-term work in agriculture across the
The demand aspects of the program often overshadow the complex economic
and political changes occurring in the region and in Mexico as a whole,
fostering increased levels of documented and undocumented migration.6 The
height of the nation's modernization drive in agriculture coincided with the
program. Agriculture was to provide Mexico with the foreign exchange and
cheap food needed to spur the industrialization process (Arizpe 1981). Land
redistribution was at its peak under the Cardenas administration (1934-1940)
and came to a near halt after 1940.7 Between 1950 and 1960 agricultural
policy shifted away from land distribution and development of rainfed
smallholding plots to public investment in large scale irrigated agriculture
and livestock raising (Dinerman 1982). The emphasis on irrigated agriculture
discouraged the production of basic foodstuffs in favor of more profitable
By 1960 the polarization of Mexican agriculture into large scale
agribusiness and small rainfed units was complete. Fifty percent of the
landholding units had less than five hectares of land. These smallholders
controlled about 14% of the total arable land and produced 4% of agricultural
output. On the other hand, 0.5% of the land units accounted for 28.3% of the
arable land and 32.3% of agricultural production (Arizpe 1981:167).
One of the manifestations of this new model of development was an increase
in the number of landless agricultural wage workers. Between 1950 and 1960
rural laborers increased by 50%, making up almost half of the agricultural
work force. By 1970 they represented 54% of the labor force in agriculture
(Pare 1979). In short, the "modernization" process had seriously eroded the
economic viability of smallholding rainfed i agriculture and produced an
increase in the number of landless agricultural wage workers. A major
consequence of this process was an outflow of rural people to the cities and
across the northern border in search of work.
The development of commercial agriculture in the region had similar
effects. In the 1950s and 1960s, a group of capitalist farmers in Calvillo
was able to consolidate landholdings and undertake commercial guava
production.8 Private and public bank credit financed the construction of
dams and other irrigation systems; improved strains of guava, insecticides,
herbicides, chemical fertilizers and other high technology inputs became
essential ingredients in the production process.
Calvillo's entrepreneurs came mainly from the small proprietor class. By
the early 1970s, they had brought guava production into full swing. Calvillo
became the most ,important producer of guava in Mexico, accounting for
two-thirds of national output and 30% of the state's agricultural income (INIA
1980). In contrast, basic crop cultivation dropped sharply for two major
reasons: 1) government regulated corn prices eroded producer incentives to
the point where land was sold, leased or abandoned altogether;9 and 2) many
producers, attracted by the lucrative nature of guava, began planting guava
trees in place of basic food crops.
As the number of hectares planted to guava increased throughout the 1970s,
strategic portions of the production process became concentrated in the hands
of a few producers. Packing and storage, processing, distribution to national
and international markets, and sales were controlled by the minority of
growers who had access to financial and commercial networks. More and more,
capital accumulation in guava controlled the productive activities of the
majority of direct producers. Increased !capital accumulation further
differentiated producers, thus reinforcing the capitalist character of small
private producers (Esteva 1980).
The process of capitalist development not only increased socioeconomic
differentiation in Calvillo but also created a relative surplus population
comprised of marginalized subsistence producers and a growing sector of
landless laborers.10 For this population, migratory wage labor became the
primary means of survival.
Although migration is a Widespread phenomenon in the region, households
have not been affected equally. The next section turns from a discussion of
the macroeconomic causes and consequences of the agrarian crisis to an
analysis of rural class structure. We contend that an examination of the
development of the rural productive structure las presented here provides only
a partial explanation of differential migration rates and patterns. A fuller
understanding of migration involves relating the overall process of
macroeconomic change to the household's specific role in the productive
structure, i.e., class position.
Determinants of Rural Class Structure in Calvillo
Traditional economic studies of the peasantry tend to select a single
index of ownership over the means of production as the major determinant of
class; the most common indicator is size of landholding.l1 In a number of
studies, land quality or the value of agricultural output complements the
analysis.12 The consideration of. property relations in isolation does not,
however, adequately capture class divisions in the agrarian sector. First,
simple land-size categories do not reflect differences in the quality of
land. Second, and more important, land-size categories do not capture the
organization of production on the landholding.
To better understand class stratification within the peasantry, the nature
of the productive process as well as the social and economic relations that
emerge from such a process must be considered. Unlike the standard criterion,
this approach to class status enables us to view households' unequal access to
land and other means of production as part of broader structural mechanisms,
the process 3f production, accumulation, and exploitation, in operation in the
At the empirical level, the task is to identify indicators of the
household's access to the means of production and the associated set of
relations in which it participates. The incidence of commercialized
agriculture and the extent of proletarianization in Calvillo give rise to
three major interrelated indicators that serve as approximate measures of
The first and most important measure of class status is the household's
participation in the labor market. By examining the buying and selling of
household labor, this first indicator attempts to account for the household's
level and form of integration into rural relations of production. We then
introduce two variables that reflect important facets of the household's
access to productive resources and, thus, of its form of participation in the
production process; these are the type of land in use and the kind of crop
produced. The final class indicator is the household's ownership of means of
production other than land, i.e., tools, work animals, and other farm
animals. This indicator, an additional measure of the household's economic
standing, complements and reinforces the class structure given by the first
Using the data from the Calvillo survey, the following subsections analyze
each of the three indicators to arrive at a multivariate picture of class in
the study region.
1. Participation in the Labor Market. In the literature concerned with
class status within the peasantry on the empirical level, the methodology
developed by the Indian economist, Utsa .Patnaik (1976), represents a
groundbreaking contribution to the field. In her analysis of Indian
agriculture, she argues that using the size of the landholding unit to measure
the concentration of means of production is not a sufficient foundation upon
which to indicate class status,. While examining the resource position of the
household may indicate that the peasantry is' segregated into more or less
distinct land-size divisions, it fails to capture qualitative differences in
land type and in the ways in which production is organized.14 Stressing
that no one index can fully measure class status, Patnaik proposes that the
labor exploitation criterion or "the use of outside labor relative to the use
of family labor would be the most reliable single index for categorizing the
peasantry" (1976:84, Patnaik's emphasiss.
Patnaik's index defines class position in the rural sector through two
related criteria:; the possession of the means of production and the
exploitation of labor arising directly from the production process itself.
The uneven distribution of the means of production in the rural sector
reflects a process in which certain households accumulate most productive
resources and thus require more labor than can be provided by family members,
while other households have so few resources as to necessitate selling their
labor power. Additionally, the total amount of labor used and, thus, its
division between family and hired labor reflects the intensity of cultivation
as well as the level of technology. At a general level, then, households can
be classified by the extent of their participation in the labor market.
Patnaik presents the following "E" index or labor exploitation
criterionl5 to categorize "mutually exclusive economic classes":
E = (XI XO)/Y
where Xl equals total labor days hired ini by the household; X0 equals
total labor days hired out by the household- and Y represents family
(household) labor days on the operational holding.16
The numerator of the E index determines whether a household is a net
seller or net buyer of labor power. The relationship between net labor
(X1-XO) and family labor (Y)i thus indicates the household's relative
dependence on wage labor for subsistence. For example, a fully
proletarianized household--lacking land and other means of production--neither
hires in labor (X1=O) nor performs family labor (Y=O). In this case the E
ratio tends toward negative infinity, since the household participates in the
labor market only as a seller of labor power (XO>O). At the other extreme,
a pure capitalist household depends exclusively on the labor of others
(X1)0, XO=O, Y=O) for production; E therefore approaches positive infinity.
For those classes not identified as exclusively capitalist or proletarian,
the sign and size of E determine whether a peasant household is a net
appropriator of labor or whether it is exploited on the whole. In the middle
peasantry, self-employment (Y) by definition is of primary importance for
household subsistence and therefore exceeds net labor. The numerator will be
positive and E positive if the household on balance hires in more labor than
it hires out. If no outside labor is involved on the operational holding
(Xl=0) or if the household is more dependent on wage labor for subsistence
(Xo)X1), then E is 0 or negative (and small). Poor peasants or the
semi-proletarians are poorly equipped with land and other productive
resources; off-farm labor thus provides the greater part of households'
subsistence so that E tends to be large and negative.
In applying the E criterion to the Calvillo survey data, several salient
trends emerge (see Table 1). First, the class divisions given by the E ratio
show that only a small percentage of the landed, i.e., the capitalist/rich
peasantry17 control the demand for labor in each region. Second, the data
show evidence of a self-sustaining middle peasantry that on balance neither
exploits labor nor is itself exploited (X1=0, XO=0, and E=0). And
finally, the greatest percentage of households are located within the lower
two strata of the rural class structure, the poor and full-time laboring
groups. Significantly, proletarianization is pervasive among poor peasants.
Indeed, the average number of labor days hired out by the poor peasantry is
greater than that of the landless or full-time laborer class.
2. Land Type and Crop Type. an examination of the type of land in use
and the kind of crop produced by the household strengthens the analysis by
focusing on the highly differentiated character of agricultural production in
the region. In Calvillo, basic crops, corn and beans, are largely cultivated
on rainfed land while the commercial cropping of guava is done on irrigated
land. Table 2, based on 1970 Census data,10 provides an insight into the
uneven distribution of land by type and crop in the region.19
The most outstanding division within rural groups appears with respect to
crop type. While ejidos devote a minimum amount of land (2.5%) to commercial
crops, private units of production devote between 32% (for holdings greater
than 5 hectares) and 53% (for holdings of 5 or fewer hectares) of their land
to commercial crops. Among rural groups, the distribution of land by crop
type is equally skewed. Private units control 97% of the land under
commercial agriculture and 63% of basic crop land whereas ejidos cultivate 3%
and 37% of commercial and basic crop land, respectively. In terms of land
type, the census data show that both irrigated and rainfed land are
concentrated in large private units of production.
The marked differences across peasant households in the type of land owned
and the kind of crop cultivated translate into severe inequalities in
households' expenditure and revenue structures and, consequently, in class
position. From the point of view of household expenditures, the production of
guava--a commercial crop cultivated on irrigated land--requires large
investments in fertilizers, pesticides, wage labor, transportation, and
year-round upkeep and packaging; thus, it is a viable enterprise only for
those households with a sufficiently strong economic base, namely the upper
strata of the peasantry. Corn and bean production, on the other hand,
requires neither irrigated land nor large capital outlays in the form of wage
labor and means of production and is, therefore, an option open to all landed
From the perspective of household revenues, land and crop type are also
indicative of the household's socioeconomic position in the region's agrarian
structure. In comparing (gross) revenues derived from subsistence and
commercial output, we find that on average each hectare devoted to the
cultivation of guava generated $467,647 as opposed to $1,534 pesos20 for
corn and beans. Similarly, a Ihectare of irrigated land generated an average
of $372,511 in contrast to 1,411 pesos for rainfed land. Clearly, then, these
two variables, crop type and land type, significantly define and distinguish
rural households in Calvillo.
Those producers cultivating guava and having irrigated land thus provide
an initial juxtaposition between the upper; (capitalist/rich and middle
peasants) and lower strata (poor peasants and full-time laborers). Given the
large capital outlays necessary for the cultivation of commercial crops, the
capitalist/rich sector of the peasantry can be identified by the possession of
irrigated land and the production of guava. Middle peasants, although
possessing irrigated land, are more likely to engage in the cultivation of
basic crops, an activity that does not require large capital investments. The
poor peasantry lack irrigated land, thereby impending the production of more
profitable crops. The class structure shown in Table 3 is based on the
analysis of the type of land in use and the kind of crop produced by the
In contrast to the census data given in.Table 2, the sample survey data
capture more precise patterns ofi land distribution by type and crop across
classes. According to Table i3, the capitalist/rich stratum owns a smaller
percentage of the total arable land in the sample than either middle or poor
peasants, yet it dominates the production of the most profitable crop in the
region, guava, and accounts for a significant: portion of the corn and bean
output. For the area under study, then, absolute land size is secondary to an
analysis of land quality and the kind of crop produced on the operational
Furthermore, these two indicators provide a clearer picture of the
polarization of classes than can Jbe gleaned from a consideration of household
participation in the labor market. In particular, the middle peasantry, a
group that lacks a precise classification on: the basis of the buying and
selling of labor power, acquires a more specific character. In contrast to
the capitalist/rich class, the participation of middle peasants in commercial
agriculture is minor; however, by holding irrigated land they are able to
produce both absolutely and relatively larger amounts of staple crops than
poor peasants with larger holdings.
3. Access to Nonland Meansl of Production. The third class indicator
examines access to productive resources other than land. The two previous
indicators supported the existence of distinct classes. A large group of
fully proletarianized households stands out in contrast to a small
capitalist/rich peasant sector Icontrolling commercial agriculture in the
region. The polarization of economic groups Idenotes a level of capitalist
development in which the means of production are concentrated in the hands of
a few, while the vast majority of rural inhabitants must resort to the sale of
labor power to subsist. It is expected that an analysis of the household's
access to nonland means of production will further clarify the pattern of
concentration of resources in the hands of an economically superior class.
The marked difference in the types of production prevalent in Calvillo
makes it necessary to distinguish between mechanized (tractors, trucks, and
pick-ups) and nonmechanized (tools and draft animals) instruments of
production. As indicated earlier, commercial crops require a relatively high
degree of mechanization in contrast to basic crop cultivation which can be
undertaken with draft animals (horses, mules, and donkeys) and a few basic
implements (shovels, hoes, and axes).
The ownership versus rental of means of production, in particular
mechanized instruments, must also be considered in examining rural class
structure. For example, the ownership of a truck or a tractor represents not
only an instrument of production on the operational holding but also an
instrument that can be rented-out to generate a monetary income. Ownership of
mechanized instruments thus implies relative independence in production and an
additional income source that complements agricultural production. On the
contrary, renting-in of means of production may imply a greater dependence on
usury capital and a monetary outflow from the household.
In classifying Calvillo's households with this third indicator, it is
hypothesized that ownership of mechanized means of production is associated
with a capitalist/rich peasant class, whereas renting-in mechanized
instruments characterizes poor and middle peasants. The use of draft animals
in production establishes the opposite situation; poor and middle peasant
households favor draft animals because of their scant hold on mechanized
Ownership of other nonwork farm animals (cows, pigs, and chickens)
provides a further insight into the socioeconomic status of the household.
The animal stock represents both an additional source of farm income and a
means to meet basic subsistence requirements. For poor peasants in
particular, animals are a "reserve fund" to be drawn from in times of economic
difficulty and emergencies. On the other hand, for capitalist/rich households
financially able to amass large animal stocks, they represent an expansion of
farm investments. These differing uses of farm animals suggest that the
possession of animals also represents a useful complement of class status.
Table 4 presents the class distribution of households based on access to
mechanized and nonmechanized means of production. The most important
phenomenon seen in Table 4 is the highly uneven distribution of the means of
production in Calvillo. Capitalist/rich households (7% of the total number of
rural units) control the totality of owned mechanized instruments, over a
third of the tools used in production, and the vast majority of nondraft
animals. Significantly, the middle peasantry (22.8% of households) rents in
approximately two-thirds of all (rented) mechanized instruments. In contrast,
poor households (24.6% of the total) depend heavily upon animal power to
undertake production, own 15% of nonwork animals, and, to the extent that they
use mechanized instruments of production, rent them in. As expected, the
bottom 45.6% of households, the full-time laborers, have little or no access
to means of production.
This skewed distribution of production instruments by class is directly in
line with the class structure that emerged from the consideration of the
buying and selling of labor power and land/crop type. In fact, the
distribution of households is not altered in moving from the second to the
third class indicator. This strongly indicates a cohesive interaction among
the household's extent of participation in the labor market, the type of
land/crop under cultivation, and access to nonland productive resources.z9
Class, Household Structure, and Migration
Using the Calvillo survey data, this section examines how differential
migration rates by class affect and are affected by households' internal
structure, composition, and the sex and age division of labor. This
framework, which introduces particular aspects of the household unit into the
study of migration, is adopted on the assumption that household organization
and structure are dynamic components of class. As a social unit, the
household responds to and acts upon changes occurring in the wider economy.
In this respect, we are in agreement with Pessar in defining the household as
"an evolving nexus of social relations which originates within a larger field
of social relations and institutions through which it is transformed and which
it may in turn modify" (1982:3).
The first part of this section outlines migration patterns and rates by
class and addresses the importance of migration for each class. The second
part links differential migration rates to household structure and composition
characteristics. In particular, women's roles in production and in the
household are highlighted as key components shaping and defining the migration
trajectory of the household. Additionally, the impact of migration on the
household division of labor is considered.
1. Migration Patterns by Class. Table 5 presents data on rural
out-migration by class in Calvillo for the survey period 1981-1982. The table
shows that migration is inversely related to class. The number of households
engaged in migration increases as class status decreases. Almost one-half
(46.1%) of full-time laborer households had members who migrated during the
study period whereas no households within the capitalist/rich sector reported
anyone migrating. The size of household income and the importance of
remittances for household reproduction provide key measures of the role class
plays in migration. For example, among completely proletarianized
households--numerically the most importanti group in Calvillo's rural
sector--migration to the United! States represents an important means to
supplement household income and allows many households to secure reproduction
requirements substantially above a bare subsistence level.
Table 6 shows that over 90% of migrants from the proletarianized class
sent remittances and that remittances constituted a significant percentage of
total household monetary income, 28.5%. In almost half (48%) of full-time
laborer households, the cash influx was directed toward the purchase of
consumer goods such as television sets, radios, bicycles, and household
appliances. Health and educational expenses were also taken care of through
remittances. A smaller percentage (28%) were able to invest migration income
in "home improvements" such as, painting the house, constructing a new room or
house, or installing indoor plumbing. For another group of households (20%),
however, remittances went solely toward the repayment of debts, some incurred
to finance previous migration expenses.
Among the poor peasantry, on the other hand, the relatively high
percentage of households engaged in migratory wage labor (42.8%) reflects this
sector's extreme state of impoverishment. For poor households, the
marginalization of rainfed production of staple crops has meant that wage
labor rather than production on the land satisfies immediate consumption
needs. These households thus have more in common with full-time laborers than
with their landed counterparts. An examination of the data, however, shows
this class to be economically worse off than landless or completely
proletarianized households and casts doubt on whether minimum subsistence
requirements are being met.
Table 7 presents data on household incomes for all strata. The poor
peasantry have both the lowest average annual cash income and average total
income.2 A comparison of poor and landless households, the most
proletarianized groups, shows that the poor have an average cash income 30%
less than full-time proletarian households; average total income only slightly
decreases the income gap between the two groups.2t
A second factor, related to total income formation, is also suggestive of
this sector's impoverished condition. With the exception of a single
household, agricultural production of corn, the main staple, was held for
personal consumption. Yet in no instance were corn output levels sufficient
to supplement, much less cover, dietary needs throughout the year. On
average, food crop production lasted about three months, although in several
cases the poor quality of the crop made it suitable only for animal feed.
In poor peasant households, then, migratory wage labor represents an
important means to meet basic consumption needs. Indeed, Table 6 shows that
this class derives almost one-third of its total monetary income from
migration remittances. Through the permanent and cyclical migration of some
household members, these units acquire resources that contribute to household
survival. In contrast to landless households, nearly 70% of households from
the poor peasantry receiving remittances used these remittances to purchase
such basic necessities as food and clothing rather than provide for an
improved standard of living.
Direct producers classified as middle peasants account for 22.8% of the
households in the survey. Within this stratum, approximately one-quarter of
all households had members engaged in migratory wage labor. All migrants from
the middle peasantry sent remittances; remittances, however, constituted less
than 15% of the total cash income of these households. In contrast to the
lower strata of rural households, remittances in this class were invested in
the purchase of land and other means of production rather than being used to
purchase basic needs or consumer goods.
In the majority of middle peasant households receiving remittances from
migrants (64.9%), farm equipment and livestock were the two principle
purchases. Migration by members of this class thus appears to be motivated by
the need to maintain a competitive balance' in the region's productive
structure. For a significant group of middle peasant producers, resources
gained from international migration appeared to keep households afloat as
producers of staple food crops. For a minority of middle peasants looking to
expand production, remittances provided a critical means to finance capital
outlays in the production of the region's most profitable crop, guava.
The capitalist/rich units of production, which account for 7% of the
households in the survey, dominate the agrarian productive structure in
Calvillo. Their superior position in production is linked to the underlying
process of capital accumulation in agriculture.l These households are involved
in the production of the most profitable crop in the region, guava. The use
of wage labor accompanied by a high degree, of inputs and mechanization
characterize the production process. It is primarily for these reasons that
labor migration in this class is not a characteristic feature. A number of
factors relating to household reproduction support this hypothesis.
In capitalist/rich households, agricultural commodity production
constitutes the primary source of income. Nearly the whole of their income
(93%) is derived from the sale of agricultural goods. The type of income
earning activity engaged in by Ithese units produces marked disparities in
income levels. Because commercial agriculture is very profitable, the upper
stratum of the peasantry has an average gross total household income more than
ten times greater than that of the middle peasantry (see Table 7). Financial
stability within this class, then, appears to account for the absence of
migration in this stratum. In region where wage labor migration represents
an important means to secure a variety of reproduction requirements, we would
not expect this type of migration to characterize the upper class.
The destination of migrants also appears to relate to the household's
position within the agrarian class structure. Table 8 shows the percentage of
migrants that sought work within Mexico or in the United States. Migration to
the United States constitutes the outstanding trend among the proletarianized
stratum. Migration to the U.S. also predominates among the middle peasantry.
Among poor peasant units, however, internal migration is the most pronounced
trend. The prevarious economic base of poor peasant units appear to restrict
international migration. Instead, internal! migration and the general
proletarianization; of households members constitute the primary
2. Household Structure and Migration. The migration index presented in
Table 9 gives a view of the extent of migration by sex across classes. Male
migration dominates in all class strata, but there are significant differences
Full-time laborer households have the highest rate of male out-migration
as evidenced by the migration ratio, 0.47. In this class, nearly half of all
males between the ages of 15 and 59 migrated on a temporary or permanent basis
during the survey period.24 Furthermore, this is the only class with a
relatively strong incidence of female migration.25 Although the poor and
middle sectors of the peasantry lag behind the migration rates established for
the landless, the middle sector exhibits a higher incidence of male
out-migration than the poor.
Among landless and poor peasant households, male migration relates
strongly to the existing gender and age division of labor within the region
and in the household. There are two key factors: the lack of permanent,
steady employment for men in the region and women's dual roles in productive
and reproductive activities.26 In Calvillo, the majority of completely
proletarianized households depend on wage work to meet consumption needs.
Wage income is primarily obtained from work in the guava fields or through
maquila domdstica, the home assembly of women's blouses, lingerie, and
infants' clothing. Within this strata, 65% of households engage in
agricultural wage labor and 57% take in piece work. Overall, more than 75% of
households depend on either guava or maquila for employment.
These two activities reveal the rigid sexual division of labor prevalent
in the region and in the household. Guava, for example, employs a male work
force.27 Maquila, on the other hand, employs women exclusively and uses
female children from the age of 7 or 8 as unpaid family workers.28
The limited range of permanent, salaried work available to male household
members shapes the pattern and intensity of migration. Employment in guava is
seasonal, the harvest extending from late September through early February.
Few jobs are available in guava during the off-season and other employment
opportunities--in construction (bricklaying), petty commerce or odd jobs in
the community (carpentry, plumbing)--are sporadic. During the off-season,
temporary migration is at its highest with migrants leaving after the harvest
and returning in the fall.
Women's roles in productive and reproductive activities are equally
important contributory factors in shaping the composition of the migrant labor
force. Women's work in maquila provides the household with a dependable
source of income throughout the year. Even though maquila earnings constitute
less than 25% of total income among rural proletarians, the availability of
steady, albeit poorly remunerated, work29 allows the male head of household
to migrate knowing the basic needs are being met in the interim before his
remittances arrive. The availability of maquila may also account for women's
overall lack of participation in migration.
Significantly, the nature of maquila work--paid work in the home--means
that the day-to-day responsibilities of household maintenance and child care
can be performed simultaneously with wage work. In short, women's combined
and interdependent roles in productive and reproductive activities ensure the
ongoing economic as well as social reproduction of the domestic unit.
The sexual division of labor both defines the household's relationship to
the labor market among the poor! peasantry as well as contributing to the
tenuous economic status of this class. In fact, low household income in this
class emerges in part from the social relations embedded in the gender
division of labor. First, this sector's extensive involvement in wage labor
as evidenced by the relatively large contingent of household wage earners is
due to women's greater participation in salaried work. That is, insofar as
wage employment accounts for the household's primary income source, the burden
of paid work falls on women. Slightly over half (52%) of all wage earners in
the poor peasant class are women as opposed to 38% of the landless category.
Second, the sexual division iof labor assigns to women the most poorly
remunerated work in the region, maquila domestica. Indeed, low household cash
income for this sector directly arises from the fact that a high percentage of
these households, in contrast to the full-time; laborer class, derive a major
portion of their cash earnings from maquila. Nearly 60% of households receive
between 40 and 100% of their income from home manufacture activities.
The consequences for migration, particularly international migration, are
clear. The meager income obtained from maquila restricts the household's
capacity to cover migration costs. Despite the attraction of higher wages in
the United States, this sector!does not have the economic stability to send a
migrant across the border. Migration patterns are decidedly regional and
national in character.
A number of factors relating to households' internal structures further
distinguishes these internal migrants from the international migrants of the
landless group. For example, who in the household leaves varies across these
two strata. Both sectors are dominated by male migration yet, as Table 10
indicates, male head of households largely comprise the international migrant
labor force among the full-time worker group while sons account for half of
the internal migrant stream among the poor.
The tendency for sons to migrant in the poor peasant class relates in part
to our earlier discussion of regional labor demand and the sexual division of
labor. In this class, however, increasing demographic pressure on the land in
the face of fixed or decreasing resources! compounds the effects of a
restricted labor market and a rigid sexual division of labor in production
(Young 1978). In ejidatario !households, for example, only one son, usually
the oldest, will inherit his father's plot. (With the exception of widows,
women are generally excluded from inheriting land). Sons of sharecroppers
have no guarantee of future access to land or other productive resources. The
extent of the problem can be gleaned from an examination of household size and
Table 11 shows average household size and -composition according to social
strata. Although households are large in all strata, the poor peasantry have
on average significantly larger units than either full-time laborers or middle
peasants. Furthermore, the poor peasantry is the class with the largest
percentage of extended family units, 36%.
Other studies in Mexico have argued that the extended family structure is
most characteristic of landholding units with greater resources (e.g., Arizpe
1980; Dinerman 1982). In Calvillo, however, the extended family household
appears to occur among the landed poor for reasons of basic survival. Where
wage work is absolutely essential to meet consumption needs, it is in the
economic interests of households to have as many potential laborers as
possible (Young 1978).
The large and extended household structure has been particularly
supportive of migrant households. Over 80% of migrants from the poor
peasantry come from either large or extended households. If land rights
depend on annual cultivation of the soil, as in the case of ejidatario units,
the household can sponsor a single migrant, usually a son, while other family
members remain to do agricultural work. Large and extended families, while
crucial to overall household viability for the poor peasantry, have had
differential effects on particular household members. In an area where access
to land and employment are both limited and circumscribed by the sexual
division of labor, sons have the highest tendency to out-migrate.
Male out-migration is also the dominant pattern within the middle peasant
sector. Three quarters of all male migrants from this class are dependent
sons involved in international migration. The drive toward acquisition of
means of production by these households and the high rates of participation in
migration by sons suggest that, to a certain extent, land resources represent
a viable employment option. In this class, land inheritance is not restricted
to a single male child--90% of these units are private owners or pequehos
propietarios; sons in these households may be investing in farming to secure
and maintain at least a portion of future subsistence requirements.
Nonetheless, the relatively small size of middle peasant plots, 3.7 hectares
on average, suggests that this may be an option available to one son only.
The higher median age of male migrants from this class (see Table 10) suggests
that older sons are vying for land resources.
The strong tendency to expel male household members and retain women has
in turn provoked changes in the economic and social organization of peasant
households (Margolis 1979). Among the poor peasantry in particular, a greater
tendency for men to migrate has altered the traditional sexual division of
labor. Women and children (12 years and younger) from this sector constitute
a significant proportion of the labor force on the family holding, comprising
43.7% of family farm workers. Additionally, in 60% of poor households women
assume major responsibility for agricultural production. In addition to their
tasks of weeding, spreading fertilizers, cutting beans, and husking corn,
women are involved in tasks traditionally performed by men--plowing, sowing
In both poor and landless households, women's participation in wage work
has increased in response to the high incidence of male migration. As noted
earlier, women's incorporation !into wage work is an integral part of the poor
and landless household strategy for subsisting and reproducing itself. During
periods of male migration, women supplement household income in a variety of
ways. In addition to doing maquila work, women take in laundry and ironing
and daughters work in domestic service, primarily in Calvillo City.
In contrast to the lower strata of rural households, the middle peasantry
does not appear to have altered the traditional gender and age division of
labor in response to migration by male household members. In this class (as
in the case of the capitalist/rich sector) women are not involved in paid
labor either within or outside of the home. !Women of the middle peasantry
contribute to household income formation through other income earning
activities. In almost half of these households, women work for pay as
self-employed seamstresses, but their earnings! account for less than 12% of
total household income. Women's work in agricultural production in the landed
upper classes largely consists of preparing and taking a mid-day meal to
husbands and sons in the fields.
In sum, household structure and organization, in concert with the
household's class position, isi both responsive to and the result of
migration. Economic pressures provoke multiple strategies of survival and
reproduction among different rural groups, affecting both the household's
productive base and the sexual and age division of labor.
This study has shown that the factors shaping migration processes must be
analyzed within a framework that locates larger economic, political, and
social issues within a concrete study of the regional structure of production,
class position, and household structure and organization. The particularly
strong interaction evidenced between household class position and migration
patterns in Calvillo underscores the importance of differentiating social
groups in the rural sector on the basis of their place in the relations of
For the four |classes identified in the region--full-time laborer, poor
peasantry, middle peasantry, and capitalist/rich peasantry--decisions to
migrate are uniquely grounded in the household's productive base. The least
commercialized units--completely proletarianized and poor peasant
households--have the highest propensity to migrate at the household level.
For landless households, migration to the United States appears to
substantially improve the household's standard of living. In contrast,
internal migration among the poor peasantry isl less an "option" than a vital
means to secure consumption requirements that contribute toward the very
survival of the domestic unit. Among the middle peasantry, migration is
primarily a means to enhance the household's productive base. The absence of
migration among the capitalist/rich peasant stratum suggests that its economic
dominance within the agrarian productive structure makes migratory wage labor
The focus on class status also shows how the internal structure of
different rural households affects and is affected by migration. The sexual
division of labor--specifically women's roles in production and
reproduction--and household size and composition vary widely across classes
and have a decisive effect on households' migration patterns. This analysis
also sheds light on why men and not women are the migrants in Calvillo.
The sexual composition of the migrant pool also reinforces key components
of class. Among landless and poor peasant households in particular, male
migration has important consequences for the household division of labor.
Women increase their participation in agricultural production and wage work
while retaining their traditional responsibilities for child care and family
welfare. Thus, the intensification of women's labor in paid and unpaid work
and productive and reproductive activities sharpens the analysis of class
structure and migration. It reveals how and which household members are most
vulnerable to and marginalized by changes in the household's productive base.
In summary, the consideration of class and household characteristics in this
study help not only to clarify why migration occurs but also to discern the
uneven effects of migration on rural households.
1. See C. Vasquez and R. Chiapetto (1981) fori a recent detailed examination
of the Mexican migrant population.
2. Calvillo, the second largest of nine municipios or counties in the state
of Aguascalientes, has approximately 37,000 inhabitants. A stratified
random sample of rural communities in the region was taken by the author
in 1982. Eight communities and 59 households were selected for analysis.
The Calvillo study forms part of a larger comparative project undertaken
by the author, analyzing the relationships among class, household
structure, and migration in the three most important agricultural regions
in Aguascalientes--El Valle, IEl Llano, and Calvillo. Unless otherwise
noted, all data in this paper refer toI the author's 1982 survey of
Calvillo. Funding for the project was generously provided by the Social
Science Research Council and the Inter-American Foundation.
3. An ejido is a landholding unit in which ownership and administration is
legaTTy vested in a community responsible for allocating cropland to
individual ejido members, ejidatarios. Ejido land may not be sold, rented
or transferred to nonmembers of the ejido (World Bank 1978).
4. According to the 1930 Agrarian Census of Aguascalientes, 75% of Calvillo's
landholding units were held by small property owners. There were no
ejidatarios. In 1944 the Delegaci6n Agraria reported 216 ejidatarios in
the municipio of Calvillo, about 1.4% of the total number of ejidatarios
in the state of Aguascalientes. In terms of land size, 47% of the
landholding units over one hectare in 1930 had an average size of 4.5
5. Prior to 1940, the principal migrations occurred after the introduction of
the railroad (1900) and during the turmoil of the civil war (1910-1917).
In both cases migration Was primarily to the United States and northern
border areas (Rojas Nieto 1981).
6. According to Bustamante (1975) undocumented migration to the United States
first emerged as a wide-scale phenomena during the years of bracerismo.
Internal migration was largely toward Mexico's major urban centers, Mexico
City, Guadalajara, and Monterrey.
7. By 1940, 22% of all farmland, including ,47% of the cropland, had been
distributed to over half ofi the country's rural population (World Bank
8. Guava production accelerated in the 1950s. In this period, only 52
hectares of guava were cultivated as opposed to 3,068 hectares of corn and
beans. At this time, the production of guava was exclusively controlled
by private units. By 1960 nearly ten times more land had been taken into
cultivation of guava while areas sown fori basic crops less than doubled.
During this period, private producers controlled all but a single hectare
of the land used for guava. In the 1970 Agrarian Census, land for guava
was reported to be 1,666 hectares, that is, an increase of 227% over
1960. Furthermore, for the first time, the amount of land devoted to
guava was greater than that devoted to corn and beans. While private
producers maintained their dominance over guava production, ejido units
now accounted for 3% of the land devoted to guava. (Censo Agricola,
Ganadero y Ejidal 1950, 1960, and 1970).
9. From 1957 to 1973, the government regulated the price of corn (Arizpe
1981). The idea behind the price controls was to ensure rural income
levels as well as to control the price of the nation's primary staple. In
practice, however, government-guaranteed prices meant the displacement of
corn production, because producers could buy corn on the market cheaper
than they could produce it for themselves and their families. A 1982 ECLA
study states that 33% of the smallholding peasantry are net buyers of corn
10. In my 1982 Agrarian Survey of Calvillo, 53.8% of landless households
reported having had land previously.
11. Most agrarian censuses of Latin America divide rural groups on the basis
of landholding size. Governments and international financial
institutions, e.g., the World Bank, typically rely on this kind of
division for their analyses of the rural sector.
12. In the Latin American context, researchers associated with the Centro de
Investigaciones Agrarias (CDIA) and the Economic Commission for Latin
America (ECLA) have used this approach. See Reyes Osorio, et al. 1974 and
Domike and Baraclough 1972.
13. In her analysis of the Mexican rural proletariat, Pare (1979:42) addresses
this issue in her critique of Stavenhagen's (1968) classification of rural
classes. She writes, "Though Stavenhagen's conception has the merit of
revealing the proletarian character of the landless peasant ... it
emphasizes income levels and standard of living and leaves aside the
problem of accumulation and exploitation."
14. See Lenin (1972) for an equally powerful argument against classifying the
peasantry on the basis of the size of the peasant holding.
15. Patnaik's E, as a quantified measure of exploitation, does not correspond
to the rate of exploitation or S/V defined by Marx (1975) as the ratio of
the surplus value to the variable capital.
16. Patnaik's original equation specifies an additional type of (indirect)
labor relations: renting-in/-out of land where labor is indirectly
appropriated through rent payments. In this study we do not consider
Patnaik's land renting concept in classifying rural households for two
reasons. First, leasing-in and -out of land was difficult to document in
the Calvillo survey because 'of ejidatarios' reluctance to reveal land
renting practices. Second, a rigorous empirical application of the theory
of rent has not been employed by Patnaik.
17. In my agrarian survey of Calvillo, the capitalist and rich peasantry
comprise a single class. This seems an appropriate conflation because the
upper strata of the cultivating population evidence a homogeneous social
and economic structure vis-a-vis other classes. Additionally, the
capitalist units of production surveyed in the three regions have not
reached a level in which the division of labor can be characterized by an
absolute separation between manual labor and supervisory tasks; thus the
concept of a pure capitalist as defined by Patnaik does not hold.
18. At the time of this study, the 1980 census data for the state of
Aguascalientes were not available.
19. In Table 2 the census category "private unit of production" refers to a
single production unit or household, whereas the "ejido" category refers
to a group of, landholding households. Thus, the four ejidos that are
reported to exist in Calvillo do not accurately reflect the distribution
of land among ejidatario households.
20. Before the devaluation of the peso in February 1982, one U.S. dollar was
equivalent to approximately 26 Mexican pesos. Between February and August
1982, one U.S. dollar was equivalent to 49 pesos.
21. The simple correlation coefficient between the E ratio and irrigated land
is 0.49. The correlation' coefficient between work animals and rainfed
land is 0.79.
22. In Table 7, the category "total income" includes all income sources--from
marketed and nonmarketed agricultural output, wage labor, income generated
from self-employed activities, sharecroppihg revenues, rents, government
subsidies, migration remittances, etc. It does not include goods or
services received from family, friends, land formal institutions. The
category "money income" is total income minus the value of nonmarketed
23. In considering the poor peasantry's low income level, it is interesting to
note their extensive involvement and dependence on wage work. For
example, wage income accounts for about 80 to 90% of total income earned
by poor and completely proletarianized households, respectively. The
average number of wage earners in poor households, however, is greater
than among theilandless strata, 3.4 as opposed to 2.6.
24. In this study, a temporary migrant is someone who left the community for
at least oneI month for work purposes and returned within the study
period. A permanent migrant is someone who was living and had lived more
than half a year out of the community when the survey was taken. The
tables on migration refer to both temporary and permanent migrants.
25. In the full-time laborer class, four of the five women migrants migrated
to the United States with their husbands. In general, however, the small
size of the female migrant pool does not permit an adequate treatment of
the phenomenon of female migration in this study.
26. The term reproduction has meaning on several different but interrelated
levels: biological reproduction; the daily maintenance of the labor
force; and social reproduction, or the reproduction of the whole society.
27. Although the widespread practice of paying workers according to the number
of kilos of guava picked and boxed (a destajo) often encourages entire
families to engage in agriculture wage work, adult men and male children
above the age of 13 generally constitute the contracted work force.
28. In contrast to the guava industry, the maquila industry has no local
base. Textile and clothing manufacturers from the state capital,
Aguascalientes City, account for some of the distribution of materials
among rural households in Calvillo. For the most part, though, large
national firms from Mexico City and Guadalajara, Jalisco, have set up
elaborate networks of intermediaries in charge of distribution, collection
of the finished product, and payment.
29. Wages vary widely within the maquila industry. Elaborately hand-stitched
blouses (deshilados), for example, take a single highly skilled woman 4 to
5 days to complete, working an average of 8 hours a day. For each blouse,
women receive $150 pesos. Even with many family members involved, few
households can finish more than 4 blouses a week. Machine sewn designs on
children's blankets are remunerated at 6 pesos per blanket. Between 60
and 80 blankets can be assembled in eight hours. Women
machine-embroidering designs on women's lingerie are paid between 5 cents
and 1 peso per garment. It takes approximately 3 hours to earn 20 pesost
In addition to receiving extraordinarily low wages, women doing maquila
work face harsh working conditions. Long hours of closely detailed work
with poor or no illumination has many women complaining of severe
headaches and loss of eyesight after several years. If sewing machines
are used (usually rented-in), overhead costs such as electricity and
maintenance are borne by the household. Thread and needles must also be
provided by the worker.
CLASS STRUCTURE I. THE BUYING AND SELLING OF LABOR POWERa
Household Number Average Average Average Average E Ratio
Classifi- of Family Labor Labor Net X1-X
cation House- Days Days Days Labor
holds worked Hired in Hired out Days
(Y) (Xl) (XO) (X1-XO)
Capitalist/ 3 338.0 501.0 5.0 496.0 1.46
Middle 17 356.0 30.0 30.0 0.0 0.00
Poor 13 158.0 0.5 795.0 -794.5 -5.02
Full-Time 26 0.0 0.0 502.0 -502.0 -
Source: 1982 Agrarian Survey of Calvillo
aLabor days (family, hired in and hired out) are calculated on a per person
per day basis. Hired in labor accounts for wage workers employed on the
operational holding only; hired out labor refers to wage labor performed in or
outside of the household. In households where petty commercial activities are
involved (such as fruit and vegetable street vending) or self-employment
exists within the home (such as working as a seamstress or owning a small
store) and no land is held, the households have been classified as full-time
laborer. Numbers in parentheses indicate percentage of total households
DISTRIBUTION OF ARABLE LAND BY CROP AND TYPE
Census Number Land Devoted to Land Devoted to Total
Category of Commercial Crops Basic Crops
Units (Guava) (Corn and Beans)
Rainfed Irrigated Rainfed Irrigated
Ejidos 4 7 48 1,399 662 2,116
Units 342 39 281 237 48 605
Units 514 94 1,356 2,945 208 4,603
Total 856 140 1,685 4,581 918 7,324
Source: Censo Agricola, Ganadero y Ejidal, 1970.
aOne hectare equals approximately 2.5 acres.
CLASS STRUCTURE II: LAND TYPEIAND CROP TYPE
Household Number Land Land Type Crop Type
Class Of Owned Irrigated Rainfed Output
Type Households (%) (%) (%) Guava Corn/Beans
Capitalist/ 4 22.9 67.9 13.2 98.0 11.9
Middle 13 33.8 30.9 34.5 1.9 48.1
Poor 14 43.1 0.0 52.3 0.0 48.0
Full-Time 26 0.2 1.2b 0.0 0.1 0.0
(l0.O)a 100.0 100.0 100.0 100.0 100.0
Source: 1982 Agrarian Survey of Calvillo
aTwo missing values. Numbers in
bConstitutes a family garden.
parentheses indicate percentage of total
CLASS STRUCTURE III: NONLAND MEANS OF PRODUCTION
Household Number Instruments of Productiona
Class of (%)
Type Households Mechanized Nonmechanized
Own Rent Tools Animals
Capitalist/ 4 100.0 14.3 34.1 5.8 66.2
Middle 13 0.0 64.3 26.1 31.9 16.0
Poor 14 0.0 21.4 29.1 58.0 15.0
Full-Time 26 0.0 0.0 10.7 8.3 2.8
Total 57c 100.0 100.0 100.0 100.0 100.0
Source: 1982 Agrarian Survey of Calvillo
aAn index was constructed for each category.
percentages based on an index of 1981 market prices of cows, pigs, and
CTwo missing values.
MIGRATION BY CLASS, 1981-1982a
Class Type Number of Number of Households
Households with Migrants
Capitalist/Rich 4 0
Middle 13 4
Poor 14 6
Full-Time 26 12
Laborer (45.6) (46.1)
Source: 1982 Agrarian Survey of Calvillo
aClass type in this and subsequent tables refers to the classification of
households given in Tables 3 and 4.
PERCENTAGE OF MIGRATION REMITTANCES BY CLASS, 1981-1982
Class Number of Households Households Remittances
Type Households w/Migrants Sending as a % of
Remittances Total Monetary
Capitalist/Rich 7.0 0.0 -- --
Middle 22.8 30.7 100.0 14.1
Poor 24.6 42.8 66.6 31.7
Full-Time Laborer 45.6 46.1 91.6 28.5
Source: 1982 Agrarian Survey of Calvillo
AVERAGE ANNUAL GROSS MONEY INCOME AND AVERAGE
(HUNDREDS OF PESOS)
ANNUAL TOTAL INCOME
Class Average Annual Average Annual Gross
Type Gross Money Income Total Income
Capitalist/Rich 11,911 11,942
Middle 865 1,338
Poor 800 828
Full-Time Laborer 1,136 1,136
Source: 1982 Agrarian Survey of Calvillo
DESTINATION OF MIGRANTS BY CLASS, 1981-1982a
Type Aguascalientes Mexico United States
Capitalist/Rich -- -- --
Middle -- 17.6 82.4
Poor -- 67.1 32.9
Full-Time Laborer -- 22.2 77.9
Source: 1982 Agrarian Survey of Calvillo
aDestination refers to whether migrants sought work within the
Aguascalientes (outside of Calvillo), elsewhere in Mexico, or
in the United
WAGE LABOR MIGRATION
INDEX BY CLASS AND SEX, 1981-1982a
Class Migrant Actual Index Migrant Actual Index
Pool Migrants Pool Migrants
Cap./Rich 8 0 -- 7 0 --
Middle 15 5 0.33 19 1 0.05
Poor 29 7 0.24 30 2 0.07
Laborer 38 18 0.47 40 5 0.12
Source: 1982 Agrarian Survey of Calvillo
aThe index is calculated by dividing the total number of actual migrants by
the migrant pool or potential migrants, i.e., persons between the ages of 15
CHARACTERISTICS OF MALE MIGRANTS
(AGE AND KINSHIP) BY CLASS
Class Median Age Kinship (%)
Type (Years) Head Son Otherd
Middle 28 25.0 75.0 0.0
Poor 26 33.3 50.0 16.6
Full-Time Laborer 30 84.6 7.6 7.6
Source: 1982 Agrarian Survey of Calvillo
aIncludes son-in-law, brother, grandson, and brother-in-law.
Class Average Household Composition
Type Household (%)
Size Nuclear Extended Single
Capitalist/Rich 8.2 100.0 0.0 0.0
Middle 6.1 76.9 15.4 7.6
Poor 8.3 64.2 35.7 0.0
Full-Time Laborer 7.2 88.4 7.7 3.8
Source: 1982 Agrarian Survey of Calvillo
Archivo General del Estado de Aguascalientes (AGEA)
1945- Government documents on the operation of the bracero program in
1964 the state of Aguascalientes.
1980 La migraci6n por relevos y la reproducci6n
Centro de Estudios Sociol6gicos. Mexico:
1981 The Rural Exodus in Mexico and Mexican Migration to the United
States. In: The Border That Joins, ed. by P. Brown and H. Shue.
pp. 162-183. New Jersey: Rowman and Littlefield.
1975 Espaldas mojadas; material prima para la expansion
notreamericano. Centrol de Estudios Sociologicos.
Colegio de M6xico.
Econ6mica para America Latina (CEPAL)
Economia campesina y agriculture empresarial.
xico: Siglo XXI.
Corona Vasquez, R. and C. Ruiz Chiapetto
1981 Migrantes internacionales con y sin antecedentes de migracion
internal: algunas caracteristicas socioec6nomicas. Mexico,
Deere, Carmen Diana
1978 The Development of Capitalism in Agriculture and the Division of
Labor by Sex: A Study of the Northern Peruvian Sierra. Ph.D.
Dissertation. University of California: Berkeley.
Migrants and Stay-at-Homes:
from Michoacan, Mexico. San
A Comparative Study of Rural Migration
Diego: iCenter for U.S.-Mexican
General de Estadfstica, Mexico
Censo Agricola, Ganadero y Ejidal 1930, 1940, 1950, 1960, and
1970. Mexico: Direcci6n General de Estadfstica.
Domike, A.L. and S.L. Baraclough
1972 La structure agraria en siete paises de am4rica latina. In:
Lecturas sobre desarrollo agrfcola, ed. by E. Flores. Mexico:
Fondo de Cultura Econ6mica.
1980 La batalla en el Mexico rural.
Mexico: Siglo XXI.
Nacional de Investigaciones Agrarias (INIA)
Plan regional de investigaci6n agricola, Aguascalientes Mexico:
1972 The Development of Capitalism in Russia. Moscow: Progress
El papel de la mujer en la agriculture del Bajio.
1975 Capital, Vol. I, New York: International Publishers.
1979 El proletariado agricola en Mexico: campesinos sin tierra o
proletarios agrfcolas? Mexico: Siglo XXI.
1976 Class Differentiation Within the Peasantry: An Approach to the
Analysis of Indian Agriculture. Economic and Political Weekly
1982 Kinship Relations of Production in the Migration Process:
of Dominican Emigration to the United States. Unpublished
manuscript. New York University.
Reyes, Osorio, S., et al.
1974 Estructura agraria y desarrollo agricola en Mexico. Mexico: Fondo
de Cultura Econ6mica.
Rojas Nieto, Beatriz
1981 La destruccibn de la hacienda en Aguascalientes. Mexico: El
Colegio de Michoac6n.
1968 Neolatifundismo y Explotaci6n. Mexico: Nuestro Tiempo.
1979 El problema del proletariado agricola. In: Polemica sobre las
classes sociales en el campo mexicano, ed. by L. Pare. pp. 85-96.
1978 Land Reform in Latin America. Washington, D.C.
1978 Modes of Appropriation and
study from Oaxaca, Mexico.
A. Kuhn and A. Wolpe. pp.
the Sexual Division of Labor: A Case
In: Feminism and Materialism, ed. by
124-154. London: Routledge and Kegan
OFFICE OF WOMEN IN INTERNATIONAL DEVELOPMENT
MICHIGAN STATE UNIVERSITY
The WOMEN IN INTERNATIONAL DEVELOPMENT PUBLICATION SERIES were founded in
1981 to disseminate information rapidly to national and international
specialists in universities, government, and private institutions concerned
with development issues affecting women. The two series, WORKING PAPERS ON
WOMEN IN INTERNATIONAL DEVELOPMENT and the WID FORUM, publish reports of
empirical studies and projects, theoretical analyses, and policy discussions
that illuminate the processes of change in the broadest sense and encourage
manuscripts that bridge the gap between research, policy, and practice.
Publications in the series address women's historical and changing
participation in economic, political, and religious spheres, intra- and
inter-family role relationships, gender identity, women's health and health
care, and the sexual division of labor.
EDITOR: Rita S. Gallin
MANAGING EDITOR: Patricia Hubbard-Garcia
EDITORIAL ASSISTANT: Mary Sullivan
PRODUCTION MANAGER: Conni Zellar
DISTRIBUTIONS MANAGER: Stacy Pollock
EDITORIAL BOARD: Marilyn Aronoff, Sociology; Anne Ferguson, Anthro-
pology; Ada Finifter, Political Science; Peter
Gladhart, Family & Child Ecology; John Hinnant, Anthro-
pology; Susan Irwin, Anthropology; Akbar Mahdi, Soci-
ology; Anne Meyering, History; Ann Millard, Anthro-
pology; Nalini Malhotra Quraeshi, Sociology; Barbara
Rylko-Bauer, Anthropology; Judith Stallmann, Agri-
cultural Economics; Paul Strassmann, Economics
NOTICE TO CONTRIBUTORS: To provide an opportunity for the work of those
concerned with development issues affecting women to be critiqued and refined,
all manuscripts submitted to the series are peer-reviewed. The review process
averages two months and accepted manuscripts are published within five to six
weeks. Authors receive ten free copies, retain copyrights to their works, and
are encouraged to submit them to the journal of their choice.
Manuscripts submitted should be double-spaced, sent in duplicate, and
include the following: (1) title page bearing the name, address and
institutional affiliation of the author; (2) one-paragraph abstract; (3) text;
(4) notes; (5) references cited; and (6) tables and figures. The format of
the article may follow any journal of the author's choice. Submit manuscripts
to Rita Gallin, Editor, WID Publication Series, Office of WID, 202
International Center, Michigan State University, East Lansing, MI 48824-1035.
TO ORDER PUBLICATIONS: Publications are available at a nominal cost and
cost-equivalent exchange relationships are encouraged. To order publications
or receive a listing of them, write to Office of WID, 202 International
Center, Michigan State University, East Lansing, MI 48824-1035, USA.
ACKNOWLEDGMENT: The Women in International Development Publication Series are
partially funded by a Title XII Strengthening Grant.
MSU is an Affirmative Action/Equal Opportunity Institution