The Sociodemographic Effects of the Crisis in Mexico

Chapter Five: Men's and Women's Patterns of Intragenerational Occupational Mobility During Mexico's Boom and Crisis[1] Agustín Escobar Latapí

Introduction

During the late seventies, Mexico enjoyed its last period of economic growth, the so-called "Oil Boom"; the last period of prosperity in the "Mexican Miracle". In 1982, despite abundant oil exports, Mexico's external debt became unsustainable and Mexico plunged into its worst economic crisis in the twentieth century.[2] Up to 1984, no major economic reforms were made, save for "austerity" and wage controls. After 1985, substantial reforms began, in foreign exchange policy, wage policy, trade liberalization, financial markets, and the parastatal sector. But no economic improvement was visible until 1989, after the imposition of a new "pact" among the representatives of all major social, economic and political forces, and the renegotiation of Mexico's foreign debt. From that date until the end of the Salinista sexennium in 1994 Mexico's economy was more stable, with modest growth. The economic "model" has drastically changed, the occupational structure is changing, and so are the patterns of social mobility and class formation. So far, the new model has concentrated wealth (See Cortés and Rubalcava in this volume) and substantially reduced the opportunities for employment, particularly among middle and upper occupational strata.

This chapter examines the patterns of occupational mobility of men and women during Mexico's oil boom (1976-1982) and the subsequent economic crisis (1982-1990) in Guadalajara, this country's second largest city and home to over three million people in 1990. Rarely has there been such a marked contrast between two contiguous economic periods. While the boom was characterized by growth averaging above 6 percent and rates of job creation above demand, the subsequent crisis witnessed zero absolute growth in modern employment and negative per capita economic growth. With some peculiarities, the boom epitomized the social processes that transformed Mexico from a rural, agrarian country to a mostly urban, semi-industrial peripheral capitalist economy[3]. During the long period known as ISI, Mexico's middle and working classes were remade on new bases; they grew from insignificance to become the country's most important classes, and Mexico's occupational structure was modernized as self and family employment dropped rapidly.

During the crisis, on the other hand, self and family employment grew rapidly, and classes previously identified with the modern economy lost ground in the occupational structure. But although we possess some information on the changes in the structure of employment, there are no analyses of the changes in the patterns of mobility prevailing in these two contrasting periods.

This chapter analyzes two main questions: 1) To what extent did the crisis alter the prevailing patterns of occupational mobility for men and women, and 2) if (as expected) there was a clear change: did it affect men's and women's chances of mobility roughly equally, or differentially?, i.e. did it affect gender inequality as evidenced in occupational mobility? My analysis will instead center on the comparative evolution of the (occupational ) opportunity structures and on the gap in men's and women's chances of different kinds of mobility or immobility.

To answer these two questions, I examine the intragenerational or lifetime occupational trajectories of 3,600 men and 1,400 women living and working in Guadalajara, Mexico, in June-August 1990. These men and women were interviewed in the course of the (residential based) quarterly urban employment survey carried out by the government Statistics Institute (INEGI). To the questions usually asked in the survey, which deal mostly with the respondents' jobs during the previous two weeks, the author added two kinds of schedules, one aimed at the household head and one at each worker. This last schedule recorded the respondent's job characteristics in 1975 and 1982[4], as well as other basic life events. The schedule is therefore not a classic life history schedule, but a fixed format, short version of it.

Most men's class affiliation is based on their occupational status (Blau and Duncan 1967, Featherman and Hauser 1978, Glass et al. 1954). Although they may derive their income from multiple sources (traditionally in rural areas but also in most urban settings throughout peripheral capitalist economies), their class identity and their place in the social structure are usually believed to depend on one fundamental activity. The examination of men's mobility and achievement in the labor market and within society's occupational structure has therefore been recognized as a fundamental step in the definition of the class structure, of the range of opportunity within that structure, and of equality or inequality of condition and opportunity offered by it.

Women's class affiliation, on the other hand, cannot be said to rely to the same extent on their occupation[5]. Goldthorpe et al. outline three basic approaches and justifications for the study of women's class mobility. First, what they call the "conventional" approach is based on the recognition of women's dependence on men for their class affiliation. In this approach, women's class mobility is viewed as mobility from a family whose class or occupational affiliation is based on their fathers, to another, in which this affiliation is based on their husbands[6]. This kind of analysis (marital mobility) has usually revealed fairly close resemblances between men's intergenerational occupational mobility and women's intergenerational marital mobility, at least in Great Britain, France and Sweden (Goldthorpe et al. 1987, Portocarero 1989).

Secondly, what they call the "individual" approach studies women's occupational mobility as if it could be approached just like men's, that is to say, irrespective of their fathers' or husbands' occupational status. Although they regard this approach as radically flawed from the point of view of the definition of the class structure, Goldthorpe et al. find fundamental differences: women show far more downward social mobility than men. These differences virtually disappear, however, when viewed against the intergenerational change in women's overall occupational distribution. In other words, there is roughly as much social mobility of women within the total structure of the positions they occupy as there is mobility of men within the positions they occupy[7].

The last kind of analysis, which Goldthorpe et al. perform for Britain based on Erikson (1981), is referred to as the dominance approach, and allocates to women their own occupational status if they are unmarried or separated. If married and living with a husband, it allocates to a woman the status of the conjugal partner standing higher in terms of work-time dominance order. Lastly, if both partners stand equal in terms of the work-time dominance order, it allocates women to their own status if it is equal or superior to their husbands'. In this analysis, therefore, women are "assigned" to a social class corresponding to their own occupation more frequently than in the "conventional" approach, and they are assigned to other persons' class positions only when 1) they do not work, 2) they work less time than their husbands, or 3) their occupational position is lower than their husbands'. When this analysis was performed for Great Britain it fell short of perfect "commonality" with a social fluidity model derived from men, but it approached that model more closely than the second approach.

I intend to compare men's and women's individual occupational mobility tables to explore changes in the patterns of mobility and in the degree of openness and equality of the occupational structure for both men and women. In doing so, I am not equating women's occupational mobility with men's as an index of changes in the class structure; identical patterns of occupational mobility for men and women would in fact point at different social processes (Portocarero 1989). My focus lies instead on the changes brought about on the occupational chances of men and women by succeeding periods of rapid economic growth and pronounced crisis and adjustment. Rather than focusing on the class structure, therefore, I will focus on changes in the nature of the gender - specific processes of allocation and transference of individuals to different positions within dependent and independent employment.

But I am also concerned with men. Gender is a divide which, in the occupational structure at least, is historically constituted. Although at any given time the occupational structure and the labor market show gender segmentation, over time the lines dividing these jobs or occupations change, especially with economic restructuring. My concern is how these processes reshape men's and women's occupational chances and trajectories. As women flow into the market, employers are confronted with new choices. In some cases this results in "feminized" strata, such as among dependent professionals and dependent manual workers, where some men have been displaced. Also, the surge of women in the market has abetted the government's policy aimed at lowering wages. This in turn has led to the growth of self-employment.

The following analysis differs from most other gender analyses of social mobility in three respects: it refers to two historically specific periods; it deals with the population converging in a city after two sharply contrasting, short historical periods; and the focus is on intragenerational occupational mobility. In the near future, however, we hope to analyze both intergenerational mobility and intragenerational mobility over longer time spans and in a selection of Mexican cities.

2. The National Occupational Structure, 1975-1990.

As said before, Mexico's occupational structure modernized rapidly after 1935, with agriculture losing importance, and urbanization and industrialization triggering rapid changes. Waged manual workers in industry, transport, construction and the services did increase in importance within the total occupational structure from 1940 to 1980; their share of employment grew from 32.8% to 36.5%. Nevertheless, they changed drastically in character, organization and income. Towards 1975 they were working mostly in modern enterprises of different kinds, half were unionized and protected by Mexico's modest but valuable social security program[8] and they received the highest wages ever earned by workers in Mexico. At the same time, self-employed and family workers decreased their share of employment, from 37.9% to 18.6%, as did domestic service workers, who went from 10.7% to 5.3%. Changes in the urban manual occupations were based only partly on the loss of agricultural jobs. The dynamics of modern urban employment were more influential.

The strata gaining importance during the period were nonmanual workers, with employers, managers and professionals growing from 4.5% to 13.4% and lower nonmanual workers growing from 14.1% to 21.6%.

A sector by sector analysis of this change in the higher strata reveals that, although increased government employment caused a great deal of it, another significant portion is explained by the modernization of industry and the accelerated growth of modern private services, especially financial ones (Escobar and Roberts 1991)[9]. As we explained in the chapter cited, the transformation of the structure seemed to slow towards the late seventies, with self, family and traditional service employment rising once again (García 1988). This was accompanied by growing problems in managing urban growth, rising real interest rates, and soaring external debt. The debt became unsustainable in 1982, and this triggered the crisis.

The 1980's brought drastic changes in several areas of the occupational structure. Firstly, since both government and corporate employment stagnated in absolute terms and lost ground in the structure, nonmanual strata also lost considerable ground. On the other hand, lower wages undermined household incomes pushing more workers into employment, which expanded the structure at its base in "marginal" occupations[10]. Self and family employment, which started rising slowly in the late seventies, soared in the eighties, and so did informal employment in its many guises[11].

Second, women's participation in employment rose. This tendency had been evident in the earlier years of ISI when women entered mostly lower nonmanual occupations in such great numbers that their employment structure showed, compared to men's, proportionately larger numbers of nonmanual workers (in teaching, nursing and office employment). But during the crisis women entered the lowest tiers of the structure in greater numbers than ever. As a consequence, their employment structure suffered more than men's. Women professionals lost ground within women's employment structure, even though there were proportionately more new female entrants to professional occupations than men[12]. At the other end of the structure, manual occupations are now much more feminized than at the beginning of the last decade, and so are the self-employed and family workers: unpaid family workers are mostly women, and they comprise a larger share of women's employment than at any other time during the last fifty years.

The changes in the relative importance of different occupations for women correlate with women's participation in the occupational structure (García and Oliveira 1991). In 1976, participation was clearly concentrated among more highly educated, single women, and the presence of small children in the household strongly inhibited it. By 1987, education no longer had a positive impact on participation, and the presence of small children in the household was not inhibiting it either. In other words, unschooled, married women were remaining in employment - or entering it - in spite of their small children and of the fact that child care centers, nominally a part of the Mexican Social Security Institute (IMSS), are inadequate and scarce[13].

3. - Guadalajara: General Tendencies

Guadalajara, the second-largest city in Mexico, with approximately 3.5 million inhabitants in 1990, developed rapidly after 1940 as an important regional center producing consumer goods and providing services to Western Mexico. By the late seventies it had lost its "regional" character and had become one of three national urban industrial centers (De la Peña 1986). Manufacturing concentration in the Mexico City Metropolitan Area meant however that Guadalajara only produced 4% of Mexico's GNP, an amount proportional to its population. Guadalajara's comparatively scant contribution to manufacturing GNP together with the (mostly) small scale of its industry have helped label it a "service center". In 1990, however, it was more heavily industrial -- in its employment structure -- than either of the other two metropolitan areas of Monterrey and Metropolitan Mexico City, partly because of the small scale and labor intensity of its industry. Its primacy over the state of Jalisco is absolute (it is twenty seven times larger than the next city in the state), and it is becoming synonymous with the state demographically, since it houses over 70% of its total population.

The employment structure seems to have changed in Guadalajara in the same general direction as in the rest of urban Mexico. Incompatibilities, large amounts of "insufficiently specified" occupations in the 1980 census and lack of a table breaking down occupation by gender in the 1970 census bar a detailed appraisal of this change. But some tendencies can be gathered from the available tables. Professional, technical, teaching and managerial strata seem to have increased from approximately 11.9% in 1970 to 15.6% or 17.3% in 1980, depending on the inclusion or exclusion or "insufficiently specified" occupations. Agricultural employment declined from 5.6% to just over 1% in the same period[14], as the metropolitan municipios were totally urbanized, and the importance of nonagricultural (manual) workers increased, from 35.9% to 39.6% or 43.9%, again depending on inclusion or exclusion of insufficiently specified occupations.

The lower tiers of Guadalajara's manufacturing and government labor market prior to the crisis have been analyzed in detail elsewhere (Escobar 1988, Roberts 1990). I can briefly summarize our argument. Towards the end of the boom Guadalajara possessed a mobile, competitive labor market, in which there was considerable exchange between large and small enterprises. There were also a few well established, capital intensive industries which required highly specific skills in which workers tended to stay a long time. The recent transformation of the urban employment structure was evident in the mobility of labor: workers with full primary (six years) or secondary (nine years) schooling frequently reported recently having moved to better paid jobs, and unschooled recent rural migrants were easily entering formal factory jobs. Women reported that their wages had improved recently much more often than men, and there was evidence of some displacement of women from domestic service employment to factory jobs, which at the time involved doubling or more their income.

In an economy in which large firms coexisted with small ones in every branch of production, informality seemed a key element of the world or work. Most firms could be arranged along a continuum both in terms of the formality of labor relations and in terms of the formality of firm - government relations. Even when firms lacked any formal relationship with the government, i.e. permits, registration as taxpayers of different kinds, recognition by labor or sanitary authorities, etc., they had established some sort of stable relationship with one or more state agencies, be it through bribes, an informal agreement on payment of certain taxes, or tolerance of inspectors. In other words, informality had nothing to do with the government apparatus's technical capacity for detection - and control - of these firms, but rather with the government and the firms' ability or willingness to perform legally and pay taxes. Some small employers provided significant benefits and security to their workers, while many large ones did not. Lastly, skilled men and schooled women moved easily between very large firms and small workshops. Men earned equally well in both settings (around 2.5 minimum salaries at the time), and women earned less in the shops but they had the possibility of working flexible hours. This meant that the city's work force, in manufacturing at least, could not be split into a "formal" and an "informal" work force, both because of the variable distribution of formality of employment and because skilled men, and some women, moved between typically "formal" and "informal" work settings frequently. Large and small firms were parts of the same opportunity structure.

The global changes in Guadalajara's employment structure and labor force during the eighties can be explained by the specificity of its local economy and the absence, not the presence, of powerful actors. Guadalajara's decline into the crisis was gradual, unlike Monterrey's, and the expansion of maquiladoras[15], although visible, had as of 1990 no significant impact on the structure of employment. This distinguishes Guadalajara from the northern border towns. After 1982, the general decline in wage levels and the dispersed nature of economic units in Guadalajara pushed considerably more individuals into the labor market, which further depressed wages, especially in those small-scale waged enterprises where cheap labor could be bought for its market price. This slowed or halted mobility between large and small enterprises, and "reinformalized" the small-scale manufacturing sector (Escobar, 1988). A follow-up study of poor households in the city found that the participation rate of women over 15 years of age increased 25% from 1982 to 1985, and that these women's new occupations were concentrated in the informal services (González de la Rocha, 1988). Small manufacturing workshops also changed their wage levels and employment structures: wages fell in these workshops more drastically than in large factories, and the percentage of young workers in workshops increased rapidly (Escobar, 1988).

In Guadalajara as elsewhere in Mexico, nonmanual public sector employees lost over half their 1980 purchasing power, but recovered part of the lost ground in 1990. Nonmanual private sector employees, especially those in modern multinational enterprises, suffered a small loss in purchasing power from 1982 to 1987, but later made sharp gains, and ended the decade with better overall incomes than in 1982, according to the data I obtained from employee compensation consultancy firms in Mexico City and Guadalajara. Employment in big firms and in the public sector, however, failed to increase during the crisis, meaning that it lost importance in the general structure of employment, since the growth of the total labor force was very strong. Based on the results of a quarterly employment survey, Rendón and Salas (1991) report that, between 1987 and 1989, local self-employment increased at a rate of 30%, while unwaged work increased at a rate of 27%[16]. In both cases, the rate of increase is faster for women than for men. These rates are lower than those for Mexico City, but provide a clear indication of the changes in the structure of employment.

How does Guadalajara contrast with other Mexican cities during the crisis? This depends significantly on the period of reference. The crisis of the eighties cannot be understood as a linear phenomenon in economic or social terms. There was an about face within the crisis in 1988, and this changed Guadalajara's position. Up to 1988 Guadalajara ranks fairly high: compared to the border towns, Guadalajara was slightly worse off due to the lack of a dynamic modern sector such as maquiladoras, which were not simply increasing their blue collar employment, but whose expansion gave those economies a general boost derived from construction and commerce. Compared to Mexico City, Guadalajara was better off, because its independence from government expenditure exempted it from the blow that hit Mexico City during the years of extreme austerity. Vis-à-vis Monterrey, Guadalajara was much better off because its industry was far less indebted, most of its debts were with Mexican, not foreign, banks (Alba, 1986), and Guadalajara's cheap, semi-durable consumer goods suffered less of a downturn than Monterrey's consumer durable and capital goods production.

Although there are no comparative figures for the period after 1988, a number of indicators point at a worsening of Guadalajara's position among Mexican cities. Nationally, the period after 1988 brings several important changes: 1) trade liberalization together with stabilization of the economy and lower inflation; 2) a recovery in the government's "programmable"[17]and social spending; 3) growth in exports and some growth in industrial output generally, with stabilization of wages at a thirty - year low. One of the consequences of these changes is the recovery in construction and the transport and capital goods sectors, after a period of restriction in this area by both firms and individuals.

What this means for Guadalajara is mostly negative: the amount and kind of goods bought by workers has not increased significantly, and a growing proportion of their needs may be satisfied through imports of cheaper consumption goods like clothing and shoes, previously among Guadalajara's strong industries, and now undersold. There is very little capital goods or automobile-related industry in the city, and consequently the recovery in this sector has not helped Guadalajara's recovery. Growth in the government's programmable and social spending has benefited Mexico City considerably, but Guadalajara has failed to attract resources through government social programs like Solidaridad. Although some local corporations have succeeded in expanding outside the state of Jalisco and beyond Mexican borders (Escobar, 1990) - the two most remarkable would be Grupo Sidek, whose expansion into resort development after 1987 has been extremely successful, and Arancia, a highly integrated food consortium - these activities have failed to spur generalized growth. Unlike the guarded optimism in Mexico City, Monterrey or the border in 1991, the feeling in Guadalajara was that things were worse then than "during the crisis" (before 1988).

International emigration from Guadalajara increased during the eighties. The proportion of Guadalajara's work force with a background of international migration has dropped from 17% since our first labor market study in 1982, to 4% in our last study, in 1990. On the other hand, the number of working class or lower nonmanual households reporting one or more members in the U.S. has increased[18]. This is most likely due to the more frequent departures and longer U.S. stays of those workers with migration experience[19] reported by the Ministry of Foreign Affairs (SRE), to the shift in workers' comparative assessment of life and work in Mexico and in the U.S., and to the process of legalization many of them started under IRCA (Immigration Reform and Control Act).

Assessing the changes in the structure of employment between 1980 and 1990 should be easier because there was a better categorical match in the 1980 and 1990 censuses. This is however only partly useful because, as several analysts contend, the 1980 census overenumerated workers and allocated an undue amount to "insufficiently specified" occupations, while the 1990 census underenumerates some workers, especially women. From 1980 to 1990, according to these two censuses, women became less important in the labor force: while they accounted for 32% of the work force in 1980, the 1990 census reports they only accounted for 30.5% of the work force. This runs counter to the findings of all major studies on women's employment carried out during the decade of the eighties (García and Oliveira 1994), and to our own study, which found that some of women's occupations were made up largely of women without work experience, or of women who returned to employment after a long period away from the labor force: among women the following percentages in 1990 had not been in the work force in 1982: 58% of the managerial-professional stratum, 69% of those in dependent manual work, 69% of those in clerical and sales occupations, and 61% of the self employed. The corresponding proportions for men are much lower: 18%, 40% and 41% respectively, which suggests that women's participation in the labor force did in fact increase much faster than men's. This should have produced a greater importance of women in the labor force in 1990 than before.

Though risky, comparing the results of the 1980 and 1990 censuses produces qualified good news for women's employment structure. Professional and managerial categories account for a larger share of women's employment in 1990 than in 1980: the first grows from two to four percent and the second from 1.3 to 2.3%[20]. The importance of domestic service seems to drop a little, from 10.5 to 9 percent, and women account for a larger proportion of office workers than before. But the category expanding the most is "street vendors", which grows from 0.7% to 2.8%. As will be seen below many of them are women. Still, the figures are not perfectly comparable: all these changes are so small that they could be accounted for by excessive allocation to insufficiently specified occupations in the 1980 census.

4. - Changes in Men's Mobility

The first two tables in this section summarize men's mobility from 1975 to 1982 and from 1982 to 1990. The analysis aims to establish the extent and nature of mobility, and how it changes from one period to the next.

This and all other tables exclude individuals moving to or from direct agricultural employment (68 in this case, and less in all the others). Individuals working in agricultural related activities and services are nevertheless included.

Categories are as follows: I. university degree professionals earning a salary or commission, managers and employers of more than one person[21]; II. teachers, technicians (including nurses); III. office workers not falling in the previous two categories and sales workers; IV. all dependent manual wage workers in manufacturing or the services, and V. workers who are self-employed who do not employ more than one person themselves, and street and stall vendors.

Table 1 describes the occupational mobility of men from 1975 to 1982, with outflow percentages in the lower part of each cell. This includes some occupational mobility to and/or from places other than Guadalajara. In other words, not all this mobility took place within the city. Also, the transition tables for men and women during the boom are subject to the "survivor" problem inherent in this kind of study, namely that they underestimate the kinds of mobility that take a person out of the sample such as death, departure from the city or retirement. Furthermore they overestimate the degree of permanence in occupations. The marginal 1975 distribution of occupations therefore only approximates the actual occupational structure at that time. The table reveals the kinds of mobility that took place among categories, the impact of this movement on the structure, and the openness and inequality of the structure. The reader should bear in mind therefore that this is not a local study from 1975 through 1990, but rather a study of the trajectories of a population converging in the Guadalajara occupational structure in 1990, a population with many different histories but one common outcome.

[Table 1 goes in here]

Differential survival is not a major problem for men. Most men remain in the labor force continually until senility in Mexico, since retirement plans provide only token pensions for most Mexicans. They either refuse to retire, or perform new occupations on nominal retirement. The men's marginal 1975 distribution (considering categorical deviations) is consequently a fair approximation to the 1975 occupational structure. It does not approximate the women' case as clearly, as will be explained below.

The categories employed in the table are similar to those used in other countries, with one significant exception. I have chosen to distinguish all those individuals in self-employment, whether nominally professional or not, because self-employment has remained important in Mexico throughout the twentieth century, and because this category grows rapidly during the eighties. Distinguishing the self employed allows us to establish whether the occupational structure was "modernizing": whether large scale and corporate employment tended to grow or diminish, and the extent to which this caused different kinds of mobility. I have elsewhere explored the meaning of self-employment both in the context of the formality of work and of the relative social position of self-employment vis-a-vis other categories (Escobar, 1993).

It seems that the transition of professionals to self-employment moves them down socially and economically. Women moving to self-employment from all other major categories lose a significant part of their previous income, work long hours and deploy skills which are not socially recognized (petty trade, laundering, domestic outwork). The most important factor in this present analysis of self-employment, however, is whether people are moving out of or into the corporate employment structure.

Although I do not subscribe to dualist theories of social structure for Mexico, the market difference between conditions of employment in the modern, corporate or formal sector and self-employment is very important. I prefer the term "self-employment" rather than "informal worker" or "informal sector" because of the marked lack of agreement among analysts on the nature and definition of "informal" as well as notorious social, occupational and economic disparities within it (i.e. the owner of a large illegal workshop is occupational distant from one of his apprentices or from a street vendor but both are often classified as belonging to the "informal sector")[22]. There is some heterogeneity within those self employed, but it is much less than within "informal sector worker", and describes work relations better. Self-employment is a far more accurate category to gauge people's responses to crisis and restructuring than informality. The move to self-employment has almost saturated many typically small scale activities, like petty commerce, repair shops, and taxi driver/owners to the extent that the transition to self-employment ceased to be attractive. Further employers had to raise the wages of some workers with "general" skills, especially mechanics.

Table 1 shows that men experienced a high degree of occupational stability during the boom. This is understandable in view of the short time span covered. The most stable categories are I, IV and V, where more than 80% of those starting the period are immobile. Categories II and III, which correspond roughly to "Intermediate" categories in other analyses, show less stability, but the patterns of mobility differ between them. The single largest group leaving category II (technical workers and teachers) goes "up" to the professional-managerial stratum, while the single largest group leaving category III (office and sales workers) goes "down" to dependent manual work. However, both lower manual categories contributed significantly to the growth of the professional-managerial stratum, and so did even the dependent manual work stratum, which, in spite of the modest percentage of outflows to the professional managerial stratum makes a significant contribution to its composition. Even during the boom, self-employment grew via occupational mobility from other categories, - not only from newcomers - but the contribution of each category to this growth is small. Those leaving manual dependent employment, the largest single group moving out of that category, are an exception. The author had previously documented an important life cycle phenomenon among dependent manual workers in Guadalajara, where employers laid off older workers (who then set up on their own) or else helped them set up their own shops when they arrived at a certain age and a certain level of expertise, and could not aspire to promotions within the limited hierarchical structure of this city's industrial firms (Escobar, 1984).

Men's mobility during the crisis (1982-1990) differs in some important respects from that of the previous period. There is much less stability in Table 2 and the patterns of mobility are also different. While the most stable category during the boom was that of professionals and managers, during the crisis the most stable category (in outflow terms) is self-employment (V). In general, all categories lose significantly more members to self-employment than during the boom, and this is no longer a consequence of the progression in the sample's working lives. Among workers under 25, self-employment goes from 11.2 percent in 1982 to 15.1 percent in 1990; and among middle age workers over 35 it increases from 22.,9 percent to 36.0 percent.

[Table 2 goes in here]

Remarkably, however, mobility "up" from all other categories to stratum I does not diminish. Even self-employment contributes more to category I than during the boom. The net losers are stratum III and IV, and the fastest growing category is category V, followed by category I.

Tables 1 and 2 can only be compared in general terms. A formal appraisal of change is not possible due to the existence of a few empty cells (a consequence of concentration of respondents on the diagonal) and the low Ns in some others. An analysis of changes in social fluidity lies in columns I and V. In column I, it can be seen that categories II through V increase their outflows to category V. So, while there is an increase in the movement towards self employment and this becomes the most stale category, there is at the same time an increase in mobility towards the highest stratum. The increase in mobility towards the professional managerial stratum is much less, both relatively and absolutely, than the increase towards self employment.

4. - Changes in Women's Mobility

Women's occupational mobility during the boom differs from men's in two related ways. First, the nonmanual strata (I, II and III) are far more stable, due, in part to differing employment conditions in these strata. Women in stratum I tend to work for large organizations and the government far more frequently than men, and permanence in these large organizations is crucial for promotion. Similarly, in stratum II, women tend to be teachers or nurses, while men tend to be technicians working for industrial or service enterprises. For teachers and nurses, both subject to bureaucratic work conditions, seniority is fundamental. Women in category III, who do office or sales work, however, normally lack any job security or promotion expectations, and they were also stable, although less so than others. Women in strata I, II and III have more to lose and less to gain by moving out of these occupations than do women in the manual occupations.

[Table 3 goes in here]

At the lower end of the structure women are less stable than men. Both the manual worker (IV) and self employed (V) strata show greater outflows than men did during the same period. Basically, these two strata show considerable exchanges between them, constituting a "circuit" of labor. Stability in the upper strata and the evidence that women in the lower two strata seem to move between them indicate little "long distance" vertical occupational mobility[23], both ascending and descending. Women in stratum I stood very little chance of falling to other strata, but women from lower strata also had very few possibilities of ascending to stratum I.

Another difference between the sexes is that self-employed women were not increasing via occupational mobility during the boom. In other respects men's and women's mobility tables for the boom are similar. The professional stratum expands through occupational mobility, while dependent manual work contracts. This suggests not so much structural change, as a life cycle effect: stratum I is a destination for most men and women, one that they usually enter after performing other occupations, while dependent manual work is a port of entry to employment in general.

Table IV is the one most subject to distortion due to the different lifetime employment cycles of women in different strata, that is to say its marginal distributions are distant from the occupational structures of 1975 and 1982. Women in higher strata tend to stay in employment throughout their career, while women in lower strata tend to intermittence in their careers, more so during the boom than afterwards. Keep in mind that we are analyzing mobility, not the structure itself.

[Table 4 goes in here]

During the crisis there is a very large amount of downward mobility from stratum I, into technical stratum (II) and self-employment (V). Stratum II is still very stable. Stratum III (office workers), though, expels individuals to self-employment and some upwards to I and II. The interaction between IV and V remains high. In fact, the only significant outflow from V is towards IV, while IV (dependent manual work) interacts with the upper categories a little more (it sends 14.6 percent of its 1982 members upwards to II and III but none to I).

There are uneven changes in the nature and extent of mobility among women during the crisis. The professional-managerial category changes very much, while category II (teachers and nurses) remains unchanged and is by far the most stable. Strata III and IV become less stable (especially the latter), and category V, which had been the least stable during the boom, becomes the second most stable during the crisis. An Inflow table would show, however, that it was very unstable since many workers transited to it from other categories.

5. - Changes in Gender Inequality

For men as well as women, the crisis brings much more mobility and a greater risk of downward mobility. This "greater risk" is partly the consequence of general structural change, since opportunities in the corporate and government sectors close, but it is also relative to each sex's particular opportunity structure. For example, while the top stratum among women expands from 14 to 20 workers during the boom, it stagnates at 35 during the crisis. Although in some respects men's and women's opportunity structures change in the same direction, they differ at the top: among men, stratum I continues to expand via occupational mobility during the crisis, from 225 to 255. In other words, differing chances depended on the evolution of men's and women's separate opportunity structures. The following table summarizes the change in men's and women's odds of upward versus downward mobility separately. The "odds" column for each period measures the likelihood of upward mobility with 1.0 meaning "even".

[Table 5 goes in here]

Table 5 is a simple summary of the changes in men's and women's opportunities for upward mobility to the right of the diagonal in the basic tables as "downward" and all that to the left as "upward". Correct in a broad sense (assuming that self-employment can in fact be represented as the lowest stratum), this reveals that women's odds, although better than men's throughout, worsened much more from the boom to the crisis. Also, while during the boom, men's and women's employment structures tended to equalize, this tendency was much reduced during the crisis. The finding that women's odds of upward/downward mobility were far superior to men during the boom is at first surprising, but it must be remembered that, in the context of a structure which was quickly expanding at the top, a "tight" labor market and previous marked inequality, women were much more likely to improve their situation than men (just like, as in the study of Guadalajara's labor market in 1982, they were much more likely to improve their income than men).

But the separate components of mobility behave differently. They are examined next as disparity rations between the sexes, to measure the extent to which the gap in men's and women's opportunities for upward, downward, "outward" and "inward" mobility changed from the boom to the crisis. Outward and inward refer to the transit to self-employment ("out" of dependent employment and back).

The analysis is based on disparity rations, that is, the ratio of men's probability of a particular change to women's probability of the same change. So, for example, if men in category A stand a .2 probability of mobility to B, and women .1, the ratio is 2:1 or 2.

So far our occupational classification revealed changes from the boom to the crisis. The small number of cases on cells away from the diagonal results from the limited time periods under consideration (which concentrate individuals on the diagonal) and because few women worked without interruption during the boom. To reduce distortions, these demand some aggregation. In this final section categories II and III are aggregated into one. A new terminology is also adopted. As in Goldthorpe et al. (1987), stratum I is called S, for service stratum, strata II and III called I for intermediate nonmanual. W and SE represent dependent manual workers and the separate category we posited for self-employment, respectively.

[Table 6 goes in here]

In general, we expect men to show more upward mobility than women (numbers greater than one) and less frequent downward mobility (ratios less than one). But we cannot assume differences between the boom and the crisis: weaker groups could become attractive to employers, displacing previously privileged groups and narrowing inequality. The worsening economy could hit weaker groups harder, reducing their negotiating power. This second possibility corresponds better to the observed outcome. Table 5 shows that women's odds for upward/downward mobility dropped far more than men's. In general, then, restructuring did not give women more chances of promotion than men; it has narrowed their previous advantage (better odds of advancement than men). Therefore, women were less likely to achieve occupational equality, and further they usually advanced only to lower nonmanual levels. To appreciate the meaning of men's and women's odds for upward/downward mobility it is convenient to disaggregate mobility by strata.

In general, upward mobility conforms to expectations assuming gender inequality and a widening of the gap with the crisis. Long range mobility from dependent manual work to the service stratum starts with some male advantage which accrues, as men's probability of this type of transition increases from 2.1% to 3.5% and women's decreases from 1.4% to 0%.

Short range mobility from I to S also conforms to expectations, but there is a slight decrease in the ratio, although men remain more than twice as likely to undergo this kind of mobility. The other kind of short range upward mobility, W - I, behaves contrary to expectations. During the boom, women were more likely to move up from dependent manual work to office and sales work than men, and they remained so during the crisis: only 0.7 men moved up for every woman who did. This finding is repeated in other studies elsewhere. Evidently, the separation between dependent manual work and dependent sales or office work is not as wide for women as it is for men. In all, however, the most aggregate ratio of upward mobility within the structure of dependent employment (I + W to S) is favorable to men and increases with the crisis.

Again, in the other direction findings generally conform to expectations. Though shifting slightly during the crisis, long distance downward mobility is greater for women than for men. Men's probability of falling increased more than women's, but men's advantage remains.

Short range downward mobility from S to I could not be compared during the boom because no women suffered it, while during the crisis proportionately more women did than men. In all, this is a significant increase in men's advantage over women. On the other hand, many more men than women went down from office or sales work to dependent manual work during the boom. (In fact, skilled dependent manual work was often much better paid for men than clerical work. During the crisis women were much more likely to suffer downward mobility, and their wages as dependent manual workers were usually lower than their male counterparts.)

Lastly, the ratio of aggregate downward mobility within dependent employment (S to W+I) changes. Men were more likely to drop than women during the boom to one and less likely to do so during the crisis. While the ratio of men's to women's aggregate probability for advancement only increases slightly from the boom to the crisis, the ratio of their aggregate probabilities for falling is reversed. In sum, the gender gap in terms of men's and women's probability for upward and downward mobility within dependent employment widened during the crisis, more so in downward than in upward terms.

While the above disparity ratios indicate the tendencies in gender inequality, they overlook changes in the relative chances of permanence in different strata, and they refer to only a part of the structure, that is dependent employment. The last table examines these two issues. The inequality hypothesis assumes that men enjoyed a better chance of permanence in the higher strata and a lower risk of permanence in the lowest strata and that the difference widened during the crisis. But the character of the move to and from self-employment is more problematic, mostly because a detailed analysis of the disaggregated occupations of men and women in self-employment shows large differences. For most women, transition to self-employment lowers their income, does not involve valuable skills and increases their working hours. This move can correctly be conceived as an involuntary move, whether for employment or family reasons. For men, on the other hand, the move to self-employment produces a smaller drop in their income levels and they tend to deploy more valued skills and to work shorter hours than self employed women. This qualitative difference is one reason why the move to self-employment is labeled a move "out", rather than a move "down".

The last table describes the ratio of men's to women's chances of getting to - or staying in - one rather than another stratum. The first line refers to their chances of staying in the service stratum rather than descending to other kinds of dependent employment. As can be seen, during the boom men had less of a chance of staying

than women did, but during the crisis the ratio is inverted. Though everybody's risk of falling increased, men stood a better chance of staying in the highest category.

[Table 7 goes in here]

In the lowest category of dependent employment, change is much smaller. During both periods men were unlikely to ascend to Intermediate or Service positions. This is due, as we saw before, to women's better chances of climbing from W to I, but not to S. The crisis only narrows the difference in this ratio.

The second group of odds ratios detail men's versus women's chances of mobility "out" of dependent employment and into self-employment. According to the first line, men stood a greater chance of moving to self-employment rather than descending during both periods. In the second line there is substantial change, from a situation in which men were less likely to move out rather than to other positions within dependent employment to another, in which they were more likely to do so. Lastly, movement between dependent manual positions and self-employment (rather than mobility up) also shows major change. Previously which men were less likely to move "out" instead of climbing, later they are more likely than women to do so.

The last group of odds ratios describes men's versus women's chances of mobility from self-employment into each one of the major positions within dependent employment rather than to the others. Men's advantage over women in their chances of moving from self-employment to a service position rather than to a lower one is large and it doubles with the crisis. But women remain more likely to move from self-employment to intermediate positions, with no appreciable change in this ratio from one period to the next. Lastly, women are also more likely to move from self-employment to dependent manual work rather than to higher positions, and this difference increases with the crisis. This last set of odds ratios shows no change in the gender patterns of interaction between self-employment and the major positions within dependent employment. There is a definite advantage in men's chances of long range ascending mobility from self-employment or via self-employment, while among women self-employment interacts more with the lower strata. This is so in both directions only during the crisis, since men from I and W stand a greater chance of moving to SE during the crisis.

6. - Discussion

The passage from the boom to the crisis affected men's and women's social mobility in different ways. For men, both downward and upward mobility increased. For women, downward mobility increased much more than upward mobility, and in particular, the highest positions ceased to grow via mobility, which meant that the structure of opportunity for men and women diverged at the higher levels. During the boom, women had a much better aggregate chance of upward mobility than men, which was leading to more occupational equality in gender terms. During the crisis, women still have a slight aggregate advantage, but it is restricted to movement from self-employment to dependent manual and lower nonmanual positions, and from dependent manual work to the lowest nonmanual positions (our category III, or lower intermediate).

The magnitude of the change in the patterns and chances of mobility is much greater for women than it is for men. In other words, women's patterns of mobility are much more context-dependent than men's. They lose more during the crisis. This is reflected in gender disparity and odds ratios, that is to say, in men's versus women's relative chances of different kinds of mobility, which are in many instances reversed, from a situation favorable to one unfavorable to them.

When only the structure of dependent employment is analyzed, that is, when self employment is left aside, men stand an overall better chance of upward mobility, and this advantage increases with the crisis. Disaggregated disparity ratios show that women's chances of upward mobility are more restricted and specific than men's, and that there is a greater change in gender-based ratios of downward mobility than in upward mobility.

Since women's risk of falling from the highest positions increases much more than men's, men gain an advantage in their ability to stay in the highest positions, which had previously belonged to women. The stratum from which women refuse to fall more than men is the intermediate nonmanual, where many women who work in education and health retained their previous positions, while men tended to drop or move out of dependent employment. In sum, the analysis showed that the passage of one period to the next did have an impact on gender inequality as evidenced in men's and women's patterns of mobility.

This analysis does not and could not, however, account for some of the major influences on changes in men's and women's mobility. It excludes two populations which, from the point of view of these processes of allocation of individuals, should be analyzed. One is individuals who quit employment, left the city or died, and the other is individuals who enter the occupational structure during either period. The occupational constrains on men and women are known to be different. The bulk of the pressure to join the labor force has fallen on women, who have counteracted failing male wages through an increase in their participation in the labor force. In the absence of expanding opportunities among the middle and higher strata, they have joined the lowest strata, and worsened the chances of upward mobility of other women. In other analyses I have shown that the difference in the comparative distributions of experienced versus inexperienced men and women increased from 1982 to 1990, with the two highest strata showing a remarkable increase in their impermeability to newcomers. This, in spite of the fact that, as said before, young, professional women replace older women with less schooling in the highest stratum.

The other major influence is international migration which as we discussed ion the first sections, increased notably during the eighties. It probably accounts for significant gender differences in mobility. Up to at least 1988, many more men migrated to the US. than women.[24] This should have two effects. One is to produce an "artificial" openness of the structure for men, due to other men leaving and the other is to force women who cannot rely on migrants' remittances into whatever employment can be found, which increases competition among them and could be responsible -in addition to dropping wages- for an increase in the numbers of women in the lowest tiers.

Finally the interaction of the life cycle with social mobility can only be partially assessed during these two short periods, as in the breakdowns of self employment by age. Partly this is the consequence of the periods themselves, and partly of the dwindling number of cases as analysis retreats in time. Since the life cycle has been shows to influence men's and women's trajectories, particularly in Guadalajara (Escobar, 1984), to provide a comprehensive review of the processes of change in the patterns of mobility of individuals within the structure of employment it would have to be taken into account more fully, through repeated surveys and larger samples. Hopefully, this analysis will be carried out in this city and the rest of Mexico in the near future.

BIBLIOGRAPHY

Alba Vega, Carlos

1986 "La industria de Guadalajara ante la crisis" en Revista Encuentro, No. 10 (Vol. III, No. 2):23-50

Blau, Peter M. and Otis D. Duncan

1978 [1967] The American Occupational Structure, New York: The Free Press.

Cordera Campos, Rolando and Enrique González Tiburcio

1991 "Crisis and Transition in the Mexican Economy" in González de la Rocha and Escobar Latapí (Eds.) Social Responses to Mexico's Economic Crisis of the 1980's La Jolla: Center for U.S.-Mexican Studies, U.C., San Diego.

Cortés, Fernando and Rosa María Rubalcava

1995 "Cambio estructural y concentración: un analisis de la distribución del ingreso en México, 1984-1989." (this volume).

De la Peña, Guillermo

1986 "Lo público y lo privado: el grupo doméstico frente al mercado de trabajo urbano" in de la Peña, Guillermo and Agustín Escobár (eds.) Cambio regional, mercado de trabajo y vida obrera en Jalisco. Guadalajara: Colegio de Jalisco.

Erikson, Robert

1981 "Social Class of Men, Women and Families". Quoted by Goldthorpe et al, 1987.

Escobar Latapí, Agustin

1984 Dependent Industrialization and Labor Market Structure Ph.D. Thesis, University of Manchester.

1988 "The Rise and Fall of an Urban Labor Market: Economic Crisis and the Fate of Small Workshops in Guadalajara, Mexico" Bulletin of Latin American Research, 7 (2):183-206.

1990 "Las grandes organizaciones de servicios y la región" paper delivered at Primer Colquio de Occidentalistas, Guadalajara.

1993 "Autoempleo e informalidad en Guadalajara: 1975-1982-1990" Mimeo, CIESAS Occidente.

Escobar, Agustín and Bryan Roberts

1991 "Urban Stratification, the Middle Classes and Economic Change in Mexico" in González de la Rocha and Escobar Latapí 1991.

Featherman, David L., and Robert M. Hauser

1978 Opportunity and Change, New York: Academic Press.

García, Brígida

1988 Desarrollo económico y absorción de fuerza de trabajo en México: 1950-1980. Mexico: El Colegio de México.

García, Brígida and Orlandina de Oliveira

1990 "Recesión económica y cambio en los determinantes del trabajo femenino". Mimeo. Mexico. El Colegio de México.

1991 "Motherhood, Work and Woman's Condition in Mexico", informe final presentado a la Fundación Rockefeller, Centro de Estudios Sociológicos, El Colegio de Mexico, March.

1994 Trabajo femenino y vida familiar en México. Mexico. El Colegio de México.

Glass, D.V. (ed.)

1954 Social Mobility in Britain. London: Routledge and Kegan Paul.

Goldthorpe, John H.

1987 Social Mobility and Class Structure in Modern Britain. Second Edition. Oxford: Oxford University Press.

González de la Rocha, Mercedes

1986 Los recursos de la pobreza: Familias de bajos ingresos in Guadalajara. Guadalajara: El Colegio de Jalisco/CIESAS/SEP.

1988 "Economic Crisis, Domestic Reorganization and Women's Work in Guadalajara, Mexico" Bulletin of Latin American Research, 7 (2): 207-223.

González de la Rocha, Mercedes and Agustín Escobar Latapí (eds.)

1991 Social Responses to Mexico's Economic Crisis La Jolla: Center for U.S. - Mexican Studies, University of California, San Diego.

Portocarero, Lucienne

1989 "Trends in Occupational Mobility: The Gender Gap in Sweden" in Acta Sociologica (32), 4:359-374.

Rendón, Teresa and Carlos Salas

1991 "El mercado de trabajo no agrícola en México: Tendencias y cambios recientes". Documento en el seminario Mercados de trabajo: Una perspectiva comparativa, tendencias generales y cambios recientes. Mexico. CES/COLEF y La Fundación Friedrich Ebert.

Roberts, Bryan

1990 "Employment Structure, Life Cycle and Life Chances: Formal and Informal Sectors in Guadalajara" in Portes, A., M. Castells and L. Benton (Eds).), The Informal Economy, Studies in Advanced and Less Advanced Countries. Baltimore: Johns Hopkins.

Rubin-Kurtzman, Jane

1992, "Female Employment, Demographic Change and Economic Deterioration: Mexico City, 1970-1976." Paper delivered at the XVII LASA Meeting, Los Angeles.

Table I

Men's Intragenerational Mobility: 1975 to 1982

(Ns and outflow percentages)

1975	1982
	I	II	III	IV	V	Total
I	116 86.6	1 0.7	5 3.7	6 4.5	6 4.5	134 8.4
II	9 12.3	53 72.6		6 8.2	5 6.9	73 4.7
III	18 9.0	3 1.5	131 65.8	29 14.6	17 8.5	198 12.5
IV	19 2.1	12 1.3	47 5.1	749 81.4	90 9.8	917 57.9
V	5 1.9	2 0.8	9 3.5	31 11.9	213 81.9	260 16.4
Total	167 10.5	71 4.5	192 12.2	821 51.9	331 20.9	1582 100

Source: CIESAS - INEGI 1990 survey.

Table 2

Men's Intragenerational Mobility, 1982 to 1990

1982	1990
	I	II	III	IV	V	Total
I	146 64.9	17 7.6	20 8.9	5 2.2	37 16.4	225 10.0
II	16 13.9	58 50.4	8 7.0	19 16.5	14 12.2	115 5.1
III	35 11.7	51 17.0	96 32.0	47 15.7	71 23.7	300 13.4
IV	41 3.5	24 2.0	102 8.6	781 66.1	234 19.8	1182 52.8
V	17 4.1	9 2.1	9 2.1	46 10.9	338 80.7	419 18.7
Total	255 11.4	159 7.1	235 10.5	898 40.1	694 30.9	2241 100

Source: CIESAS-INEGI 1990 survey.

Table 3

Women's Intragenerational Mobility 1975-1982

1975	1982
	I	II	III	IV	V	TOTAL
I	13 92.9			1 7.1		14 4.8
II	3 4.9	55 90.2	1 1.6		2 3.3	61 21.0
III	2 3.0	2 3.0	54 80.6	3 4.5	6 9.0	73 25.2
IV	1 1.4	1 1.4	6 8.2	53 72.6	12 16.4	67 23.1
V	1 1.3	2 2.7	4 5.3	14 18.7	54 72.0	75 25.9
TOTAL	20 6.9	60 20.7	65 22.4	71 24.5	74 25.5	290 100

Source: CIESAS-INEGI 1990 survey.

Table 4

Women's Intragenerational Mobility 1982-1990

1982	1990
	I	II	III	IV	V	TOTAL
I	18 51.4	6 17.1	4 11.4	1 2.9	6 17.1	35 6.5
II	4 4.0	91 90.1	3 3.0	1 1.0	2 2.0	101 18.8
III	12 7.4	14 8.6	106 65.4	13 8.0	17 10.5	162 30.2
IV		6 4.6	13 10.0	86 66.2	25 19.2	130 24.2
V	1 0.9	2 1.8	3 2.8	13 11.9	90 82.6	109 20.3
TOTAL	35 6.5	119 22.1	129 24.0	114 21.2	140 26.1	537 100

Source: CIESAS-INEGI 1990 survey.

Table 5

Summary of Mobility 1975-82 and 1982-90

	1975-82		1982-90
	TOTAL MOBILITY	ODDS: UP/DOWN	TOTAL MOBILITY	ODDS: UP/DOWN	02 - 01
MEN	22.3	.80	41.2	.58	-22
WOMEN	18.3	1.19	28.9	.56	-63

Source: Tables in previous sections.

Table 6

Gender Inequality: Disparity Ratios for Upward and Downward Mobility

During the Mexican Boom and Crisis

TYPE OF MOBILITY			BOOM	CRISIS
ASCENDING:			1.5a	n/ab
long range, (W-S)			2.54	2.03
short range, (I-S)			0.7	0.70
from I and W to S			1.19	1.41
DESCENDING
long range, (S-W)			0.63	0.79
short range, (S-I)			n/ac	0.58
short range, (I-W)			5.56	0.41
from S to I and W			1.25	0.60

Source: tables in previous section.

a: As explained, numbers greater than 1.0 indicate greater mobility odds for men as compared to women.

b: There are no women moving from dependent manual to service occupations during the crisis.

c: There are no women dropping from service to intermediate categories during the boom, but many do during the subsequent period.

Table 7

Gender Inequality: Selected Odds Ratios for Immobility, "Outward",

and "Inward" Mobility During the Mexican Boom and Crisis

(Odds>1.0 indicates men are favored)

			BOOM	CRISIS
ODDS RATIOS OF IMMOBILITY VERSUS MOBILITY WITHIN DEPENDENT EMPLOYMENT
A) S - S/S - I, W			0.74	2.11
B) W - W/W - I, S			1.45	1.04
ODDS RATIOS: MOBILITY "OUT, FROM DIFFERENT STRATA
A) S - SEa/S - I, W			.5/0	1.62
B) I - SE/I - S, W			0.35	1.11
C) W - SE/W - S, I			0.77	1.06
ODDS RATIOS: MOBILITY "INTO", FROM DIFFERENT STRATA
A) SE - S/SE - I, W			2.75	5.4
B) SE - I/SE - S, W			0.78	0.80
C) SE - W/SE - S, I			0.95	0.61

Source: tables in previous section.

a: SE = "self-employment"

[1]This chapter is based on a survey commissioned by the author to INEGI (The National Statistics and Geography Institute) and carried out during the third quarter of 1990. The project was possible thanks largely to a donation from the Ford Foundation. The first version of this chapter was written while the author was a Fulbright visiting scholar at the University of Texas at Austin. The editors of this volume, Humberto Muñoz, Bryan Roberts and Patricia Chalita all made helpful comments on an earlier draft. The usual disclaimers apply, however.

[2]For a detailed account of the Mexican economy up to 1987, see the Introduction to this volume and Cordera Campos and González Tiburcio (1991).

[3] This simplifies developments during the period of import substitution industrialization. Modernization of the occupational structure occurred rapidly from some time in the 1930's until the mid seventies, when self and family employment seem to have started to rise after a long period of decline. Also, there were several minor economic swings from 1934 to 1982. Some analysts of female employment have termed the Echeverría sexennium (1970-1976) one of economic deterioration (Rubin-Kurtzman 1992). This is not exact. Although per capita growth of GNP was slightly lower than during the preceding period, it was positive, from 9,200 constant pesos in 1970 to 10,258 in 1976, for per capita economic growth of 11.35% during that sexennium. Growth during the subsequent boom (1976-1982) was spectacular by present standards, averaging 6.5% annually, and the crisis (1980-87) was characterized by a drop in per capita GNP of 17%.

[4] Most of the data on the informant's 1990 job were derived from the government's schedule and coded onto a special sheet added to the author's schedule. Some questions were however asked about the worker's present job, in addition to those in the official survey.

[5]This is generally recognized in the literature, and has been the main reason for not performing parallel analyses of men's and women's occupational mobility.

[6]Women who head households are allocated to their own occupational status.

[7] I will return to this point later.

[8] Mexico's leading labor unions were at the time considered the strong hold of a politically powerful union elite, and one of the bases of the stability of the Mexican political system.

[9]In fact, the growth of manual wage workers during this period is due mostly to the rise of those employed in personal and repair services, not in industry. So, it is a rise both in service functions and in service activities that accounts for the change in the structure from 1940 to 1980. This rise is however just as important in the modern services as it is in others.

[10] The first research which showed systematically how this happened was undertaken by González de la Rocha in Guadalajara (1986).

[11] According to ECLAC (1989), informal work increased 80% from 1980 to 1987 in absolute terms, and accounted for 33% of the work force in 1987.

[12] This statement is based on my own estimate and on García and Oliveira (1990), rather than on the censuses, which indicate relative growth of managers and professionals among working women.

[13] Approximately a quarter of the contributions made by employers and workers to this institute fall under "guarderías" (Child care centers). The fact is, however, that these centers could not service more than 5% of potential demand. Anthropologists have stressed the help of neighbors and non household kin in this regard. Their role, however, seems not to be too important. In our 1990 survey of 3,000 Guadalajara households, less than 8% of the households with children under seven resorted to non resident kin or neighbors for child care. Work does not relieve the housewife of child care (which often means the children are left alone), and when it does, the most frequent "helper" is a young woman residing in the household, the housewife's younger sister or the children's older sister. Murphy et al. (this volume) stress the same point for Oaxaca, Oax.

[14] It is likely that some of these agricultural workers were insufficiently specified in 1980. Also, the data for 1970 result from adding the partial results for the three municipios which comprised the metropolitan area at the time, while the 1980 results are for the city itself (the locality). This naturally excludes rural areas in the metropolitan municipios.

[15] Maquiladoras, or in-bond plants, were the subject of special legislation and the most dynamic agents of growth of modern employment during the crisis

[16] Rendón and Salas do not specify whether the rates are annual or for the two-year period. The material used for the present chapter, as well as some other previous findings, suggest that the rates are lower although still sufficient to indicate a change in employment patterns.

[17] That portion of the federal budget which can be allocated to new infrastructure and special programs, and which virtually disappeared during 1983-1987 (Cordera and González Tiburcio, 1991).

[18] Data supplied by González de la Rocha, along with several other studies agree. The Ministry of Foreign Relations, for example, estimates that Jalisco lost 1.5 million emigrants or 30% of its 1990 population, between 1980 and 1990 (El Occidental, page 1a, October 20, 1992).

[19] The alternative is to suppose that migrants interviewed in 1982 have dropped out of the labor force or died. Everything indicates that fewer people dropped out of the labor force during the eighties.

[20] I will compare the percent distributions only. In absolute terms women expand extremely rapidly in all categories of employment. In comparing the two censuses I only refer to identical categories, which may nevertheless differ due to changes in other categories and in the rules dictating inclusion in "insufficiently specified". Addition of women in professional and managerial categories still provide a lower percentage of women in this stratum than the figure I provide based on the CIESAS - INEGI survey for 1990. The only apparent difference between the two is the inclusion in my category of those women in positions of responsibility over others within the school system, which in the census are grouped with rank and file teachers under the heading "workers in education".

[21] Category one comprises those individuals performing a profession typically performed by university degree professionals, even if they did not actually hold a degree, such as practitioners of the law who do not yet hold a degree.

[22] A separate analysis shows that, for urban Mexico as a whole, legality of employment is a factor which increases income, while self employment in general brings better income than dependent employment. Also, moving to self employment is not identical to having a long trajectory in self employment: those moving recently to self employment were not doing as well as those who had been self employed for a long time.

[23] Long distance occupational mobility involves "jumping" intermediate categories: from self employment and dependent manual work to stratum I and vice versa. This is a commonly used notion in analyses of occupational or class mobility.

[24] After that date, as more Mexican migrants entered the process of legalization started under IRCA, an increase in family migration has taken place.