Navegación – Mapa del sitio

Social mobility in Mexico. Trends, Recent Findings and Research Challenges

Patricio Solís Gutiérrez
p. 7-20

Resúmenes

La recesión de los años ochenta y la subsecuente reestructuración económica tuvieron un profundo impacto en la sociedad mexicana. No obstante, sus consecuencias sobre la movilidad social no fueron analizadas hasta finales de los noventa, cuando una serie de estudios empíricos revelaron las continuidades y los cambios en los patrones de movilidad social. En este artículo se discuten cuatro tendencias recientes: la continuidad de las altas tasas absolutas de movilidad intergeneracional; la reducción de las recompensas monetarias asociadas a la movilidad ocupacional; la creciente rigidez en las tasas relativas de movilidad; y el ajuste del caso mexicano al patrón de movilidad general propuesto por Erikson y Goldthorpe. El panorama que resulta de estas tendencias es el de una sociedad que, a pesar los efectos negativos de la crisis y los cambios estructurales de los años ochenta y noventa, ha mantenido altas tasas de movilidad social, pero sufre en otros aspectos como la calidad de las oportunidades de movilidad ascendente y la creciente desigualdad de oportunidades asociada a los orígenes de clase. El artículo concluye con una discusión sobre posibles líneas futuras de investigación de los estudios sobre movilidad social en México.

Inicio de página

Texto integral

1During the 1960s and 1970s three major research projects addressed social stratification and mobility patterns in urban Mexico. These studies were the “Monterrey Geographic and Social Mobility Project”, developed by Balán, Browning and Jelin (1973), and two studies carried out in Mexico City, the first coordinated by Muñoz, Oliveira and Stern (1977), and the second by González Casanova (Contreras 1978). There are several features common to these three studies that reflect the main topics of sociological inquiry in Latin America at that time. First, they intended to identify the effects of rapid industrialization on the occupational structure and the impact of these transformations on occupational opportunities. Second, they explored the mechanisms of assimilation of rural migrants into urban labor markets, as well as the links between rural‑urban migration and the so called “urban marginality”. Third, they shared an interest on the effects of social stratification on the universe of values and attitudes of individuals, whether on the sphere of family values or in the realm of political attitudes and positions. Obviously, there was also an interest in studying social mobility, both for city natives and rural migrants. Finally, in the Monterrey project there was a remarkable effort to make a comparative analysis of the status attainment process in Monterrey and the Unites States, within the framework proposed by Blau and Duncan (1967).

  • 1 Even when social mobility can be studied across several dimensions (i.e. education, income, status, (...)

2Based on these studies we can identify three major features of intergenerational social mobility1 during the industrialization by import-substitution (ISI) period. First, absolute upward mobility rates were remarkably high. High mobility rates were explained by a combination of rural-urban migration, which implies by definition social mobility, and rapid industrialization, which provided opportunities of mobility into unskilled and skilled manual positions. At that time, the distinction between “structural” and “relative” mobility that shapes current debates on the field was not as relevant, so there was no explicit attempt to separate these two forms of mobility. However, the prevailing interpretation focused on the role of urbanization and industrialization as the main engine for upward social mobility, thus emphasizing the importance of structural mobility.

3The second feature is that chances of upward mobility in urban areas were available both for city natives and rural migrants. Upward mobility rates were similar for individuals raised in the cities and those who came from rural communities. This finding contributed to refine the concept of urban marginality, which became simultaneously a structural phenomenon and a condition that might be temporary across the life course of rural migrants.

4The third feature is a distinctive profile in the status attainment process. This profile was characterized by four elements. First, social origins, measured through father’s and mother’s education and father’s occupation, had a strong influence on educational attainment. This influence was even higher than the observed in the United States during that period. Second, education was the most important determinant of occupational attainment. Third, social origins had only a marginal direct influence on occupational attainment, thus suggesting that their influence was mainly channeled through education. Fourth, there was a trend towards the reduction of the effect of social origins on education.

5In sum, the landscape of social stratification and mobility during the ISI period was one of high rates of mobility, where inequality of opportunity was evident −particularly regarding educational attainment−, but was somewhat blurred by plentiful opportunities of upward mobility that were available for individuals coming for different social backgrounds and migratory origins.

6An important question addressed after the economic recession of the 1980s and structural transformations of the 1990s was to what extent these changes altered patterns of social mobility in Mexico (Escobar Latapí 2001). However, at first it was not easy to answer to this question because during the 1980s and the first half of the 1990s there was not a follow-up of the pioneering social mobility studies of the 1960s and 1970s. It is interesting that studies on social mobility were disregarded precisely during a period of structural turmoil and increasing social polarization (Filgueira 2001). The absence of studies may be attributed to several reasons. One of them is the increasing dominance of the “historical-structural” paradigm in Latin American social sciences, which despised anglosaxon inspired stratification studies and downplayed survey research −and also modern statistical techniques− as useful tools for social research. Other possible reason is the displacement of research interests to other issues, such as the study of household survival strategies, women’s labor force participation, or the measurement of poverty.

7In any case, by the end of the 1990s it was not clear whether the characteristics of social mobility identified in the pioneering studies were still present or, as many scholars suggested but could not demonstrate due to the absence of empirical evidence, had suffered significant changes after the intense economic, social and political transformations of the 1980s and 1990s.

  • 2 See, among others: Behrman, Gaviria and Székely (2001), Solís (2002, 2005, 2007), the series of art (...)

8A series of new studies published since then have helped to shed some light on these issues.2 The purpose of this article is twofold. On one hand, it summarizes the most significant findings of research carried out by myself (on my own or along with other colleagues) on this field. On the other, it reevaluates these findings with additional evidence, thus validating whether they can be replicated with different data sets and can be generalized to the national context. Finally, the paper concludes with a discussion of the main challenges for research on social mobility in Mexico.

Continuities and Changes in Social Mobility during the 1980s and 1990s

9In this section I discuss what I consider to be the four most important findings regarding recent patterns of social mobility in Mexico. I focus on intergenerational occupational mobility, although some of these findings may extend to other forms of mobility, such as educational mobility. I present these findings as stylized facts, that is, generalizations that may require further examination. The presentation is supported with selected empirical evidence.

The continuity of high upward mobility rates

10The first finding is that, despite the negative social effects of the crisis and economic restructuring, absolute upward mobility rates remained high, with levels similar to those observed during the 1960s and 1970s. Intergenerational absolute mobility rates reported in several studies are reproduced in Table 1. High mobility predominates independently of the geographical context (Monterrey, urban areas, or national), the classification scheme (4, 6 and 7 classes) or even the strategy used to distinguish age and cohort effects (occupation at age 30 or the more common cross-sectional approach to mobility tables).

Males

Females

Up

Down

Total

Up

Down

Total

Urban areas (>15,000), 1998 /1/

55

13

68

-

-

-

Monterrey, 2000 /2/

52

8

60

-

-

-

National, 2005 /3/

44

15

59

62

18

80

National, 2011 /4/

49

18

67

53

19

72

Table 1 - Intergenerational Social Mobility Rates in Mexico According to Recent Studies.

/1/ Mobility between father’s and respondent’s class at age 30, birth cohort 1966-1968, four-class scheme (Zenteno and Solís 2006).

/2/ Mobility between father’s and respondent’s class at age 33, birth cohort 1955-1967, four-class scheme (Solís 2007).

/3/ Mobility between father’s and respondent’s current occupation, respondents between ages 30 and 59 in 2005, six-class scheme (Solís and Cortés 2009).

/4/ Mobility between father’s and respondent’s current occupation, respondents between ages 30 and 59 in 2011, Modified Erikson and Goldthorpe (1987) seven-class scheme. Solís (forthcoming).

  • 3 See Torche and Wormald (2004) for Chile and Valle Silva (2004) for Brazil.

11Depending on the study, estimated mobility rates for males fluctuate between 59% and 67%. Most of this mobility was in an upward direction (between 44% and 55%). In order to establish whether these rates are high or low it is necessary to perform international comparisons with a comparable class scheme. The most widely used classification is the so-called “CASMIN” scheme, proposed by Erikson and Goldthorpe (1987). The last row in Table 1 presents absolute mobility rates estimated after the application of a slightly modified version of this classification. The total mobility rate for males under this scheme amounts to 67%. This value is similar to the observed in European countries (Breen and Luijkx 2004) and also to that reported in other Latin American countries, such as Chile (74%) and Brazil (70%).3

  • 4 Mobility rates for woman are calculated from social mobility tables where origins are measured usin (...)

12Mobility rates are considerable higher for women, partly because they reflect not only changes in the occupational structure and other factors typically affecting men’s mobility rates, but also the effects of occupational segregation by gender.4 An analysis of sex differences in destinations (not shown) suggests that higher mobility rates among women are produced by the “deficit” of cases in destinations typically dominated by men (jobs in agriculture or skilled manual positions in manufacturing, for example), and the “excess” of cases in “feminized” occupations, such as clerical positions and unskilled service positions.

13Although the finding of persistent high mobility rates is somewhat surprising, it can be explained by three contextual determinants of social mobility: a) the direction and timing of long term changes in the occupational structure; b) demographic trends; c) the fact that observed absolute mobility rates are not only dependent on overall changes in the class structure, but also on the degree of circulation between classes net of these changes. Before moving on to the next section I will briefly discuss these contextual determinants.

  • 5 The micro-data were provided by the IPUMS-international Project of the University of Minnesota (Min (...)

14Long term changes in the occupational structure can be analyzed in Table 2, which presents the distribution of the working population by occupational groups in 1960, 1970, 1990, and 2000.5 In 1960, agricultural occupations were still dominant (50.5%), followed by manual occupations in manufacturing activities (21.2% if we include in this group craft and related trades workers, plant and machine operators, and workers in elementary occupations in manufacturing). Top-level positions (Legislators, Senior Officials, Managers, Professionals, and Technicians) accounted for only a minor fraction of the working population (6.1%).

Occupational position

1960

1970

1990

2000

Legislators, Senior Officials and Managers

0.9

2.7

2.6

2.1

Professionals

3.9

4.5

7.9

8.8

Technicians and Associate Professionals

1.4

2.1

3.7

3.4

Clerks

6.1

8.5

10.3

9.5

Service Workers and Shop and Market Sales

10.0

11.2

13.3

17.8

Agricultural and Fishery Workers

50.5

39.0

20.9

14.9

Crafts and Related Trades Workers

15.0

18.3

18.6

18.5

Plant and Machine Operators and Assemblers

5.5

7.4

10.8

10.8

Elementary Occupations in Manufacturing

1.2

0.6

4.2

4.3

Elementary Occupations in Services

4.8

5.4

7.5

9.7

Armed Forces

0.7

0.4

0.2

0.2

Total

100.0

100.0

100.0

100.0

Table 2 - Changes in the Occupational Structure (population 15-64), Mexico, 1960-2000.

Source: Own calculations based on the micro-data sample files of the 1960, 1970, 1990 and 2000 population censuses (Minnesota Population Center 2011).

15The changes observed during the 1960s reveal the fast transformation of the occupational structure characteristic of the ISI period. By 1970, agricultural activities reduced their share to 39.0%, whereas manual positions in manufacturing increased to 26.4%. Top­level positions showed also some increase (9.2%), and the remaining change spread among middle­ and low‑level positions, both manual and non­manual.

16These transformations continued in the 1970s and 1980s, but with occupations linked to services having the highest expansion rates. Service-oriented positions expanded not only at the top and the middle (clerks, sales workers), but also at the bottom of the occupational hierarchy (elementary occupations in services). Finally, in the 1990s the pace of change decreased: occupations in agricultural activities kept declining (14.9% in 2000); manual jobs in manufacturing maintained their share in relation to 1990 (with about one third of the working population), and the only significant expansion was among middle-level and “unskilled service positions”: service workers and shop and market sales (17.8%), and elementary occupations in services (9.7%).

17In sum, in the long term the occupational structure of Mexico experienced significant transformations, even when the pace of change reduced after 1990. This is important because intergenerational mobility rates are more susceptible to long-term than to short-term changes in the occupational structure, and therefore the persistent high mobility rates are partly explained by such long-term changes.

18There are also demographic factors that may have facilitated social mobility. Two trends are particularly important. First, the onset of the fertility decline in the 1970s produced an increasing gap in fertility levels across social classes (with lower fertility rates among the upper class), probably facilitating a redistribution of opportunities towards the bottom of the occupational structure. Second, higher migration rates, domestic and to the United States, may have also produced an increase in intergenerational mobility, not only through the internal redistribution of the population into regions with higher opportunities (a typical result of rural-urban and interregional migration), but also by selecting out international migrants that otherwise would have competed for opportunities of upward mobility within Mexico.

19Finally, high absolute mobility rates have also persisted because social mobility is explained not only by macro changes in the occupational structure but also by circulation mobility that takes place beyond structural change. For example, in the 2011 intergenerational mobility table, the dissimilarity index between the marginal class distributions of fathers and sons was only 20%. This percentage represents the intergenerational mobility that is “forced” by intergenerational shifts in the occupational structure. If these shifts were the only force driving social mobility, the expected absolute mobility rate would also be 20%. However, the observed mobility rate is much higher (67%), thus suggesting that, in addition to the transformation of the occupational structure, other allocation processes at the meso- and micro-level are also producing social mobility.

Reduction of income gains associated to upward occupational mobility

20One of the main underlying assumptions of the sociological approach to social mobility is that the position of individuals in modern society can be derived from their position in the labor market (Grusky and Kanbur 2006). According to this assumption, occupations are the institutionalized bridges that link individuals to valuable resources and assets. These resources come in “reward packages” of different sort that include not only monetary assets, but also other kind of assets such as power, prestige, etc. Based on this assumption, sociological research tends to equalize occupational mobility with social mobility. If an individual experiences occupational mobility, then his/her access to valuable assets will significantly increase or decrease and therefore he/she will experience upward or downward social mobility.

21However, the assumption that links social mobility to occupational mobility may become problematic if the association between occupational position and rewards changes or weakens over time. Some of this may have happened in Mexico as a result of the recession and economic restructuring in the 1980s and 1990s. Based on survey data on wages by occupation in 1965 and 2000, Solís (2007) reported an overall reduction in real wages in Monterrey during this period, a finding that is not surprising given the negative effects of the crisis and economic restructuring. However, drops were higher for nonmanual occupations, with a consequent reduction in the wage gap between manual and lower nonmanual positions. This implies that the economic advantages associated to upward mobility from manual to nonmanual occupations were reduced.

  • 6 National employment surveys were implemented in the 1970s, but the micro-data necessary to estimate (...)

22A question is whether these conclusions can be generalized to the national context. Analyzing long term changes in wages by occupation at the national level is difficult because there are not comparable surveys for the years before 1992.6 However, the micro-data samples of the 1960, 2000 and 2010 population censuses can help us to explore these changes. Figure 1 displays the average monthly wages by occupation for males. Values are adjusted for inflation so that they reflect the equivalent value in pesos for 2010. Between 1960 and 2000 the wage gap between top-level occupations and the rest increased. In addition, real average wages for professionals and lower non-manual positions such as office clerks and sales workers decreased. The wages for skilled manual occupations (plant and machine operators and craft workers) and low-level positions (elementary and agricultural/fishery occupations) remained basically unchanged. This situation accentuated between 2000 and 2010 (although mean wages for top-level occupations decreased significantly, perhaps due to the financial crisis experienced by the country in 2008-2009).

Figure 1 - Monthly Average Wages by Occupation in Mexico. Males, 1960, 2000 and 2010*.

* Average wages for the working population aged 20-64. Values adjusted for inflation to the equivalent in pesos in the year 2010.

Source: Own calculations based on the micro-data sample files of the 1960, 2000 and 2010 censuses (Minnesota Population Center 2011).

23These results are indicative of a shift in stratification of wages and also in the expected economic rewards associated to upward mobility. During the ISI period, upward mobility from manual to lower nonmanual positions was associated to significant gains in wages. After the structural changes of the 1980s and 1990s these gains were more uncertain. Only long distance mobility to highly skilled technical and professional positions, or top-level occupations in the public or private sector, guaranteed upward mobility in wages.

24This reduction in the financial retributions of upward occupational mobility from higher manual to lower non-manual occupations is a pattern that might be generalized to other Latin American countries. According to Kessler and Espinosa (2003), who studied the Argentinean case, the declining social rewards for the middle class during the 1980s and 1990s derived in a situation where occupational mobility into destinations typically associated to this sector was no longer followed by mobility in social rewards, thus producing what they called “spurious mobility”.

25Going back to the Mexican case, “spurious mobility” may have significantly increased in the 1980s and 1990s. However, it would be an overstatement to argue that occupational hierarchies are no longer important for social stratification. Indeed, when instead of income we use other measures of standard of living (such as the availability of household appliances), important differences among occupational groups emerge (Table 3). In addition, there are important distinctions in tastes and cultural consumption patterns among members of the different occupational groups (Solís 2002, 2005). A proper evaluation of the social distance among occupational groups must consider other dimensions of stratification, such as hierarchies in cultural capital, social capital and social status.

Auto

Hot water

Computer

Refrigerator

VCR

Washing machine

Legislators, senior officials and managers

81

89

51

98

84

88

Professionals

66

79

37

94

74

80

Technicians and associate professionals

52

68

21

92

65

77

Office clerks

49

70

20

92

65

77

Service workers and shop and market sales

35

45

4

81

44

63

Plant and machine operators and assemblers

43

55

11

83

52

67

Crafts and related trades workers

31

40

6

72

40

57

Elementary occupations

22

32

5

65

33

49

Skilled agricultural and fishery workers

18

12

1

34

13

21

Table 3 - Availability of Household Appliances by Occupation in 2000, Mexico (%)*.

Auto - % of individuals with automobile in their household.

Hot water - % of individuals with hot water in their household.

Computer - % of individuals with computer in their household.

Refrigerator - % of individauls with refrigerator in their household.

VCR - % of individuals with Video Cassete Recorder in their household.

Washing machine - % of individuals with washing machine in their household.

* Working population between ages 20 and 59.

Source: Own calculations based on the micro-data sample files of the 2000 Census (Minnesota Popula

Increasing rigidity in class mobility

26As mentioned before, absolute social mobility rates have remained high in Mexico. However, contemporary sociological evaluations about social mobility and social justice are usually based not on absolute mobility rates but in “relative” mobility, that is, in “how strongly parental social class predicts the social class in which the child will be located when he or she is an adult” (Breen 2010). The emphasis on relative mobility stems from its relationship with inequality of opportunity: a perfectly “fluid” society, that is, a society where parental social class is not a determinant of individuals’ class destination, would represent an extreme example of equality of opportunity. On the contrary, a perfectly “rigid” society (where there is a one-to-one association between origins and destinations) would represent the extreme case of inequality of opportunity. Evidently, perfect fluidity and rigidity are only “ideal types” and no known society adjusts to these patterns. Therefore the question is not whether a society is “fluid” or “rigid” but how much fluidity exists and how it changes over time.

27Few studies in Mexico have empirically evaluated changes over time in social fluidity (Cortés and Escobar Latapí 2005; Solís 2002, 2007; Zenteno and Solís 2006). The study of Zenteno and Solís reviews the association between parental social class and the destination of males belonging to birth cohorts born between 1925 and 1968. The study is limited to the city of Monterrey and national urban areas. Ordered logistic regression models are adjusted where the dependent variable is the respondent’s social class at age 30 and the independent variables are father’s social class, the size of the community of origin, and years of schooling. Results are presented in Table 4.

National

Monterrey

1936-1938

1951-1953

1966-1968

1925-1932

1948-1956

1963-1967

Father’s Occupation

Non-manual (reference)

-

-

-

-

-

-

Higher manual

0.77

0.68

0.48*

0.99

0.52*

0.29**

Lower manual

0.81

1.53

0.22**

0.67

0.38*

0.25**

Community of origin

Rural (reference)

-

-

-

-

-

-

Urban

1.97

0.98

1.50**

0.99

0.97

1.10

Education (years)

1.39**

1.39**

1.50**

1.58

1.47**

1.48**

Pseudo R squared (McFadden)

0.19

0.17

0.24

0.27

0.26

0.21

n

146

180

171

233

336

177

Table 4 - Effects of Social Origins and Educational Attainment on Occupational Attainment (around age 30). Odds Ratios from Ordered Logit Models. National Urban Context (15,000+) and Monterrey, by Birth Cohort*.

Source: Zenteno and Solís (2006).

28Results are similar for Monterrey and the overall context of national urban areas. There is no significant effect for father’s social class in the models for the oldest cohort (1936-1938 in the national context and 1925-1932 in Monterrey). This indicates that during the ISI period (these birth cohorts reached age 30 during the 1950s and 1960s), social origins had little or no effect in class attainment, once the indirect effect through education was controlled in the model.

29In contrast, in the youngest cohorts (1966-1988 for the national urban context and 1963‑1967 for Monterrey), the magnitude of the coefficients associated to social class increases and they become statistically significant. Thus, for example, at the national level a son of a skilled manual worker had odds 52% lower of attaining a higher social class than a son of a non-manual worker. In the case of unskilled manual workers the odds were four times lower. These disadvantages are additional to those that may come from differences in educational attainment.

  • 7 One of them is Russia before and after the Soviet era, see Gerber and Hout (2004).

30In sum, these results suggest that, despite the persistence of high absolute mobility, the fluidity of social stratification in Mexico may have decreased for birth cohorts exposed to the crisis and structural change of the 1980s and 1990s, at least in urban areas. Although more research is needed, these results are remarkable because there are only a few international examples that show a similar trend,7 and they oppose what has been reported in other Latin American countries such as Brazil, where the pattern of association between origins and destinations has become more fluid (Costa Ribeiro 2007).

Prevalence of Erikson and Goldthorpe’s Core Model

31Finally, the fourth feature is that, despite Mexico’s historical and institutional peculiarities, relative mobility patterns conform quiet well to the core model of social fluidity proposed by Erikson and Goldthorpe for industrialized countries (Erikson and Golthorpe 1987, 1993).

32An important branch of contemporary sociological research on social mobility has focused on the identification of patterns of association between social class origins and destinations through the empirical analysis of the mobility table. The predominant approach is to adjust log-linear models to the observed frequencies in the table. These models may include different sets of parameters, which in turn reflect the hypothesized pattern of association postulated by the researcher. Erikson and Goldthorpe’s Core Model includes eight parameters, and reflecting four theoretical effects (inheritance, sector, hierarchy, and affinity).

  • 8 As in Table 1, in the case of females the parental class is measured through father’s occupation at (...)

33Solís and Cortés (2009) adjusted a modified version of the Core Model to a 144 cell three‑way table featuring a 6x6 class intergenerational mobility table across 4 regions of the country (see tables A1 and A2 in the Appendix for details of the occupational classification and the specification of the loglinear coefficients). Their results are presented in Table 5. The table also includes results for the “main diagonal” and “quasi-perfect mobility” models, which are simpler models and are presented for comparison purposes.8

Males

G2

df

Δ

1. Main diagonal, region-fixed effects

741.3

99

13.5

2. Quasi-perfect mobility, region-fixed effects

354.5

94

7.6

3. Modified version of Erikson and Goldthorpe’s Core Model, region-fixed effects

177.7

93

5.0

4. Modified version of Erikson and Goldthorpe’s Core Model, region-specific effects

153.1

72

4.6

Females

G2

df

Δ

1. Main diagonal, region-fixed effects

964

99

12.9

2. Quasi-perfect mobility, region-fixed effects

642.8

94

9.6

3. Modified version of Erikson and Goldthorpe’s Core Model, region-fixed effects

300.5

93

6.9

4. Modified version of Erikson and Goldthorpe’s Core Model, region-specific effects

258.2

72

6.1

Table 5 - Goodness of Fit Statistics for Log-linear Models of intergenerational Occupational Mobility Tables, Mexico, 2005.

Source: Solís and Cortés (2009).

34Results show that the modified version of the Core Model (model 3) fits quite well the frequencies of the table for males. The dissimilarity index between fitted and observed frequencies (Δ) is 5.0%, a figure of similar magnitude to that reported by Erikson and Goldthorpe (1987, 1992) for industrialized nations. In addition, the inclusion of interaction parameters for region-specific effects (model 4) does not produce significant improvements in the model, thus suggesting that a unique model is appropriate to account for the pattern of association across different regions. The model fits also reasonable well for females (Δ=7.6), indicating that the Core Model appropriately describes relative intergenerational mobility patterns between fathers and daughters.

35What do these models tell us about the pattern of intergenerational class mobility in Mexico? First, at least in this aspect Mexico is not an exception or an isolated case. The core model works reasonably well in Mexico and this may help to include the country in comparative international studies contrasting the strength of inheritance and other forces driving occupational mobility.

36Second, the model also fits quite well for women, suggesting that it may be used to open an avenue of research −the intergenerational occupational mobility of women− scarcely explored in Mexico. The odds ratios obtained from the models (Table 6) are informative of the forces shaping mobility patterns in Mexico. The most important effects are those associated to inheritance and affinity. The magnitude of inheritance reflects the strength of intergenerational reproduction, particularly among the service class and farmers. The first affinity parameter (actually a dis-affinity rather than an affinity parameter) reflects how low chances for mobility between farm occupations and the service class are; the second affinity parameter shows the higher chances of mobility between the service class and other routine non­manual positions, as well as between skilled and unskilled manual positions.

Males

Females

Parameter

Hierarchy 1 (hi1)

0.81*

0.77*

Hierarchy 2 (hi2)

0.92

0.56*

Inheritance 1 (in1)

2.16*

1.56*

Inheritance 2 (in2)

2.47*

2.34*

Sector (se)

0.63*

0.70*

Affinity 1 (af1)

0.49*

0.46*

Affinity 2 (af2)

1.88*

2.09*

* p < 0.05

Table 6 - Odds Ratios for the Parameters of the Modified Version of Erikson and Golthorpe’s Core Model.

37Finally, the only major difference between males and females is that the second hierarchy coefficient is statistical significant for women. The estimated effect is of a similar magnitude than those associated to the inheritance and affinity dimensions (eB = 0.56). Thus, in relation to males, females confront additional barriers for long term mobility between unskilled manual/farm occupations and the service class.

Discussion and Conclusions: Future Research on Social Mobility in Mexico

38At the end of the 1990s knowledge about social mobility patterns in Mexico was scarce. It was not possible to answer basic questions about the impact of social and economic changes on social mobility. A number of studies carried out since then have helped to shed some light on this topic. The purpose of this paper has been to summarize the most important findings of recent studies of social mobility, evaluating as well whether these features can be generalized to the overall national context.

39I have discussed four major findings: a) the continuity of high overall and upward mobility rates; b) the reduction of income gains associated to upward occupational mobility; c) the increasing rigidity in relative rates of occupational mobility; and d) the extent to which the Mexican case conforms to the core model of social fluidity proposed by Erikson and Golthorpe.

40The overall picture emerging from this findings is of a society that, notwithstanding the negative effects of the economic recession and structural changes of the 1980s and 1990s, maintained its high rates of structural mobility, but suffered in other aspects such as the decrease of the economic rewards associated to upward mobility, as well as the increasing rigidity in relative mobility patterns.

41Although these findings are important, there are many aspects that require further study. I will finish this article by briefly mentioning to four of them.

  • 9 Part of the results reported in this article come from the 2005 survey, see (Rabell 2009). The 2006 (...)

42First, more and better data sources are needed. There have been significant improvements in this aspect. Information on occupational mobility was included for the first time in a national survey in 2005, and there have been two additional national surveys carried out in 2006 and 2011,9 as well as surveys for specific cities such as Mexico City (Solís 2011). However, the inclusion of social mobility as a regular topic in official surveys has not been accomplished. Incorporating measurements of social mobility into official statistics would produce more reliable measures and also contribute to enhance the importance of social mobility as a topic in the agenda of public policy.

43Second, it is important to direct more attention to the mechanisms and processes leading to the intergenerational reproduction of social inequalities, particularly in the fields of education and occupation. Recent studies on social mobility have documented the extensive of opportunity currently existing in Mexico. Some studies have linked inequality of opportunity to the social and economic changes that have taken place after the crisis of the 1980s, and particularly to the implementation of neoliberal economic and policies (Cortés, Escobar and Solís, 2007). However, as appealing as this hypothesis may be, inequality of opportunity was high long before the application of these policies. This suggests that the underlying meso- and micro-mechanisms producing inequality of opportunity were active before the structural changes of the 1980s and 1990s and may still be operating today. However, little is known about these specific mechanisms. For example, research on the effects of social networks on the reproduction of inequality or on the multiple ways in which inequalities in parental origins become institutionalized in the educational field is still scarce. Therefore, new studies on inequality and social mobility must pay increasing attention to the study of these mechanisms.

44Third, studies of social mobility may benefit from a comparative approach with other countries. At the same time that studies of social mobility were abandoned in Mexico during the 1980s and 1990s, the field experienced significant evolution in industrialized countries, both in theoretical perspectives and in methodological approaches. One of the major developments has been the proliferation of comparative studies that involve a growing number of countries (Grusky and Hauser 1984, Erikson and Golthorpe 1993, Breen and Luijkx 2004). Incorporating Mexico to these studies is important to establish similarities and differences in social mobility patterns with other nations. As I have discussed earlier, some features of social mobility in Mexico −such as the persistence of high rates of overall mobility− are shared with many other nations, but at the same time other features −such as the apparent rise in rigidity− may be distinctive of Mexico. A logical step to further test hypotheses about the commonalities and specificities of Mexico would be to carry out comparative analyses with countries in the region with available recent data, such as Argentina, Chile and Brazil.

45Finally, studies on social mobility cannot ignore the importance of international migration as an endogenous component of social stratification. There are close to 10 million Mexicans between ages 15 and 64 in the United States, a figure equivalent to 16% of the working-age population (Giorguli, Gaspar and Leite, 2007). An exodus of this magnitude has important implications for social stratification within Mexico. For example, the exit of migrants may alleviate pressures in the labor market and reduce competition for upward mobility, not only because the stock of individuals in search of better occupations decreases, but also because migrants transfer significant amounts of money to their families in Mexico, thus alleviating social and economic demands. On the other hand, a massive return of migrants to the country might produce the opposite effect. However, more research is needed to fully understand these and other possible links between social mobility in Mexico and international migration.

Appendix

I. Service Class

Profesionals, managers, high level officials in the public and private sector, university professors

II. Nonmanual routine workers

Middle level managers in the public and private sector, teachers, artists,office clerks, sales agents in realty and insurance services

III. Sales

Sales employees in established bussinesses

IV. Specialized manual workers

Factory supervisors, machine operatives, craft workers, vehicle drivers, specialized laborers

V. Unspecialized manual workers

Workers in street sales and personal services, domestic workers, security workers, helpers, craft apprentices, nonspecialized factory workers, nonspecialized construction workers

VI. Farm workers

Unskilled and semi-skilled laborers in agriculture

46Table A1 - Class scheme used to fit loglinear intergenerational mobility models of Tables 5 and 6.

Respondent’s class

I

II

III

IV

V

VI

Parental class

I

in1

0

0

0

0

0

II

0

in1

0

0

0

0

III

0

0

in1

0

0

0

IV

0

0

0

in1

0

0

V

0

0

0

0

in1

0

VI

0

0

0

0

0

in1

472. Quasi-perfect mobility

Respondent’s class

I

II

III

IV

V

VI

Parental class

I

in1

0

0

0

0

0

II

0

in2

0

0

0

0

III

0

0

in3

0

0

0

IV

0

0

0

in4

0

0

V

0

0

0

0

in5

0

VI

0

0

0

0

0

in6

483. Modified version of Erikson & Goldthorpe’s Core Model

Respondent’s class

I

II

III

IV

V

VI

Parental class

I

in1+in2

hi1+af2

hi1

hi1

hi1+hi2

hi1+hi2+se+af1

II

hi1+af2

in1

0

0

hi1

hi1+se

III

hi1

0

in1

0

hi1

hi1+se

IV

hi1

0

0

in1

hi1+af2

hi1+se

V

hi1+hi2

hi1

hi1

hi1+af2

in1

se

VI

hi1+hi2+se+af1

hi1+se

hi1+se

hi1+se+af2

se+af2

in1+in2

49Table A2 - Specification of Parameters for Log-linear Models presented in Tables 5 and 6.

Inicio de página

Bibliografía

Balán, Jorge, Harley L. Browning, and Elizabeth Jelin, 1973 − Men in a Developing Society; Geographic and Social Mobility in Monterrey, Mexico. Austin: Published for the Institute of Latin American Studies by the University of Texas Press.

Behrman, Jere., Alejandro Gaviria, and Miguel Székely, 2001 − “Intergenerational Mobility in Latin America.” Economia 2:1­44.

Blau Peter M. and Otis D. Duncan, 1967 − The American Occupational Structure. New York: John Wiley.

Breen, Richard, - 2010. “Social Mobility and Equality of Opportunity.” Economic and Social Review 41(4): 413-28.

Breen, Richard and Ruud Luijkx, 2004 − “Social Mobility in Europe between 1970 and 2000.” pp. 37–75 in Social Mobility in Europe, edited by Richard Breen. Oxford: Oxford University Press.

Contreras Suárez, Enrique, 1978 − Estratificación y movilidad social en la Ciudad de México. 1º ed. México: Universidad Nacional Autónoma de México.

Cortés, Fernando and Agustín Escobar Latapí, 2005 − “Movilidad social intergeneracional en el México urbano.” Revista de la cepal pp. 149­167.

Costa Ribeiro, Carlos Antonio, 2007 − Estrutura de classe e mobilidade social no Brasil. Sao Paulo: Edusc.

Escobar Latapí, Agustín, 2001 − Nuevos modelos económicos: ¿nuevos sistemas de movilidad social? Serie Políticas Sociales No. 50. Santiago: cepal.

Erikson, Robert and John H. Goldthorpe, 1987 − “Commonality and Variation in Social Fluidity in Industrial Nations. Part II: The Model of Core Social Fluidity Applied.” European Sociological Review 3:145­166.

1992 − The Constant Flux: A Study of Class Mobility in Industrial Societies. Oxford; New York: Clarendon Press; Oxford University Press.

Filgueira, Carlos, 2001 − La actualidad de viejas temáticas: sobre los estudios de clases, estratificación y movilidad social en América Latina. Serie Políticas Sociales No. 51. Santiago: cepal.

Gerber, Theodore and Michael Hout, 2004 − “Tightening Up: Declining Class Mobility During Russia’s Market Transition.” American Sociological Review, 69:677.

Giorguli Saucedo, Silvia, Selene Gaspar Olvera, and Paula Leite, 2006 − La Migración Mexicana y el Mercado De Trabajo Estadounidense: Tendencias, Perspectivas y ¿oportunidades? México: Consejo Nacional de Población; Secretaría de Gobernación.

Grusky, David B., 1994 − “The contours of social stratification”. pp. 3-35 in: David B. Grusky (Ed.) Social stratification: Class, race, and gender in sociological perspective. Boulder, CO: Westview Press.

Grusky, David B., and Ravi Kanbur, (eds.) 2006 − Poverty and Inequality. Stanford: Stanford University Press.

Kessler, Gabriel and Vicente Espinoza, 2003 − Movilidad social y trayectorias ocupacionales en Argentina. Rupturas y algunas paradojas del caso de Buenos Aires. Serie Políticas Sociales No. 66. Santiago de Chile: Comisión Económica para América Latina y el Caribe, División de Desarrollo Social.

Minnesota Population Center, 2011 − Integrated Public Use Microdata Series, International: Version 6.1 [Machine-readable database]. Minneapolis: University of Minnesota.

Munoz García, Humberto, Orlandina de Oliveira and Claudio Stern, 1977 − Migración y Desigualdad Social En La Ciudad De México. 1º ed. México: Instituto de Investigaciones Sociales, Universidad Nacional Autónoma de México.

Rabell Romero, Cecilia (coord.), 2009 − Tramas familiares en el México contemporáneo. Una perspectiva sociodemográfica. México, D.F. Instituto de Investigaciones Sociales, unam-El Colegio de México.

Serrano Espinosa, Julio and Florencia Torche (Editors), 2010 − Movilidad social en México: población, desarrollo y crecimiento. México, D. F. Centro de Estudios Espinosa Yglesias.

Solís, Patricio, 2002 − “Structural Change and Men’s Work Lives. Transformations in Social Stratification and Occupational Mobility in Monterrey, Mexico.” Doctoral Dissertation, University of Texas at Austin.

2005 − “Espacio social y estilos de vida en Monterrey. Una evaluación crítica de la propuesta de Pierre Bourdieu”. In: Isabel Jiménez (ed.) Ensayos sobre Pierre Bourdieu y su obra. México: Plaza y Valdés , 388p.

2007 − Inequidad y Movilidad Social en Monterrey. México, D.F. El Colegio de México, 358p.

2010 − “Ocupaciones y clases sociales en México”. En: Julio Serrano Espinosa y Florencia Torche (Editores), Movilidad social en México: población, desarrollo y crecimiento. México, D. F. Centro de Estudios Espinosa Yglesias.

2011 − “Desigualdad y movilidad social en la Ciudad de México”. Estudios Sociológicos XXIX-85, enero-abril. pp. 283-298.

2012 − “Desigualdad social y transición de la escuela al trabajo en la Ciudad de México”. Estudios Sociológicos 90, pp. 641-680.

Solís, Patricio and Fernando Cortés, 2009 − “La movilidad ocupacional en México: rasgos generales, matices regionales y diferencias por sexo”. pp. 395-433 in: Cecilia Rabell Romero (coord.): Tramas familiares en el México contemporáneo. Una perspectiva sociodemográfica. México, D.F.: Instituto de Investigaciones Sociales, unam-El Colegio de México.

Solís, Patricio (forthcoming), Estratificación y movilidad social en México: ¿Nuevas pautas o tendencias recurrentes?

Valle Silva, Nelson, 2004 − Cambios sociales y estratificación en el Brasil contemporáneo. 1945­1999. Serie Políticas Sociales No. 89. Santiago de Chile: Comisión Económica para América Latina y el Caribe.

Zenteno, René and Patricio Solís, 2006 − “Continuidades y discontinuidades de la movilidad ocupacional en México.” Estudios Demográficos y Urbanos 21­3.

Inicio de página

Notas

1 Even when social mobility can be studied across several dimensions (i.e. education, income, status, etc.), sociological studies on social stratification have typically focused on intergenerational occupational mobility. Occupations are usually collapsed into a limited set of “social classes” that reflect major economic and social divisions (Grusky 1994, Olin Wright 2005, Solís 2009). In this article I follow this tradition, so when I talk about “social mobility” I am referring to intergenerational mobility across occupational classes, unless otherwise noted.

2 See, among others: Behrman, Gaviria and Székely (2001), Solís (2002, 2005, 2007), the series of articles published in Cortés, Escobar and Solís (2007) and Serrano Espinosa and Torche (2010).

3 See Torche and Wormald (2004) for Chile and Valle Silva (2004) for Brazil.

4 Mobility rates for woman are calculated from social mobility tables where origins are measured using as parental class father’s and not mother’s occupation. It is not practical to generate mobility tables using mother’s class as origin because most mothers never had work experience.

5 The micro-data were provided by the IPUMS-international Project of the University of Minnesota (Minnesota Population Center 2008).

6 National employment surveys were implemented in the 1970s, but the micro-data necessary to estimate wages by occupation are only available for the years after 1992.

7 One of them is Russia before and after the Soviet era, see Gerber and Hout (2004).

8 As in Table 1, in the case of females the parental class is measured through father’s occupation at age 15, and the occupation of destination is the current or last occupation.

9 Part of the results reported in this article come from the 2005 survey, see (Rabell 2009). The 2006 and 2011 national surveys were funded by the Centro de Estudios Espinosa Yglesias, which has actively promoted research on social mobility in Mexico in recent years (see: http://www.movilidadsocial.org).

Inicio de página

Índice de ilustraciones

URL http://trace.revues.org/docannexe/image/1061/img-1.jpg
Ficheros image/jpeg, 124k
Inicio de página

Para citar este artículo

Referencia en papel

Patricio Solís Gutiérrez, « Social mobility in Mexico. Trends, Recent Findings and Research Challenges », Trace, 62 | 2012, 7-20.

Referencia electrónica

Patricio Solís Gutiérrez, « Social mobility in Mexico. Trends, Recent Findings and Research Challenges », Trace [En línea], 62 | 2012, Puesto en línea el 02 abril 2014, consultado el 22 marzo 2017. URL : http://trace.revues.org/1061

Inicio de página

Autor

Patricio Solís Gutiérrez

Profesor-investigador del Colmex. Doctor en Sociología por la Universidad de Texas (2002), obtuvo la Maestría en Población por la Facultad Latinoamericana de Ciencias Sociales, sede México en 1995, es licenciado en Sociología por la Facultad de Filosofía y Letras de la Universidad Autónoma de Nuevo León. Fue Visiting doctoral student por el Max-Planck-Institut für Demografische Forschung, Rostock, Alemania. Publicaciones más recientes: “Inequidad y movilidad social en Monterrey”, México, El Colegio de México. Centro de Estudios Sociológicos, 2007. Escobar, Agustín, Cortés, Fernando, (editores) “Cambio estructural y movilidad social en México”, México, El Colegio de México. Centro de Estudios Sociológicos, 2007. “La marginación urbana. México”, D.F.: segob-Consejo Nacional de Población, 2002.

Inicio de página