Artículos

 

Vulnerability Factors in the Middle Class: Evidence for Argentina and Mexico after the Crisis of the 1990s

 

Factores de vulnerabilidad de la clase media: Evidencias para Argentina y México luego de las crisis de los años noventa

 

Gabriela Liliana Galassi¹ y Leandro Mariano González²

 

¹ Becaria de doctorado del CONICET (Centro de Estudios Avanzados–CONICET). Dirección electrónica: gabygalassi0502@gmail.com

² Investigador del CONICET (Centro de Estudios Avanzados–CONICET). Dirección electrónica: leandrogonzalez@yahoo.com

 

Fecha de recepción: 6 de diciembre de 2010
Fecha de aceptación: 10 de junio de 2011

 

Abstract

This paper explores the profile of the Argentinean and Mexican middle classes during the most recent crisis in both countries. It combines the perspectives of social vulnerability and class analysis theoretically underlying a "matrix of vulnerability and social classes." The analysis used household surveys in Argentina for 1998 and 2003 and those in Mexico for 1994 and 1996. The results show that whereas the Mexican middle class was primarily affected during the "Tequila crisis" through its physical assets (housing conditions), in Argentina, education and the labor market were the main mechanisms affected by the 2001 crisis.

Keywords: 1. Middle class, 2. new poor, 3. vulnerability, 4. crisis, 5. demographic, economic and educational level profile.

 

Resumen

Este trabajo explora el perfil de la clase media argentina y la mexicana durante las últimas crisis en ambos países. Se combinan las miradas de vulnerabilidad social y análisis de clase, subyaciendo teóricamente una "matriz vulnerabilidad–clases sociales". El análisis emplea las encuestas de hogares de Argentina para 1998 y 2003 y México, 1994 y 1996. Se observa que mientras la clase media mexicana se vio principalmente afectada durante la crisis llamada "efecto tequila" vía sus activos físicos (condiciones habitacionales), en Argentina la educación y el mercado laboral fueron los principales mecanismos afectados por la crisis de 2001.

Palabras clave: 1. Clase media, 2. nuevos pobres, 3. vulnerabilidad, 4. crisis, 5. perfil demográfico, económico y nivel educativo.

 

INTRODUCTION

Interest in the impoverishment of the middle class emerged in the Latin American region since the 1990s revealed a profound concentration of income to the detriment of the class in question (Hoffman and Centeno, 2003).¹ Middle–class sectors in the countries in the region were affected by the 1980 debt crisis and subsequently by the structural adjustment programs, the concentration of wealth, the increased price of basis services and the deterioration of those provided by the public sector, and particularly the modification of the labor market in the 1990s. The impact was not homogeneous throughout the middle class and instead, certain sectors were affected, experiencing downward social mobility whereas others succeeded in mobilizing their assets, managing to maintain their position on or even move up the social scale. Not only did certain sectors of the middle classes become impoverished; the process of impoverishment affected the entire social structure, while some sectors actually improved their situation. This led to an increase in social inequality in general. Strictly speaking, this process of increasing inequality within the social structure is not restricted to the Latin American region but had already occurred in Europe in the late 1980s (Kessler y Di Virgilio, 2008).

It is important to note that despite the scope and importance of the phenomenon of the impoverishment of the middle class, it has yet to occupy a key position on either the academic agenda or Latin American policy. This is quite different from the situation Europe, where social classes have been the subject of attention for a long time. That is why we think it is important to review the experience in this issue, in order to begin to determine the characteristics of this phenomenon and obtain lessons for the future. This article examines the experiences of Argentina and Mexico regarding the profiles of class structure that emerged during the latest major crises in the two countries. To this end, it analyzes the demographic, economic and educational level of social classes, using household surveys from 1998 and 2003 in Argentina and from 1994 and 1996 in Mexico.

Argentina historically distinguished itself from other Latin American countries by the size of its middle–class sectors (Minujin and Anguita, 2004; González Bombal and Svampa, 2001; Kessler, 2002). There is another feature of the Argentinean middle class that sets it apart from the rest of Latin America: the strong cultural component that forms part of its identity, which takes precedence over income (Minujin and Anguita, 2004). The Argentinean middle class, which continued to rise for more than half the last century, began to show signs of weakness after the end of the industrialization model in 1975. The new model established during the last quarter of the century, characterized by the withdrawal of the state (which had traditionally guaranteed security for the middle class), coupled with the collapse of the labor market and the education system, the most effective mechanisms for social mobility, led to the weakening of middle–class sectors. During the 1980s, impoverishment operated through a drop in the purchasing power of salaries due to inflation, whereas in the 1990s, unemployment (which exceeded 20 % in 2002), severely affected the conditions of the middle classes. As a result of the 2001 crisis, several of them were plunged into poverty, creating the group of "new poor" (Minujin et al., 1992; Minujin and Anguita, 2004, Kessler, 2002).

A similar process occurred in Mexico due to the Tequila Crisis (1995). During the 1990s, unlike in Argentina, impoverishment was caused by the expansion of informality and job precariousness (with low salaries) rather than by an increase in unemployment; unemployment rates were relatively low during the decade. A process of income polarization occurred due to the increase in real salaries of certain categories (directors and workers in semi–skilled services) to the detriment of the rest (Kessler y Di Virgilio, 2008).

It is worth noting that in Mexico, as in Argentina, the 1980s saw a series of structural adjustments to strengthen the role of the market to the detriment of the state. The adjustments led to an expansion of the informal labor market (Cortés and Escobar Latapí, 2005) and a drop in the purchasing power of salaries which, together with a criterion of efficiency, produced an increase in the concentration of income (CASA, 2001). Following a sharp devaluation of the peso in December 1994 (caused by a nationwide collapse due to massive capital flight) and a battery of structural adjustments, the Mexican economy experienced its greater contraction since the Great Depression. Although the middle class was affected by these measures, some studies revealed that the short term financial impact mainly affected the highest strata (Gilbert, 2005). This triggered a process of impoverishment originated by the crises of the 1990s.

The inequality of opportunities increased in Mexico, decreasing the possibilities of social mobility, particularly in the lower classes. The rise in inequality combined with the deterioration of the usual mechanisms of mobility: the system of occupational mobility became more rigid, while education lost its role as a buffer against downward mobility. The crisis affected the middle class since it restricted the possibilities of credit, weakening small and medium Mexican firms, one of the main sources of income for the Mexican middle class (Camberos, 2009).

The importance of the middle class lay in its key role in the processes of development, evinced throughout the 20^th century in Latin America and prior to that in European countries and the United States. The scope and profile of this class, which mediates between the upper and working class, determine a society's degree of polarization. Hence the possibility of its disappearance, not in quantitative terms but as regards its identity. This poses a serious risk to the possibility of its future development.

Since crises have become a recurrent phenomenon in most Latin American countries, it is important to determine their effect on the social structure, particularly the middle class, which has shown a tendency to experience downward mobility due to the external shocks of the past few decades. It is essential to determine the characteristics of the middle–class sectors and the possible factors that make them likely to suffer downward mobility in order to design interventions focusing on strengthening their tools and offsetting their weaknesses.

The aim of this paper is to analyze the starting point and end of social classes in two experiences of Latin American crises (the Tequila Crisis in Mexico and 2001 in Argentina) in order to explore the hypothesis that the middle–class sectors have experienced a situation of relative vulnerability that makes them prone to becoming poorer when a macro–economic shock occurs. Previous results for Argentina based on the 2001 Census (Galassi, 2010b) show that some of the features causing the greatest risk for the middle (as opposed to the upper classes) include incipient ageing, larger households, a higher proportion of extended households, greater youth fertility, a lower proportion of households with dual–income couples and lower educational achievement. They also show that in marriage patterns, housing quality, proportion of immigrants, the possession of material assets and the incidence of formal employment, the middle class move away from the working class and towards the upper classes.

The following section reviews the main features of the theoretical lines of vulnerability and social classes, which are the approaches that will be used to explore the profile of social classes. A class map is then drawn up on the basis of the household surveys in the countries being analyzed in the period of study. This is used to provide a description of the social structure. Lastly, a number of conclusions on the analysis carried out are presented.

 

THEORETICAL CONSIDERATIONS

This study was undertaken within the framework of the issue of social inequality, in the line of studies on social vulnerability and class analysis. Social inequality is a crucial issue in Latin America, one of the continents with the highest degree of inequity.

In the area of social inequality, the concept of multidimensional vulnerability, offsets the limitations of the one–dimensional approach to poverty. However, the inclusion of the condition of vulnerability only provides a partial image in terms of social inequality. It can be complemented by a view of the structure underlying the society in question. Individuals in a society have varying degrees of vulnerability (an individual feature) according to their relative position in the social structure (relational feature, relative to the rest of individuals in society). This article therefore seeks to combine the two perspectives (social vulnerability and class analysis) theoretically underpinning a "matrix of vulnerability and social classes" (Galassi, 2010a). Since the two approaches are located in the sphere of social inequality, each one describes different aspects of the multidimensional phenomenon of the social situation of individuals and households.

Class Structure

Class analysis is designed to define the concept of social class² and outline the limits of classes. Although social stratification involves delimiting social classes, one should remember that they are heterogeneous (López, 2007). One of the main objectives of class analysis is to analyze the relationship between social classes and life opportunities, although patterns of behavior, action, attitudes, and values are observed, among other things (Breen, 2005).

Although social structure involves both financial resources and the forms of organization that participate in social interaction, thereby including both economic and political aspects (Icazuriaga, 1994; Sánchez, 1999), the authors that have conducted class analyses have opted for a criterion based on socio–economic elements since there is a close link between economic sphere and social classes (Cortés and Escobar, 2005).

Although position in the labor market has been a key element in social stratification, this does not mean one can ignore the fact that there are elements outside this that influence the determination of social structure (Escobar, 1995). Torrado (1992 and 2007), an Argentinean researcher who has adopted this line of argument, has developed a social stratification methodology on the basis of economic or occupational attributes provided by population censuses. Since Torrado's work is situated in 20th century Argentina and her methodology is suitable for undertaking censuses and surveys, this analysis will use her approach while adapting it to the period under study and available data sources.

Before reviewing the concepts in Torrado's approach, it is important to have a definition of the middle class. In Argentina, this social class, comprising both a self–employed stratum and a salaried one, has traditionally been identified by its cultural pattern (mainly a high level of educational attainment) and its links with the state (public employment and education) (Minujin and Anguita, 2004). For its part, the Mexican middle class, is defined by its goals: home ownership, private education and professional financial success and includes non–manual workers with high educational achievement, located in intraurban zones (Zamorano, 2007b; López, 2007; Sánchez, 1999). For both Argentina and Mexico, the operational criteria used in the studies to identify the middle class are far from perfect and in many cases, the complexity of the concept of middle class has yet to be fully grasped (Zamorano Villarreal, 2007a).

The Principal Concepts of Torrado's Analysis

The stratification methodology proposed by Torrado is part of the Marxist approach (Sémbler, 2006), in which social classes are identified on the basis of the concept of the social division of labor in keeping with the relations of production. The latter refer to the distribution of social agents in positions based on their control of the production process, determined by the possession of the means of production. Thus relations of production constitute the criteria for delimiting the sub–set of social agents that occupy an analogous social position, so–called "social classes."

However, the only agents immersed in the division of labor are those that comprise the economically active population (EAP). In order to define the position occupied by individuals outside the EAP, the criterion of relations of distribution is used, in which the inactive population plays an indirect role through its consumption. The link between the two groups (active and inactive) and therefore the relations of production and distribution, is established through income transfers made by active individuals within the family. In other words, stratification originally operates at the level of individuals in the EAP, usually heads of household, after which the position of inactive agents is defined on the basis of their belonging to a certain family.

Within Torrado's class scheme, the middle classes (Sémbler, 2006), share the border with both the bourgeoisie (becoming "petite bourgeosie owners," which includes owner–sellers, independent workers, workers in family industries, etc.) and the working class ("semi–proletariat working class" which includes commercial agents and office clerks, among others) as well as a group of ideological (teachers) and political agents (state employees).

Social vulnerability

The term "vulnerability" has appeared in intellectual and government discourse in Latin America for several decades. However, no great conceptual precision is observed when it is used to refer to social vulnerability, which is usually confused with poverty. The studies on social vulnerability gained importance in the mid–1990s as a response to the technical development regarding the new forms of social inequality derived from the economic instability of developing countries and the advance of globalization (González, 2010).

The vulnerability approach is immersed in a series of new dynamic, multidimensional approaches. Analytical dissatisfaction with the use of the notion of poverty (and its measurement based on household income) to identify reduced social situations and orient social policies has been the main incentive for developing vulnerability approaches (Arriagada, 2005; Pizarro, 2001). But in addition to its theoretical value, the social vulnerability approach is important on the political agenda. This is because the possibilities of poverty involve the risk of conducting policies aimed simply at transferring resources to "pull" households out of poverty, which, although it may yield results in the short term, is by no means a long–term solution (CEPAL, 2002; Pizarro, 2001).

Moreno (2008) identifies two lines in the studies on vulnerability: first of all, there are the conceptions that refer to the fragility, precariousness, defenselessness and uncertainty regarding the characteristics of individuals' situation (households, communities, etc.) The other perspective is based on the notion of risk. In this case, the characteristics of individuals' and households' characteristics are more important, since their effect on the units creates an unfavorable or vulnerable situation. The difference between these two lines is that the former emphasizes the distribution of attributes and material or symbolic assets while the latter explicitly introduces the notion of risk. Most of the work of international organizations and authors such as Kaztman falls within the first approach (González, 2010). The work proposed here is also framed within this conception.

The most important position on this issue in Latin America is that of the Economic Commission for Latin America and the Caribbean (CEPAL, 2002), which, in a document drawn up by the Latin American and Caribbean Center of Demography (Celade), explained its doctrine regarding the topic of vulnerability from a sociodemographic perspective, regarding it as more of an analytical approach than a conceptual category. The concept of vulnerability involves several dimensions, the most important of which are economic, demographic and social (although the latter two interact in such a way that it is possible to speak of sociodemographic vulnerability) which shapes the key profiles of sociodemographic risk for communities, households and persons.

Vulnerability simultaneously involves a potentially adverse event (an exogenous or endogenous risk), an incapacity to respond and an inability to adapt to the new scenario (CEPAL, 2002). CEPAL acknowledges three main uses of the notion of vulnerability: 1) groups at "social risk," (individuals with a tendency towards a lack of social integration), 2) groups whose members display conduct with common features that make them liable to suffer damage; 3) groups that share basic characteristics that may create similar problems (age, sex, ethnic condition, territorial location) which are useful when it comes to identifying the target populations of social policies. This last criterion is applicable to middle class social groups.

As an antecedent to the CEPAL approach and other similar ones, there is Moser and Felton (2007)'s "asset vulnerability approach, which focuses on the strategies used by poor households to cope with the crisis. The central concept of the approach is assets, which goes beyond income and expenditure as indicators of household status, since they provide a more complete image of long–term living standards that have been accumulated over time and last longer. This idea of assets at the center of the conceptualization of vulnerability is sustained by other authors (Rodríguez, 2000). One of the advantages of dealing with vulnerability on the basis of an assets approach is that it makes it possible to obtain measurable features when it comes to determining the condition of vulnerability of a certain group. That is why this work is framed within this line.

Kaztman (1999 and 2000) and Kaztman and Filgueira (2006), for their part, used the conception of assets as determinants of the condition of vulnerability, although they added the concept of opportunity structures (state, market and community) which determine the likelihood of access to goods, services or activities that affect well–being.

On the basis of all these trends regarding social vulnerability, it can be regarded as a situation that emerges from the combination of economic, demographic and social factors. These factors, combined with the opportunity structure, means that individuals unable to response to a certain situation would be pushed down the social scale if it occurred.

 

METHODOLOGY

Regarding the period of analysis, in the countries studied, the last year of real positive growth prior to the crisis and the first one after it were taken into account. In the case of Argentina, this meant that 1998 and 2003 were chosen, whereas in the case of Mexico, 1994 and 1996 were chosen. The analysis was carried out on the basis of the Permanent Household Survey (EPH) in Argentina and the National Survey on Household Income and Expenditure (ENIGH) of Mexico.

EPH is the national program for the permanent, systematic production of social indicators. Its main objective is to find out about and characterize the population from the perspective of their socio–economic insertion, through information at the individual and household level. One of EPH's limitations is that it is only conducted in the country's main urban centers, excluding rural areas. Moreover, the data are obtained through a survey rather than a census. However, weighters are provided on the basis of demographic projections to expand the sample to the population covered in the survey.

Mexico's ENIGH is designed to provide a statistical overview of the behavior of household income and expenditure as regards its amount, origin and distribution. It also provides information on the occupational and socio–demographic characteristics of household members, as well as issues referring to the infrastructure of the dwelling and the household's equipment. As in the case of EPH, the microdata include variables at both the individual and the household level, in addition to providing a weighter to expand the sample to the total population.

Due to the information provided by household surveys, which include data on individuals' insertion in the labor market and their socio–economic characteristics, they are ideal for conducting analyses of inequality.

The class map in Argentina and Mexico was drawn up using Torrado's methodology (1992) adapted to the data sources chosen. The variables in the household surveys on insertion into the labor market were used, defining a series of occupation groups that subsequently became social classes. The unit of analysis is the household, whose position in the social structure is determined by the householder's insertion in the labor force. The rest of the household members shared the householder's position due to the relations of distribution.

The variables through which this classification is drawn up are: 1) the condition of activity, which delimits the universe on the basis of which it is possible to determine the position within the social structure. In Argentina, unlike Mexico, the EPH provides a description of unemployed persons with previous experience, in addition to those that are employed; 2) occupation, which, in Argentina makes it possible to determine the hierarchy (director, boss and worker) and degree of skill in the occupation (professional, technical, operative and unskilled). In Mexico, categories are drawn up around a scale of hierarchy; 3) occupational category (bosses, salaried workers, self–employed and unpaid workers), 4) the activity sector (public or private); 5) the size of the establishment (microbusiness or business sector).

On the basis of these variables, following Torrado (1992), 11 categories were drawn up. company directors, professionals in a specific function, small business owners, technical directors, small autonomous producers, administrative employees and sellers, autonomous specialized workers, skilled workers, unskilled workers, self–employed farm workers and unspecified (residual).³ The way these groups were obtained is shown in tables 1 and 2.

After drawing up these class maps for Argentina and Mexico, in the years assigned for each, the demographic and economic profiles and the educational attainment of the classes were examined in a comparative fashion and over time. For reasons of space, the results of the estimates by interval, with a confidence level of 95 percent, on which comments are made throughout the article, have been omitted.

 

RESULTS

Class maps

On the basis of the occupational groups defined earlier, it is possible to construct a class map for each country in each year (Table 3).⁴

The first thing that strikes one is that under this classification, Mexico had a smaller middle class, of approximately 33 percent of the population, compared with 46 percent in the case of Argentina, coupled with a large upper and working class. Beyond small changes in the social structure in the period under study: an increase in the working class to the detriment of the upper and middle class in Argentina and the opposite in Mexico, and the desalarization of the Argentinean middle class and of both this and the upper class in Mexico,⁵ the class maps do not show significant modifications. This is hardly surprising, since the modifications in occupational position and social structure are long–term issues, whereas what is modified in crises are the social responses to a single position.

It is therefore essential to observe the modifications in the class profile. Most of the studies on the impoverishment of the middle class in Argentina and Mexico (Minujin et al., 1992; Kessler, 2002; Minujin and Anguita, 2004; Gilbert, 2005) refer to the change in class identity. That is why this paper proposes a review of demographic and economic profiles and educational attainment before and after the crisis. The analysis is exploratory and observes the distribution of the population in key variables in each of the aspects explored.

Demographic Profile

The most important, long–term demographic changes in individuals' living conditions (CEPAL, 2002) are caused by the demographic transition: a steady drop in birth and death rates, which reduces the growth of the population and makes it older, increasing the household dependency ratio, and urban transition. a steady increase in the urban proportion and the importance of migratory movements between and within cities and the second demographic transition: postponement of first marriage and reproductive stage and loss of the formality and increase in the fragility of the institution of marriage. These processes constitute the main socio–demographic risk profiles. Moreover, among the population in general, it is a well–known fact that persons of an advanced age and women constitute a vulnerable group, which leads them to situations of defenselessness. Thus, an initial approach to the demographic profile of the social classes is provided by the population structure by sex and age.

Middle class population pyramids in Argentina show a shortage of population in the groups aged 20–24 to 35.39 (the economically active population), a gap that increases in 2003. This may indicate the fact that in both the working and the middle class, migration is a household strategy. Since this population is of an economically active age, it produces an increase in demographic dependency, which constitutes an aspect of vulnerability.

Conversely, the middle class structure is more rectangular (older) than that of the working class, with a narrower base (lower proportion of persons aged up to 14) and less rectangular than that of the upper class, which has an even narrower base. It is interesting that despite displaying significant differences between classes, the structure by large age groups experienced an insignificant change over time. As for the composition of the population by sex, this does not vary significantly either between social classes or over time. The lack of change in the age structure and sex of the population is natural due to the fact that they involve structural characteristics.

The Mexican pyramids also show the impact of the migratory phenomenon between 1994 and 1996, since this last year saw a greater split in the 15 to 25 year age group, together with the fact that it occurred among both sexes, whereas in 1994, it was an eminently male phenomenon.

As for the phenomenon of migration in active ages, the academic community of Latin America has already expressed its concern about the problem of the emigration of young professionals in the region, especially at times of economic recession (Didou, 2008).

Chart 1

In both countries, upper and middle class households are similar in size and smaller than working class households. Moreover, the proportion of single–parent nuclear households in Argentina increased during the crisis periods analyzed, to the detriment of nuclear and extended households. However, the proportion of nuclear households, the most common in all classes in general and the Mexican middle class in particular (Icazuriaga, 1994), has similar figures in the latter to those to the upper class and is larger than in Argentina. The increase in single–parent households, which reflects an increase in the fragility of family links, can also be observed in the Mexican upper class.⁶ This weakening of the nuclear family corresponds to a global transformation within the second demographic transition, which poses a risk (CEPAL, 2002). This phenomenon has been observed in Mexican society in general by authors such as Esteinou (1999). But this has not stopped the middle class, and the upper class, in the case of Mexico, from receiving the greatest impact in this respect. Both countries have seen an increase in the proportion of women–led households in all social groups, with the exception of the Argentinean upper class.

As for marital status, the proportion of individuals living with their partners is higher among the Mexican middle class than in the other classes. This is not the case in Argentina, where figures for this class are lower and closer to that of the working class during the period studied. It is worth noting that the proportion of persons living with their partners does not provide information on the stability of these couples. Recent decades have seen an increase in the instability of couples, leading to a proliferation of separations and divorces and new unions (Esteinou, 1999), phenomena that are not recorded by the data sources used here.

Chart 1

Regarding fertility, since household surveys are unsuitable for calculating direct or indirect measures, it is only possible to obtain an approximate indicator of the phenomenon: the number of children under five for every 1 000 women. This reveals, first of all, higher fertility in Mexico than in Argentina, while the indicator for the middle class in both countries is closer to that of the middle class. Moreover, while the ratio is lower in Argentina for all classes, in Mexico it is higher for the middle class.

Table 2

In short, the advance of the demographic transition, the "brain drain" and the combination of the harmful effects of the second demographic transition (weakening of marital links and increase in single–parent and women–headed households has been observed in the middle class of both countries⁷ (CEPAL, 2002). However, these processes affect all social classes in both countries to a greater or lesser extent. Added to this is the fact that in certain aspects, such as the fertility pattern, the middle class is closer to the working class.

Economic profile

The economic factors that can indicate the degree of vulnerability include two main aspects: material conditions and insertion in the labor market.

As for the material conditions of existence, most studies used poverty measurements. In this case, due to the complexity of measuring poverty in Mexico because of the poverty line method,⁸ the concept of unmet basic needs (UBN) has been used. At the same time, housing is an essential aspect of the lives of individuals and families, meaning that its ownership and quality are also good indicators of the availability of material assets.

As far as employment status is concerned, recent changes in the labor market, due to modifications in the forms of production of the globalized economy (new productive processes, technological changes and labor deregulation, (new productive processes, technological changes and labor deregulation), have created new demands for labor based on values such as flexibility or adaptability which conflict with the current structure and increase the vulnerability of the active population. This may happen either as a result of exclusion from access to the labor market or as a result of the social risk derived from working conditions. Working conditions comprise the quality of work, characterized by the prevalence of high rates of temporary work and precariousness.

Despite the crisis, in Argentina, the proportion of households in poor condition⁹ decreased in all classes between 1998 and 2003 except in the middle class. In Mexico,¹⁰ the class whose housing conditions experienced the greatest decline is the upper class, whereas among both the working and the middle class, the proportion of households in poor condition remained relatively stable. Conditions of access to housing did not change for the middle class in either country (the proportion of home owners did not change),¹¹ with the upper class receiving the greatest impact. The share of homeowners fell by nearly 10 percentage points in Argentina and seven percentage points in Mexico,¹² approaching figures for the middle class.

A compound indicator of the existence of deprivation, often used in studies on inequality and quality of life is the unmet basic needs (UBN) indicator, which establishes a set of universal needs which, if unmet, indicate the presence of vulnerability. Although one should consider the specific situation of each country in order to define the set of shortages comprising UBN, after reviewing Rodriguez's (2005) proposal for Nuevo León, it was decided to use the same indicators in both countries to permit comparability.¹³ Following the methodology used by INDEC, a household is classified as UBN if it has at least one of the following indicators of deprivation: 1) overcrowding, over three persons per room; 2) unsuitable housing: rented room, hotel or hostel room, dwelling not intended for housing or mobile dwelling in Argentina, attic room, locale used as housing, mobile housing or refuge in Mexico; 3) poor sanitary conditions (no toilet); 4) lack of schooling, at least one child old enough to attend elementary school and 5) lack of capacity for subsistence (four or more persons per employed household member and householder who has not completed elementary school in Argentina or has had no schooling in Mexico). This measurement shows that the incidence of households with UBN has remained in a period of crisis, with the exception of schooling and health installations in Argentina and schooling and subsistence capacity in Mexico. The middle class in both countries is more severely affected by overcrowding and sanitation than by poor living conditions, as in the Argentinean working class and the capacity for subsistence, which is particularly important in Mexican working class sectors.

Table 3

As one can see, the concept of UBN does not show the process of impoverishment observed. This is because it measures structural poverty, which is a longer–term condition. In the case of Argentina, using the poverty line method, a significant increase in poverty was observed during the 2001 crisis, especially among the middle class (Galassi, 2010b), revealing the shift towards the group of "new poor."¹⁴ Although poverty as such, as measured by the poverty line, constitutes a circumstantial feature, the experience of falling into a situation of this nature, at the risk of being prolonged or repeated in the future, produces a long–term impact on individuals' identity.

As for the labor market, in Argentina, middle class activity and employment rates (which reflect access to the labor market) are similar to those of the upper class and are not significantly different between classes.¹⁵ Beyond the effect of the crises on access to employment: increase in unemployment in the middle and working class in Argentina and among the latter in Mexico, the greatest impact was produced in terms of the quality of employment. Proof of this is the marked increase in the proportion of the population that experiences difficulty in joining the labor market (unemployed, underemployed)¹⁶ in both the middle and the working class in Argentina. In Mexico, this indicator increased most in the upper class (approximately 10 percentage points). It should be noted that in this country, the indicators of precarious employment are greater than in Argentina, which has a higher employment rate.

Another indicator of the quality of employment is temporary employment, which the EPH records in Argentina by distinguishing between temporary employment positions, odd jobs or unstable jobs (of unknown duration), the proportion of which remained the same in the upper and working class and increased in the middle class. A similar indicator in Mexico corresponds to the proportion of contracts for a set time or task or jobs without contracts. This figure actually dropped during the period of crisis, significantly so for the middle class.

In general terms, the economic shock had a much smaller impact on the Mexican labor market than on the Argentinean one. The Argentinean labor market was also more affected in terms of quality of employment than as regards quantity or access.

Educational Attainment

Regarding the cultural profile, it was only possible to review the population's educational level by social class, due to the restrictions imposed by the data sources. Education, as a means of acquiring human capital, is one of the most important mechanisms of social mobility (Zamorano, 2007), which, in the case of the impoverishment of the middle class, plays a key role in preventing downward mobility. In order to analyze the population's educational profile, both educational systems were homogenized, regarding elementary school as equivalent in both cases whereas in Mexico, secondary education included both junior and senior high school. Tertiary education in Mexico corresponds to technical or commercial education with senior high school being a prior requirement.

One of the most striking features is the lower educational level of the Mexican population in general. Approximately half the Argentinean upper class has had access to university studies. A similar proportion of the middle class has completed junior high school, whereas the working class has failed to complete this level. In Mexico, between 35 and 40 of the upper class have failed to complete high school, although over a quarter have completed university, a slightly lower proportion than in Argentina. A similar proportion of the middle class at most have completed primary while nearly 40 percent of the working class have failed to complete this level.

Moreover, if one regards completed high school as the cut–off point between high and low educational attainment, the figure for all classes in Argentina is 10 percentage points higher than in Mexico. Added to this is the fact that in Argentina, the proportion of the population that has completed high school at school rose between 1998 and 2003 (to a lesser extent in the middle class), whereas in Mexico, the change in the proportion between the years under analysis is insignificant. It should be noted, however, that the low proportion of the Mexican middle class with either complete or incomplete university studies contrasts with what is found in certain cities (Icazuriaga, 1994), where the proportion of middle class children that attended university and the importance families placed on education to cope with the crisis were high.

It is also important to note that in both countries, the educational profile is significantly different between the various social classes. In both cases, the middle class is in a better position than the working class, albeit very distant from the upper class.

Tabla 4

 

FINAL REFLECTIONS

This paper has reviewed the profile of the social structure in both Argentina and Mexico before and after the last crises in each. The author sought to identify the elements that could provide indications of the causes of the varying impacts of the shocks in the various social groups, with particular emphasis on the middle class, since the literature has acknowledged the impoverishment of this class in Latin America.

The intrinsic age structure of the middle class, which is older than the working class and less fertile than the upper class, makes it vulnerable, since the proportion of dependent population is higher than in the upper class. This is exacerbated by the gaps in the active population in the middle class age pyramids, presumably due to the migration of young people, particularly in the case of women in Mexico. This pattern of vulnerability combines with the effects of the second demographic transition, which produce a change in household composition. This leads to the proliferation of single–parent homes in the middle class and also in the Mexican upper class. In keeping with this fragmentation, there has been an increase in the proportion of women–led households in all classes. The Mexican middle class is characterized by having stronger marital bonds than other classes, unlike the situation in Argentina where the middle class has approached the working class in terms of the percentage of individuals belonging to a couple. As for fertility, the middle class in both countries shares similar values to the working class, which obviously constitutes a risk factor.

Risks can also be identified in the economic sphere. In Argentina, in a context in which living conditions improved, they did not improve in the middle class, in which there were a number of households with overcrowded conditions and a lack of toilet facilities. In Mexico, the upper class has been most severely affected in terms of living conditions such as access to home ownership. Although middle–class households showed a certain incidence of overcrowding and poor sanitary conditions, they also suffered from the lack of capacity for subsistence.

The impact of the respective crises in the labor market was greater in Argentina than in Mexico, with problems regarding the quality of employment (rather than access to it) becoming particularly acute in the former. But this is offset by the fact that the indicators for the quality of employment in Mexico show considerable levels of precariousness. Moreover, the impact of the crisis via the labor market in Argentina had a greater effect on the middle and working class, whereas in Mexico, employment indicators had a strong effect on the middle class, particularly through the increase in unemployment. Lastly, the Argentinean middle class experienced a relative decline in its educational achievement, since the other classes increased their level of education more than the middle class sectors. In Mexico, however, educational attainment remained the same throughout the crisis.

In short, there are many channels through which the social structure can be affected, by creating various consequences in terms of future possibilities, depending on the assets affected and their starting point. Thus, despite the possible problems in Mexico in terms of working conditions and educational attainment, the greatest impact during the Tequila crisis occurred in the housing conditions, particularly of the upper classes, which, since they involve physical assets, are more likely to be able to recover. Argentina, which had traditionally experienced upward social mobility driven by quality employment and education, experienced the greatest problems in these two spheres, which questioned society's capacity for recovery. However, it should be recalled that in Mexico, Cortés and Escobar (2005) have experienced downward social and job mobility or fluidity as a result of which, despite the fact that the crisis has not shown a noticeable impact in terms education and work, society's lack of capacity for mobility was already underway. It should be noted that the opportunities for social mobility in Mexico, prior to the Tequila Crisis, fell during periods of economic crisis, although the middle class has been able to mobilize its assets to cushion their impact (Escobar, 1995). One should therefore qualify the conclusions on the impoverishment of the middle class in Mexico with its possibilities of using strategies which, once the crisis is over, will enable it to return to the living standards it enjoyed prior to the crises.

This analysis is designed to highlight the importance of the level of vulnerability of the middle class even in terms of the aim of alleviating poverty in the region. The middle class plays a key role in alleviating social polarization. In order to prevent its impoverishment, interventions must be designed to provide it with tools that will permit its subsistence and prevent a drop in its level of well–being. Policies must be oriented towards creating quality employment. Small and medium businesses must also be promoted through policies that include soft credits and variable tax rates, since they are one of the main sources of employment of members of the middle class. Lastly, greater access to education must be encouraged, taking care to ensure quality education standards and teach skills required for entering the labor market.

 

BIBLIOGRAPHY

Arriagada, Irma, 2005, "Dimensiones de la pobreza y políticas desde una perspectiva de género", Revista de la CEPAL, núm. 85, Santiago de Chile, pp. 101–113.         [ Links ]

Breen, Richard, 2005, "Foundations of a New–Weberian Class Analysis", in E. O. Wright, comp., Approaches to Class Analysis, Cambridge University Press.         [ Links ]

Camberos, Mario, 2009, "Neoliberalismo, globalización y empobrecimiento de la clase media sonorense", Revista de la Universidad de Sonora, núm. 24, pp. 38–41, in <http://www.revistauniversidad.uson.mx/revistas/24–6revi24.pdf>         [ Links ].

Citizen's Assessment of Structural Adjustment (CASA), 2001, "Ajuste y empobrecimiento: veinte años de crisis en México", Executive summary of document drawn up by Coordinating Committee of CASA México, member of SAPRIN NETWORK, based on reports by I. Morales, S. Cruickshank, R. Aguirre Reveles, M. López, M. Pérez Rocha Loyo, M. Oviedo, C. Cortez Ruiz, M. Unda, J–I Chaboux–Leroux, M. Verduzco and N. Torres.         [ Links ]

Comisión Económica para América Latina (CEPAL), 2002, "Vulnerabilidad sociodemográfica: viejos y nuevos riesgos para comunidades, hogares y personas. Síntesis y conclusiones", Santiago de Chile, CEPAL.         [ Links ]

Cortés, Fernando and Agustín Escobar Lapatí, 2005, "Movilidad social e intergeneracional en el México urbano", Revista de la CEPAL núm. 85, April 2005, Santiago de Chile, CEPAL.         [ Links ]

Didou Aupetit, Sylvie, 2008, "Presentación. Movilidades académicas y profesionales en América Latina: entre la ignorancia y la polémica", Educación superior, vol. 37, no. 148, pp. 71–85, poner el sitio.         [ Links ]

Escobar Latapí, Agustín, 1995, "Movilidad, reestructuración y clase social en México: el caso de Guadalajara", Estudios sociológicos, vol. 13, núm. 38, May–August 1995, Mexico, El Colegio de México, pp. 231–259.         [ Links ]

Esteinou, Rosario, 1999, "Fragilidad y recomposición de las relaciones familiares. A manera de introducción", Desacatos, Fall, núm. 002, Mexico, Centro de Investigaciones y Estudios Superiores en Antropología Social.         [ Links ]

Galassi, Gabriela, 2010a, "Hacia la matriz vulnerabilidad–clases sociales: Enfoques de Rubén Kaztman y Susana Torrado", in Leandro González, comp., Lecturas sobre vulnerabilidad y desigualdad social, Córdoba, Argentina, Centro de Estudios Avanzados/Universidad Nacional de Córdoba/CONICET/Copiar.         [ Links ]

Galassi, Gabriela [M.A. thesis], 2010b, "Clase media y nuevos pobres en Argentina con la crisis de 2001: Estudio exploratorio a partir de Censos de Población y Encuestas de Hogares", Córdoba, Argentina, CEA/UNC.         [ Links ]

Gilbert, Dennos, 2005, "La clase media mexicana y la crisis económica de mediados de los años noventa", Estudios Sociológicos núm. 68, México, El Colegio de México, pp. 465–483.         [ Links ]

González Bombal, María Inés, and Maristella Svampa, 2001, "Movilidad social ascendente y descendente en las clases medias argentinas: un estudio comparativo", serie Documentos de Trabajo núm. 3, Buenos Aires, SIEMPRO.         [ Links ]

González, Leandro, 2010, "Orientaciones de lectura sobre vulnerabilidad social", in Leandro González, comp., Lecturas sobre vulnerabilidad y desigualdad social, Córdoba, Argentina, Centro de Estudios Avanzados/Universidad Nacional de Córdoba/CONICET/Copiar.         [ Links ]

Hoffman, Kelly, and Miguel Ángel Centeno, 2003, "The Lopsided Continent: Inequality in Latin America," Annual Review of Sociology, núm. 29, pp. 363–390.         [ Links ]

Icazuriaga Montes, Carmen, 1994, "Desarrollo urbano y formas de vida de la clase media en la ciudad de Querétaro", Estudios demográficos y urbanos, vol. 9, núm. 2 (26), pp 439–456.         [ Links ]

Kaztman, Rubén, coord., 1999, "Activos y estructuras de oportunidades. Estudios sobre las raíces de la vulnerabilidad social en Uruguay", Montevideo, CEPAL.         [ Links ]

Kaztman, Rubén [ponencia], 2000, "Notas sobre la medición de la vulnerabilidad social", 5° MECOVI Workshop, Mexico City.         [ Links ]

Kaztman, Rubén, and Fernando Filgueira, 2006, "Las normas como bien público y como bien privado: reflexiones en las fronteras del enfoque AVEO", IPES series of working documents, Colección de Aportes Conceptuales núm. 4, Uruguay, Programa IPES/Facultad de Ciencias Humanas/Universidad Católica del Uruguay.         [ Links ]

Kessler, Gabriel, 2002, "Empobrecimiento y fragmentación de la clase media argentina", Proposiciones núm. 34.         [ Links ]

López Santillán, Ricardo, 2007, "Lo bonito, limpio y seguro: usos del espacio de la Ciudad de México por una fracción de clase media", Alteridades, vol. 17, núm. 34, July–December 2007, Mexico, Universidad Autónoma Metropolitana Iztapalapa, pp. 9–25.         [ Links ]

Martínez Valle, Adolfo, 2009, "Social Class, Marginality and Self–Assessed Health: A Cross–Sectional Analysis of the Health Gradient in Mexico," International Journal of Equity in Health.         [ Links ]

Minujin, Alberto et al, 1992, Cuesta abajo. Los nuevos pobres: efectos de la crisis en la sociedad argentina, Buenos Aires, Losada.         [ Links ]

Minujin, Alberto, and Eduardo Anguita, 2004, La clase media, seducida y abandonada, Buenos Aires, Edhasa.         [ Links ]

Mora y Araujo, Manuel, 2002, "La estructura social de la Argentina: Evidencias y conjeturas acerca de la estratificación actual", serie Políticas Sociales núm. 59, Santiago de Chile, CEPAL, División de Desarrollo Social.         [ Links ]

Moreno Crossley, Juan, 2008, "El concepto de vulnerabilidad social en el debate en torno a la desigualdad: problemas, alcances y perspectivas", Working Paper Series núm. 9, Miami, Center for Latin American Studies/University of Miami.         [ Links ]

Moser, Caroline, and Andrew Felton, 2007, "The Construction of an Asset Index Measuring Asset Accumulation in Ecuador", Chronic Poverty Research Center, Working Paper núm. 87.         [ Links ]

Pizarro, Roberto, 2001, "La vulnerabilidad social y sus desafíos: una mirada desde América Latina", Serie de Estudios Estadísticos y Prospectivos núm. 6, Santiago de Chile, CEPAL–ECLAC, División de Estadísticas y Proyecciones Económicas.         [ Links ]

Rodríguez Ramírez, Héctor, 2005, Propuesta metodológica para la medición de la pobrera en Nuevo León, México, Instituto Tecnológico y de Estudios Superiores de Monterrey.         [ Links ]

Rodríguez Vignoli, Jorge, 2000, "Vulnerabilidad demográfica: una faceta de las desventajas sociales", Población y Desarrollo series, núm. 5, Santiago de Chile, CELADE/FNUAP.         [ Links ]

Sánchez Saldaña, Kim, 1999, "Sectores medios en México, una bibliografía comentada", Nueva Antropología, vol. XVI, México, Universidad Nacional Autónoma de México, pp. 57–67.         [ Links ]

Sémbler, Camilo, 2006, "Estratificación social y clases sociales. Una revisión analítica de los sectores medios", Políticas Sociales series, núm. 125, Santiago de Chile, CEPAL.         [ Links ]

Torrado, Susana, 1992, Estructura social de la Argentina. 1945–1983, Argentina, Ediciones de la Flor.         [ Links ]

Torrado, Susana, 2007, "Estrategias de desarrollo, estructura social y movilidad", in Susana Torrado, comp., Poblacióny bienestar en la Argentina del primero al segundo centenario. Una historia social del siglo XX, vol I, Buenos Aires, Edhasa, pp. 31–67.         [ Links ]

Zamorano Villarreal, Claudia, 2007a, "Presentación", Alteridades, vol. 17, núm. 34, July–December 2007, pp. 3–6.         [ Links ]

Zamorano Villarreal, Claudia, 2007b, "Los hijos de la modernidad: movilidad social, vivienda y producción del espacio en la ciudad de México", Alteridades, vol. 17, núm. 34, July–December 2007, pp. 75–91.         [ Links ]

 

NOTES

¹ The issue of social mobility reappeared in Mexican social sciences in the mid–1990s (Zamorano Villarreal, 2007b).

² There is no consensus on the definition of the concept of "social class." This paper adopts an operative criterion, based on the concepts of Susana Torrado (1992). It will discuss the upper, middle and working classes in line with the terminology used by the author, referring to the class at the top, center and base of the social pyramid, even though it is a well–known fact that most of the population are located in the tertiary sector, since there are increasingly few workers, strictly speaking. Terms such as "social sectors" (or referring to the upper, middle or working classes) will be used as synonyms of "social classes."

³ In Argentina, this group includes unemployed householders who have not had previous work experience. In the total universe of the active population, these cases account for 0 percent in 1998 and 0.4 percent in 2003.

⁴ Despite the fact that in Mexico, Cortés and Escobar (2005) have drawn up a class map in which professionals are located in Class 1 (corresponding to the first one on the social scale), this paper considered professionals within the middle class, considering that the expansion of the opportunity structure in occupational terms makes it possible to distinguish between individuals with a managerial function and those who work as professionals in a specific function. The classification adopted for Mexico is intended to maintain its comparability with the other country analyzed, Argentina.

⁵ The results for both Argentina and Mexico are consistent with those found in other studies on the class structure of the two countries during the period considered (Torrado, 2007; Mora y Araujo, 2002; Martinez Valle, 2009; Camberos, 2009; Gilbert, 2005).

⁶ It should be noted that the proliferation of single–parent families in the working class may be obscured by the fact that surveys sometimes record this type of families as extended families, when the single–parent family joins another family (for example, when there is a divorce and the mother and children move in with the grandparents). This phenomenon may occur due to the lack of resources, particularly in sectors with the lowest income.

⁷ One factor not explored here but studied in Argentina in Galassi (2010b), is that the middle class fertility pattern, like that of the working class, has a dilated peak, unlike the upper class, where the peak is late. This indicates the presence of a delay in the incorporation into the second demographic transition, also described as a feature of social vulnerability (CEPAL, 2002).

⁸ There are three line of poverty, defining three types of poverty: food poverty, skills poverty and assets poverty.

⁹ Unsuitable housing is that which is not a house or apartment and built with unsuitable materials or has no water, electricity or bathroom, or has a latrine or simply a cesspool.

¹⁰ The high percentage of unsuitable middle class housing in Mexico should be qualified by the rigidity of the definition of unsuitable housing adopted. Icazuriaga (1994) find that although middle–class housing in Querétaro is usually old, it is built from sturdy materials. For his part, Zamorano (2007a) explains that middle–class areas of residence are not homogeneous but often combined with popular housing. Moreover, middle class families often buy housing on the semi–urbanized periphery, hoping that it will increase in value (Zamorano, 2007b).

¹¹ The share of middle class homeowners in Mexico is lower than that reported by Icazuriaga (1994) for the city of Querétaro.

¹² Although the difference in the proportion of home owners and unsuitable housing in Mexico among the upper class is not significantly different from zero, this may be due to the small sample size for this class (235 households in 1994 and 255 in 1996).

¹³ The results using Rodriguez's (2005) proposal are not comparable with those obtained using the INDEC methodology.

¹⁴ Although this is not advisable because of the evidence provided in this article, the creation of a group of new poor, particularly (but not exclusively) comprising former members of the middle class is a recognized phenomenon in Latin America because of the literature on the subject (Minujin and Anguita, 2004; Kessler, 2002; Kessler and Di Virgilio, 2008).

¹⁵ The slight increase in activity rates in Mexico among all social classes may be due to the increase in the share of secondary workers (women and youth), which is a common phenomenon during periods of crisis, which has been pointed out by Escobar (1995) and Esteinou (1999) in Mexico.