Social Mobility Research Paper

Conventionally, different forms of mobility are studied: horizontal, upward, and downward mobility refer to moves among positions at the same level of advantage, moves up and down along positions ordered in a hierarchy of advantage. Intergenerational mobility compares individuals’ positions in adult life with the positions of their families of origin. Intragenerational (or career) mobility studies position sequences within individuals’ own generations, for instance the moves between the successive jobs in working life. Other distinctions frequently made contrast collective vs. individual mobility, structural vs. exchange mobility or absolute rates vs. relative rates of mobility.

1. Research Questions and Traditions of Study

Classical authors have primarily seen social mobility as a mechanism of class formation or status group formation. The basic assumption is that classes, status groups, or other collectivities with some form of sociocultural or sociopolitical identity can only emerge and potentially provide a basis for collective action on condition that they achieve a sufficient degree of ‘demographic identity’ (Goldthorpe) in their membership, that is, when they include a kernel of members homogeneous in terms of their social origin and stable over time (families across generations or individuals through their working lives). Along these lines, Marx—whose main concern with mobility is otherwise with collective downward mobility of small proprietors and other middle classes into the proletariat— considers individual mobility as a crucial determinant of working class organization in his discussion of the North American class structure. Contrary to Europe, classes in the USA, he argues (much as Sombart does later) , ‘have not yet become fixed, but continually change and interchange their elements in a constant flux’ (Marx [1852] 1958, p. 225), with the effect of weakening class formation in the working class. Also stressing the connection to social mobility, Max Weber even defines a social class as an ensemble of those class situations among which high interchange of individuals typically takes place, while different classes are distinguished from each other by a lack of mobility between them. Weber’s mobility criterion has been used to empirically establish class boundaries by later researchers.

In its later development, research on social mobility has largely turned to the study of social opportunity, aiming to determine the relative fluidity or rigidity of the stratification system concerning access to positions of varying advantage. Research has in particular concentrated on identifying the degree to which individuals’ prospects in life are determined by the social conditions of their family of origin and on specifying the individual, institutional, and societal factors responsible for it. The more strongly the positions persons hold are conditioned by their social provenance, the more rigid and closed is a stratification system; the greater the mobility between social origin and a persons’ own position, the more open or fluid that system is.

Issues related to social mobility are often the object of normative and political discourse and are then often interpreted ideologically, for instance as ways of legitimizing social inequalities. In this vein, variants of liberal political philosophy from John Stuart Mill onwards regard unequal conditions as legitimate when they result from (fair) competition or when unequal chances of access to privileged positions are due to unequal achievements. In its extreme form, the functionalist theory of stratification holds unequal rewards to be deserved, as they are assumed to be distributed according to the contributions made to the efficient functioning (and ultimately survival) of society and are thought to be necessary to attract the most able and hard-working individuals to the most important positions (see Goldthorpe 1996 for a forceful critique of meritocratic and functionalist assumptions). Though at times influenced by normative preferences, research in the field is analytical and empirical. Its results may, however, serve to examine to what degree and for what reasons social reality conforms to or falls short of particular normative or political goals.

The investigation of social mobility is mainly pursued by way of observation and comparison of national populations. Comparative designs are essential to determine the extent to which fluidity or rigidity of the stratification system is related to other characteristics of societies such as their degree of modernity, their economic and political order (e.g. capitalist market economies vs. state command economies), particular institutions or policies. Research has especially focused on the evolution of social mobility in line with the development of industrial society. While in his early work Sorokin (1927) expected ‘trendless fluctuation’ rather than a development in a specific direction, later researchers have assumed a historical increase in mobility with the transition from preindustrial to industrial society, either as a precondition for the take-off of industrial development or as its consequence (Lipset and Zetterberg 1959). Others assume a steady increase in mobility as industrialism advances, and an increasing uniformity and convergence of social structures and social mobility in industrial societies, due to standardizing tendencies inherent in advanced technologies and other elements of industrialism. Finally, exceptionally high rates of mobility have been postulated for specific societies, particularly those not burdened by the legacy of the European feudal past or its class-based stratification system, such as the USA, Australia, or Japan. To test such hypotheses, numerous datasets in many different societies have been created, primarily through large-scale population surveys, but also (notably in studies on earlier periods) by using suitable information from official records. As members of social elites tend to be represented in extremely small numbers only in population surveys, mobility into elite positions is generally not covered.

For the purposes of structured summary, much of the sociological literature can be represented as a debate between two different paradigms and research traditions. To study mobility, both essentially rely on work positions, that is, the (occupational) jobs individuals hold to earn a living. The status attainment approach, on the one side, sees the principle property of these jobs in terms of their respective location in a finely graduated continuum of rewards related to the position and conveyed to their incumbents. The continuum of rewards is generally conceived in terms of social prestige or socioeconomic status and measured by corresponding scales.

The class mobility approach, on the other side, conceives the (structure of) positions among which mobility can occur as discrete classes which differ from each other by properties of a qualitatively different nature rather than only along a vertical dimension. In earlier studies, categories consisted of rather broad groupings of occupations defined in an ad hoc manner (sometimes as broadly as agricultural manual nonmanual). Newer class schemas are grounded in theoretically comprehensive conceptions of the class structure. In their now widely accepted class schema, Erikson and Goldthorpe (1992) draw on the character of employment relations as core criteria to identify different classes, distinguishing first between classes whose incumbents buy the labor of others (employer), sell their own labor (employee) or do neither (selfemployed); among employees, further distinctions are made between different classes of (manual) workers, routine nonmanual employees, and of professionals, administrators, and managers in the so-called service class or salariat, the criterion being the extent to which employment relations are regulated according to a labor contract or a service relationship. Related to the differing conception of the main properties of the positions among which mobility is seen to occur, the two traditions also differ in their theoretical understanding of the mobility process, the methods used, and the main conclusions arrived at.

2. The Process of Status Attainment

Research pursued in the status attainment framework is often—though not necessarily—embedded in a more general (liberal) theory of industrialism and shares with it assumptions about the impacts of industrialization on other aspects of social structure, notably social mobility. Industrialization is understood as a pervasive force that sooner or later affects all national societies across the globe. In this process, it is argued, the order of stratification in preindustrial societies in which ascription by family, kin, ethnic ties, gender or status group affiliation determines the position of individuals in society, is increasingly replaced by a new order in which achievement based on individual abilities, qualifications, and work accomplishments play the decisive role in social selection (Blau and Duncan 1967, Treiman 1970). A fundamental trend toward rationalism and universalism is seen to unfold. To survive in markets actors will be forced by the technological sophistication of production processes and by growing market competition to increasingly allocate individuals ‘to positions according to what they have shown they can do, rather than according to their social provenance’ (Goldthorpe 1996, p. 256). Notably formal educational qualifications are expected to increasingly determine the job level (successively) reached in bureaucratic work organizations. To enhance the development, states expand educational services and subsidize the costs of education, thus providing easier access to education for children in less fortunate economic conditions. As industrialization progresses, social placement of individuals will then be conditioned less by their social origin, and intergenerational social mobility between the positions of parents and those of their children is expected to increase.

These general hypotheses lead to an understanding of social mobility as a process of successive status attainment by individuals, in which later attainment depends on earlier attainment. In line with the conception of the occupational structure as a graded hierarchy, this process is studied with variants of linear regression models (often path or structural equations models) to determine the extent to which the social status or prestige level of a person’s job at a given point in his or her career depends on earlier job achievements, educational qualifications, and directly or indirectly on parents’ status. In Blau and Duncan’s (1967) seminal study The American Occupational Structure, education is identified as the main determinant of individuals’ occupational careers and as the core link between social origins and attained position. Parental education and occupational status also affect children’s occupational status or incomes, but their effects are weaker than those of education and are mainly mediated through education. One farreaching conclusion has been that under such conditions of status attainment—powered by individual achievement—the very basis for class solidarity and class action disappears. Extensions of the model have studied how other family conditions, personal abilities (intelligence), sociopsychological factors (parental encouragement and support, educational and occupational aspirations, peer contacts) or characteristics of schools intervene in these processes and contribute to generating them (Sewell and Hauser 1975). In evidence from a number of societies at different levels of industrial development, Treiman and Yip (1989) find confirmed the expectation that status attainment becomes increasingly achievement-based as industrialization unfolds.

The status attainment paradigm has come in for criticism on a number of counts. Its representation of the attainment process is individualistic and onesidedly supply-driven, concentrating as it does on the way familial resources and individual human capital affect occupational and other outcomes. But the transformation of individual resources into outcomes depends on existing opportunities. The latter are constrained both by the distribution of available positions and by the competition of those contending for the same positions with similar or better resources. Further, the unidimensional representation of positions in terms of status or prestige obscures the way in which opportunities are bound up with structural transformations such as the shifts in forms of employment and between sectors, industries, and occupations. Groups like farmers, skilled industrial workers, small shopkeepers, and service workers hold positions that are heterogeneous in terms of their sociostructural location and affected in different ways by particular state policies and structural transformations. However, in terms of social status or prestige, they all are similarly rated. In the hierarchical framework it thus appears difficult, if not impossible, to trace mobility processes to their causal conditions.

3. Intergenerational Class Mobility and Social Fluidity

Proponents of the class framework have attempted to overcome these shortcomings by reorienting the study of mobility to analyze the moves of individuals between a set of social class positions identified within a class structure. These moves are represented in mobility matrices, which, in their elementary form, are bivariate frequency tables with one dimension representing the set of origin classes, the other dimension the set of destination classes, and the cells of the matrix containing the frequencies of transitions observed between origin and destination. While origin and destination can be any two successive positions (for instance in working life), intergenerational comparisons between the class positions of parents and those of their children are of primary interest. Several important strengths result from this reorientation. Compared to hierarchical ratings, the class characteristics allow the derivation of more realistic hypotheses concerning the greater or lesser likelihood of mobility between particular origins and destinations. Structural transformations and changing opportunity structures can be more directly captured with class categories sensitive in this respect. The concomitant development of log-linear models (for early contributions see Goodman 1970, Hauser 1979) has brought crucial improvements in the ability to account for opportunity structures in the statistical analysis of mobility matrices.

In these models the constituent relations in a mobility table are the relative rates of mobility. These are basically a set of odds ratios which measure the relativities existing between different origin classes and the attainment of different destination classes. Odds ratios can be interpreted as measures of the competitive advantages the progeny of one origin class have over those of another origin class in obtaining or avoiding a position in one rather than in another destination class. Odds ratios have the crucial property of being independent of the marginal distributions of the mobility matrix (the relative size of origin classes and destination classes). The latter reflect the macrostructural changes between generations in positions available and can be understood as capturing the opportunity structure. The odds ratios thus express the pattern of association between origin and destination independently of a given opportunity structure. However, as Sørensen (1977) and others have stressed, opportunity structures also depend on how vacant positions available in the market are matched to individual workers, e.g., by creating more or less extended chains of vacancies within and between firms. But still it is the relative rates of mobility—expressed by odds ratios—which are now widely accepted as a proper measure of the degree of fluidity or rigidity of the class structure or the stratification system.

Assessed in terms of these relative rates, class destinations in mature industrial societies continue to depend heavily on class origins. In England or France, for instance, the odds of a son born to a service class father obtaining a service class position rather than one in the unskilled working class is more than 20 times larger than the corresponding odds for an unskilled worker’s son. Mobility is more likely to occur for intermediate classes, for example, routine nonmanual employees, while the likelihood of staying immobile in the parental class are larger in the most and least advantageous classes. Class positions that require the ownership of material or financial capital or land are ‘inherited’ to a clearly larger extent from the parental to the offspring generation than classes not requiring such assets. Short-range mobility occurs more often than long-range mobility. While there is still debate as to whether in the long run social fluidity has increased (most persistently defended by Ganzeboom et al. 1989), most research indicates that there is a high degree of temporal constancy and cross-national commonality in the fluidity pattern of industrial and postindustrial societies.

Explanations of the common characteristics of social fluidity have been proposed in mainly two ways. Researchers emphasizing the hierarchical dimension have stressed the distances (in status or other considered dimensions) between positions as the mobility generating principle (e.g., Hout and Hauser 1992). The larger the vertical distance between positions, the less mobility there will be between them. Sophisticated distance models have consistently shown considerable distance effects, but they cannot entirely explain the existing pattern. Researchers in the class framework have acknowledged the multidimensional nature of the resources that are available to the offspring of various class origins and that can be used for gaining access into classes of varying advantage and desirability.

In their major comparative study of a broad range of industrial societies, Erikson and Goldthorpe (1992) show that alongside the impacts of vertical ‘hierarchy’ the fluidity pattern is markedly shaped by ‘inheritance’ of parental class position (in all classes, but especially in the petty bourgeoisie, among farmers, and in the service class), by the ‘sector’ division between employment in agriculture and outside agriculture, and by special affinities expressing enhanced mobility among particular classes like the various nonmanual classes. To account for the high stability across time and the marked similarity between countries, Goldthorpe (2000) proposes a general explanation in terms of resource-constrained rational action, much in line with the theory suggested by Boudon (1974). Members of the various social classes pursue the same aims, giving priority to avoiding downward mobility and only in the second place seeking to achieve upward mobility. But in pursuing these same goals and priorities, the different amounts and kinds of resources typically available prompt the working and intermediate classes to use different strategies than the service class. The cumulative effect of different starting positions, resources, and strategies is so pertinent that it produces high stability in the resulting outcomes. If inequalities of opportunity are to decline and social fluidity to increase, it is most likely to occur in a context of markedly reduced inequalities of condition. High degrees of cross-national commonality and constancy over time, however, are not to be equated with complete uniformity. Some limited, but nevertheless significant differences are found between societies, and change is also found within some societies across time. Erikson and Goldthorpe thus propose accepting, albeit only in a weak form, the hypothesis originally proposed by Featherman et al. (1975), that ‘basically the same’ relative rates of social mobility exist in all industrial societies with a nuclear family system and a market economy. As changes in time do not seem to form a systematic trend (associated with industrial or some other development), results appear to favor Sorokin’s assumption of ‘trendless fluctuation’. There is also evidence against the hypothesis of exceptionally high fluidity in the ‘new’ societies (USA, Australia) and Japan. Nor does social fluidity systematically vary according to the politicaleconomic order of countries. The former communist Eastern European societies show quite similar fluidity patterns to capitalist societies. Instead, deviations from the core pattern appear to arise more from historically contingent conditions resulting from a nation’s particular economic, social, and political history or from specific institutional arrangements, for instance in the educational system, in labor market segmentation or in other institutions with an impact on specific elements of the mobility pattern. Periodically higher relative rates of mobility have also been shown to result from massive state intervention, as in the case of the long-term equalizing policies by the social democratic governments in Sweden (Erikson and Jonsson 1996) or the strong discriminatory measures pro and contra particular classes in some phases of the communist era in Hungary.

In contrast to relative rates, absolute rates of mobility refer to the observed absolute number of moves that can be read directly from the mobility matrix, and to proportions such as the percentage of children who have a different (percentage mobile) or the same (percentage immobile) class position as their parents. Other often studied absolute rates are outflow proportions (percentages of offspring of a given origin attaining particular destinations) or inflow proportions (percentages of members in a given destination recruited from the various origin classes). Even though absolute rates are more directly observed, theoretically they are conceived to be implied by the more fundamental conditions of social fluidity and by the opportunity structures captured in the marginal distributions. Under a given set of odds ratios, particular marginal distributions imply a unique constellation of absolute rates. (The distinction between relative rates of mobility and absolute rates has replaced the earlier distinction between exchange mobility and structural mobility, which had similar intentions but was statistically and theoretically unsatisfactory.)

In contrast to relative rates, absolute rates vary considerably between nations and across time. Given only little variation in relative rates, it is mainly variation in the opportunity structure that generates this outcome. In a world of economic and occupational change, mobility is induced as workers are pushed or ‘forced’ to leave areas of declining employment and gravitate to those which are expanding. Societies differ in the absolute amount of mobility both among themselves and across time because they vary in the pace of change and because differences in class composition exist even among countries with similar levels of development. The absolute numbers of such moves determine the observable mobility experience of a population. They also strongly affect the composition of class membership by persons intergenerationally stable or recruited from other class origins. Most likely they are of particular importance in processes of class formation. Though these issues are of considerable interest, they have been hardly studied.

Similar relative rates alongside varying absolute rates are also found if different population subgroups are compared. Large differences between men and women, immigrants and natives, or between different races or ethnic groups often exist in absolute rates of mobility into more or less advantageous classes because the groups are constrained to or choose jobs in different segments of the labor market and hence their distribution among classes varies. But the patterns of social fluidity hardly differ between the groups. If they sometimes do, differences tend to be small. As a result of gender-segregated labor markets, for instance, men’s and women’s absolute chances of access into different classes strongly differ, but these inequalities are generally not intensified by men and women being affected differently by their class origins. As a confirming corollary it is found that in times when married women’s labor market participation was low, social fluidity for daughters in the marriage markets has been rather similar to the relative mobility prospects of their brothers in the labor market. Among the highly debated issues concerning gender inequality one important question relates to the proper unit of stratification, class, and mobility analysis. Is it (always) women’s or men’s own position in the labor market that determines their social status or class affiliation, or is it rather the common position of the family unit? As women’s labor force participation has increased and has become more continuous, the most convincing answer is to study the issues separately, to analyze men’s and women’s labor market careers and social mobility separately, likewise their household division of labor and resource utilization, but also the common position they may share as a family unit.

In the figures on mobility or immobility between social origin and eventual destination, whether viewed in absolute or relative terms, the process of intergenerational social reproduction is only reflected in its global outcomes. It results from various intervening processes, notably educational attainment, transition from education to work and mobility between successive jobs in working life within and between firms. These processes have to be—and are—studied in their own right. Of these, the role of education has been scrutinized most intensively. The fact that education is a prerequisite for access to professional and managerial jobs makes it an important channel for upward social mobility, but evidently one which the more advantaged families also make use of to secure favorable positions for the next generation. Further, while education is found to be the main carrier responsible for the hierarchical element in the fluidity pattern, this is much less true—in some cases not true at all—of its role as a mediator of other elements or mechanisms that produce immobility (Ishida et al. 1995). Transfer of capital or business, provision of social contacts, or inculcation of the proper social skills and manners are some of the means prosperous families have to back their children outside school and which can be used to compensate for possible failures in education.

Studies pursued in the class framework so far have not found evidence consistently supporting expectations of a general trend towards increased educationbased or ‘credentialist’ social selection in all industrial or postindustrial societies, neither in the sense of education becoming systematically less dependent on social origins nor in the sense of education consistently becoming a stronger predictor of social destination. As is true for intergenerational social fluidity in general, changes that actually take place in some countries appear to be limited in extent, and the direction they take depends on contingent circumstances specific to these countries.

Bibliography:

1. Blau P M, Duncan O D 1967 The American Occupational Structure. Wiley, New York
2. Boudon R 1974 Education, Opportunity and Social Inequality. Wiley, New York
3. Erikson R, Goldthorpe J H 1992 The Constant Flux: A Study of Class Mobility in Industrial Societies. Clarendon Press, Oxford
4. Erikson R, Jonsson J O (eds.) 1996 Can Education be Equalized? The Swedish Case in Comparati e Perspecti e. Westview, Boulder, CO
5. Featherman D L, Jones L, Hauser R M 1975 Assumptions of social mobility research in the US: The case of cccupational status. Social Science Research 4: 329–60
6. Ganzeboom H G B, Luijkx R, Treiman D J 1989 Intergenerational class mobility in comparative perspective. Research in Social Stratification and Mobility 8: 3–55
7. Goldthorpe J H 1996 Problems of ‘meritocracy.’ In: Erikson R, Jonsson J O (eds.) Can Education be Equalized? The Swedish Case in Comparati e Perspecti e. Westview Press, Boulder, CO, pp. 255–87
8. Goldthorpe J H 2000 On Sociology: Numbers, Narrati es, and the Integration of Research and Theory. Oxford University Press, Oxford
9. Goodman L A 1970 The multiplicative analysis of qualitative data: Interactions among multiple classifications. Journal of the American Statistical Association 65: 226–56
10. Hauser R M 1979 Some exploratory methods for modeling mobility tables and other cross-classified data. Sociological Methodology 10: 141–58
11. Hout M, Hauser R M 1992 Symmetry and hierarchy in social mobility: A methodological analysis of the CASMIN model of class mobility. European Sociological Re iew 8: 239–66
12. Ishida H, Muller W, Ridge J M 1995 Class origin, class destination, and education: A cross-national study of ten industrial nations. American Journal of Sociology 60: 145–93
13. Lipset S M, Zetterberg H L 1959 Social mobility in industrial societies. In: Lipset S M, Bendix R (eds.) Social Mobility in Industrial Societies. University of California Press, Berkeley, CA, pp. 11–75
14. Marx K [1852] 1958 The eighteenth brumaire of Louis Bonaparte. In: Marx K, Engels F (eds.) Selected Works. Foreign Language Publications, Moscow
15. Sewell W H, Hauser R M 1975 Education, Occupation, and Earnings: Achie ement in the Early Career. Academic Press, New York
16. Sørensen A B 1977 The structure of inequality and the process of attainment. American Journal of Sociology 42: 965–78
17. Sorokin P A 1927 Cultural Mobility. Free Press, New York
18. Treiman D J 1970 Industrialization and social stratification. In: Laumann E O (ed.) Social Stratification: Research and Theory for the 1970s. Bobbs Merrill, Indianapolis, IN, pp. 207–34
19. Treiman D J, Yip K-B 1989 Educational and occupational attainment in 21 countries. In: Kohn M L (ed.) Cross-National Research in Sociology. Sage, Newbury Park, CA, pp. 373–94