The term ”social mobility” describes the nature and amount of change in social position over time. In principle, this change can be defined for any social entity. Thus, one can study the ”collective mobility” of classes, ethnic groups, or entire nations in terms of, for example, average health status, literacy, education, or gross domestic product per capita. More commonly, the term is used in connection with the movement of individuals or families. However, even though social mobility typically is defined with respect to micro units of society, the pattern of mobility across those units generally is considered a core characteristic of a society’s social structure, and the study of this mobility generally is recognized as a fundamental area of macro-level sociology.
Social mobility typically is conceptualized in terms of the quantity of movement and the distribution of its direction and distance. The different rates that together constitute the mobility structure of a society is highly complex, however, for several reasons. First, societies have more than one dimension along which mobility can occur. Thus, one can speak of occupational mobility, social class mobility, educational mobility, job mobility, income mobility, wealth mobility, and so on. In principle, one also can use the term ”social mobility” to describe movement among nonhierarchical social statuses, such as religious affiliation mobility and geographic mobility or mobility across categories that describe attitudes, belief systems, life styles, and the like. The dominant use of the term in the literature, however, concerns mobility along a social hierarchy that defines a dimension of social inequality in a society. Second, even with respect to a single hierarchy, the mobility structure is not easy to summarize. A different rate of mobility can be calculated with respect to each combination of origin and destination position along the social hierarchy in question. Empirically, it may be possible to summarize this collection of rates accurately in terms of a function of the social distance between origin and destination or in terms of specific relationships between the origin and destination categories. In general, however, an accurate summary cannot be expressed in terms of a single number. Thus, for each social hierarchy, there is not a single rate of social mobility but a core set of rates that, taken together, can be termed the structure of mobility with respect to the particular hierarchical dimension.
Social mobility is an important issue in sociology for several reasons. For one thing, it is relevant to social equity. Philosophical and moral evaluations of social inequality often depend not only on the level of inequality in a society but also on the extent to which individuals or families can leave disadvantaged states during their lifetimes or across generations. Social mobility is also an important explanatory factor in social theory. The basic stratification variables affect a wide variety of social outcomes and behaviors, but these effects accumulate over time; social mobility therefore affects outcomes by changing the states and durations of these key explanatory variables. The societal rate of mobility also may have macro-level consequences. An early conjecture in this area appears in the work of Werner Sombart, who argued that the failure of early twentieth century socialist parties in the United States stemmed in part from the high rate of American social mobility, which prevented the formation of strong class identification.
The longest-standing tradition in sociological mobility research concerns mobility in occupational groupings or social classes. Much of this work has used ”mobility tables” (cross-classifications of origin by destination position) to study ”intergenerational mobility,” that is, the extent to which the social position of adults differs from that of their parents. Another large body of work has focused on ”intragenerational mobility,” or the mobility experienced by individuals or families over the course of their adult lives. Because male labor force participation generally has been higher and more persistent than female participation and because of the somewhat controversial presumption that the status of a family derives from the status of the male breadwinner, for many years these studies focused on intergenerational mobility between fathers and sons, although more recent literature has examined the structure of mobility for women as well.
An important question in intergenerational mobility research is whether overall rates of inter-generational social mobility differ by country. Earlier in the century, scholars hypothesized that the United States had especially high rates of mobility, and some argued that those rates were a consequence of the American meritocratic value system. More recently, it became clear that the primary factors in cross-national differences in mobility rates are structural, not cultural. Differences in so-called structural mobility across countries arise from the extent to which the distribution of positions for sons or daughters differs from the distribution of positions for their fathers. Changes in this distribution across generations (as well as more subtle factors such as class differences in fertility, death rates, and migration rates) necessarily produce intergenerational social mobility. Countries whose occupational distribution is changing rapidly (high rates of structural change) therefore have greater levels of mobility than do countries whose occupational distribution is changing slowly.
Not all social mobility occurs as a result of structural change. The component of social mobility that occurs beyond the amount produced by structural change is typically called circulation mobility, exchange mobility, or relative mobility. The Featherman, Jones, Hauser (FJH) hypothesis of the mid-1970s asserts that cross-national and historical differences in social mobility are accounted for almost completely by differences in levels of structural mobility. According to the strong form of this hypothesis, once structural mobility is taken into account, the pattern of relative mobility chances is invariant over time and across countries. This pattern has three principal features: (1) relatively high immobility at the top and bottom of the hierarchy, (2) higher levels of short-range mobility than long-range mobility (moves from the top to the bottom or from the bottom to the top of the hierarchy are especially rare), and (3) a relatively small impact of origins on destinations in the middle of the hierarchy.
More recent research has determined that even though the weak form of the FJH hypothesis (overall mobility differences are due largely to differences in structural mobility) is supported by the data, the strong form (invariance of relative mobility chances) appears to be false. However, further progress on this issue has been elusive. In particular, the question of whether cross-national differences in relative mobility chances are the subject of such complex national historical differences that they are idiosyncratic or whether they are the product of a more parsimonious set of structural forces (e.g., the extent to which the political system is democratic, the level of modernization, and the level of social inequality) remains to be answered.
Another continuing challenge in mobility research concerns conceptualization and measurement of the component of mobility that is due to structural change. The specification of this causal force in terms of differences in the distribution of positions of fathers and their adult children is problematic for subtle but important reasons. Such an identification assumes that the observed destination distribution is caused by forces (such as technological change) that are not affected by (and therefore are a legitimate cause of) the observed mobility process. This amounts to assuming that the observed destination distribution constitutes a rigid supply constraint, a set of preexisting empty vacancies that are filled by the movement of sample members with respect to their origin positions. This assumption is never perfectly true. If the ”supply constraint” is not rigid (and it is unlikely to be so), the observed distribution of destination positions (which by definition represents a summing up of the mobility outcomes for a particular statistical sample) is a consequence of the mobility process as well as of the ”structural forces” that constrain the character of this destination distribution. It therefore cannot be taken to be a pure cause of social mobility. The logic of structural mobility becomes especially problematic when subgroups of the population are studied in this fashion. For example, if the distribution of women’s occupations shifts toward high-status occupations relative to the total occupational distribution, it is problematic to argue that the relative improvement of women’s destinations is a ”cause” of women’s higher levels of social mobility as opposed to being a consequence of that mobility. This problem, which is sometimes referred to as the ”reflection” problem, has not had a satisfactory solution.
Although structural change is a major part of the explanation for overall levels of intergenerational social mobility in a society, it cannot explain differences in the likelihood that particular individuals will be upwardly or downwardly mobile. The prevailing pattern of circulation mobility that was noted above (relatively high levels of immobility at the top and bottom, the predominance of short-range over long-range mobility, etc.) implies that class of origin is a significant predictor of the types of mobility that do occur. However, an explanation for destination positions that relied solely on the status of origin would be unsatisfactory in two respects: First, the predictive power of social origins by itself is relatively weak; second, the explanation does not indicate how and why social origins matter.
Efforts to redress these deficiencies stem largely from the publication of Blau and Duncan’s The American Occupational Structure (1967). A major goal of that work was to understand whether the educational system operated primarily as a device that transmitted the status of parents to their children or as an engine of social mobility that freed children from the effects of the status of their parents. To accomplish that goal, Blau and Duncan developed what has come to be known as the status attainment model. Their approach to the study of mobility assumed a dominant metric to social hierarchy: the socioeconomic status of occupations. Their research showed that, at least for men (Blau and Duncan did not study the mobility of women), education was a more important determinant of a son’s adult socioeconomic status than were his socioeconomic origins. Furthermore, while educational attainment was strongly influenced by socioeconomic origin, most of the individual-level variation in educational attainment was not explained by socioeconomic origin. Those authors also showed that most of the effect of socioeconomic origins on outcomes was indirect, derived from the effect of those origins on education. Finally, the effect of education on occupational attainment regardless of social background was much larger in the United States than was the direct effect of father’s socioeconomic background (regardless of the son’s education). These findings led many to interpret Blau and Duncan’s research to mean that the United States more closely approximated an ”achievement” than an ”ascription” society, although others pointed to the still large (even if not decisive) disadvantage arising from low socioeconomic origin along with the disadvantages associated with being a first-generation immigrant, an African-American, or a woman as constituting important qualifications to such a generalization.
The Blau and Duncan approach essentially divided the intergenerational mobility process into three segments. The first segment concerned the process of educational attainment, the second concerned the transition from school to work, and the third concerned the ”intragenerational” mobility that occurs over the working life. Leaving aside the powerful but difficult to specify force of structural change, this division may offer the best possibility for understanding the mechanisms that lie behind intergenerational social mobility as well as identifying possible policy interventions and shedding light on three processes that have great importance in their own right. Each of these processes calls attention to specific institutions (in particular, the educational system and the labor market) that facilitate, limit, or channel social mobility. The focus on how institutional forces constrain the impact of individual resources on individual outcomes sometimes is referred to as the ”fourth generation” of social mobility research (with early mobility studies being the first generation, the status attainment tradition being the second, and statistically sophisticated analyses of mobility tables being the third).
A large body of literature has grown around each of these components of the intergenerational mobility process. With respect to education, scholars have conceptualized the educational career as a set of transitions to successively higher grades and have asked whether family background has the same influence at each grade level of this transition process. Results for the United States and several other countries suggest that the effects of family background decline at higher-grade transitions, though these findings are controversial. Assuming that the decline is real, some scholars have argued that the historical raising of the minimum school-leaving age should have reduced the impact of family of origin on outcomes over time.
Again, however, while there is some evidence that the effects of family background have declined during the twentieth century and that these declines are caused by the expansion of education, empirical studies have failed to confirm this conjecture decisively.
A second major focus in the literature concerns the reasons why socioeconomic background is associated with educational performance. It has been appreciated since Sewell and associates developed the ”Wisconsin model” in the early 1970s that there is a social psychological component to mobility in which family status is related to parental expectations for the child. In combination with grades in school, peer group influences, and teachers’ expectations, this shapes a student’s educational and occupational aspirations. More recent work has reconceptualized these family advantages or disadvantages in terms of cultural resources (”cultural capital”), which sometimes are specified as a family’s participation in ”high-cultural” activities (exposure to art museums, opera, theater, dance, etc.); in other studies, they are defined more broadly (and vaguely) as encompassing all the cultural advantages a family may possess that affect a child’s ability to do well in school. Other recent literature focuses on ”social capital,” which sometimes is interpreted to mean the level and quality of interaction parents have with their children and at other times is interpreted to refer to the resources embedded in the parents’ social networks that could in principle influence a child’s outcomes. A third, rather controversial focus of attention in recent years concerns possible links between socioeconomic status and genes and the extent to which intergenerational correlations among status variables (particularly educational outcomes) indicate the presence of a genetic force. A fourth focus concerns the specific consequences of low income on children’s development and later socioeconomic outcomes. A fifth focus concerns the extent to which the characteristics of schools, neighborhoods, and communities can mute or exaggerate the impact of family characteristics on educational outcomes.
The second mobility component is the transition from school to work. A large body of literature focuses specifically on aspects of this transition, including variation in the extent to which the diplomas, degrees, and advanced degrees provided by schools are linked by law or custom to specific occupational careers; the extent to which credentials are standardized in a country; the extent to which the supply of those credentials is controlled by schools in light of estimated demand; and the extent to which students who graduate with these diplomas or degrees are provided with knowledge of the relevant job market. Many policy concerns in the United States focus on those who leave school before the tertiary level and the extent to which they are provided with a mix of academic and vocational skills and credentials that is valuable on the job market. Vocational education in particular is organized quite differently across industrialized societies, and in recent years comparative research on this transition has accelerated.
The third component concerns intragenerational mobility over the life course. This research has taken different forms. The Blau and Duncan approach largely emphasized the mean or typical pattern of life-course development as a function of education, first job, and father’s occupation. In this form, the question of mobility is reduced to a question about the average status ”return” to the resources an individual possesses on first entering the labor market. Although this approach is informative about the typical level of status advancement during the work career as a function of origin conditions, it suffers from two deficiencies: First, it does not explain how education and the first job lead to the current job; second, it does not provide an explanation of the frequency or consequences of deviations from the typical amount of status advancement during the work career.
An understanding of the full distribution of outcomes (i.e., both upward and downward career mobility) is made possible through the use of the ”mobility table” approach that has been applied to the study of intergenerational mobility between the status of the father and the status of the son or daughter. The prevalent approach in recent sociology, however, has been more institutional. One line of work has focused on structural linkages between jobs in particular occupational or organizational labor markets. This work has addressed the implications of entering these ”internal labor markets” for subsequent career advancement, with an important subset of it directed at questions about whether these institutional mechanisms reproduce, enhance, or mute racial or gender differences. Because these job linkages generally are not expressed in terms of abstract hierarchical measures such as class and socioeconomic status, studies of these organized labor markets frequently have turned away from the earlier focus on class or status and toward more concrete reward variables such as earnings and job level within an organizational hierarchy. Jobs outside of organized hierarchies that lacked other forms of institutional protection (such as professional licensing requirements) were characterized as ”open” or (if low-quality) ”secondary” labor market jobs. For several years, sociologists hoped that a parsimonious set of labor market ”boundaries” could be operationalized and used to explain labor market outcomes in terms of labor market segment early in one’s career. The promise of this ”segmented labor market” approach to career mobility has faded, however. It is now recognized that the boundary between unstable, low-paying jobs in what once was commonly referred to as the ”secondary labor market” and ”internal labor market” jobs is by no means impermeable, especially in the early years of the adult life course. The segmented labor market approach has been undermined further by the appreciation of the numerically high levels of mobility (including involuntary mobility) out of corporate jobs, often as a result of plant closings and corporate restructuring. Literature on ”displaced workers” that has developed largely in labor economics rather than sociology has attempted to quantify the short-term and medium-term career consequences ofjob displacement (the literature shows only transitory effects on employment but more durable effects on earnings). It can be assumed that job displacement is a principal mechanism by which structural change produces short-term and longer-term intragenerational (and ultimately intergenerational) occupational mobility. However, sociologists have only begun to explore the connection between job displacement and the structural mobility observed in mobility tables.
A separate body of literature has addressed the intragenerational mobility of people who at one time or another in their lives are poor. Aside from questions about the intergenerational transmission of poverty, much of this literature has focused on whether poverty is a permanent or transitory status. It has been recognized that most poverty in the United States is transient, although an important fraction of the poor remain poor for long periods, while many who escape poverty have a relatively high probability of returning to poverty in the future. Much of this literature addresses the factors that influence rates of entry into and exit from poverty.
Poverty studies use a measure of income, especially income in relation to needs, rather than class or status as the basic measure of position. They typically make the family the relevant unit of measurement because it is family income, not individual earnings or status, that most directly determines poverty status. They also direct attention to the facts that income mobility is a household, not an individual-level, concept; that income mobility can be generated by labor market events involving one’s partner as well as oneself; that public transfers can be an important source of income and can play a significant role in determining levels of income mobility; and that changes in household composition (including marriage, cohabitation, and union dissolution) can strongly influence income mobility.
More recent mobility literature has focused as much attention on instability as on stability (or stable ”career advancement”) over time. This emphasis raises important questions about an important presupposition underlying the sociological framework for mobility studies: that it was meaningful to conceptualize the socioeconomic status of the family of origin as a stable point and the ”current” status of the adult son or daughter as a ”realized” socioeconomic status that could be compared with the point of origin. Early studies were forced by the limitations of data to use parental status at a single point in time (e.g., the point at which the respondent was 16 years old) as the measurement of family status over the duration of childhood. The growing availability of multigenerational panel data that provide information about the possibly changing status of the family of origin during childhood has made it possible for scholars to study how temporal variations in the status of parents affect the process of intergenerational transmission. These more extensive data on the lives of parents and their grown-up children are allowing scholars to study intergenerational and intragenerational mobility with respect to statuses such as income, wealth, and poverty, which are perhaps more volatile than are occupational status and class position.
Questions about racial and/or ethnic or gender differences in mobility have largely been subordinated to gender and racial and/or ethnic inequality and changes in levels in inequality over time. In other words, the focus has been more on the collective mobility of these groups with respect to white males than on whether the structure of individual-level mobility within groups defined by race or gender is different from the structure of mobility for white males. This literature has perhaps paid more attention to economic outcomes than to class or status outcomes. Research on gender inequality in particular has avoided the use of status metrics or broad occupational groups, which understate the gender inequality that is visible in earnings. The issue of mobility still plays an important role in this literature because of the possible role of mobility processes in explaining how gender or racial and/or ethnic inequality comes about. In the case of women, the literature has focused on why women take a different mix of academic subjects than men do and on how gender differences in the transition from school to work and in career mobility produce differences in the average earnings of women and men over the life course. In this literature, the pattern and quantity of work experience and the different distribution of men and women across jobs and occupations (”sex segregation”) have been the issues of greatest interest. In contrast, the role of specifically intergenerational processes and their possible impact on gender inequality or on the well-documented declining level of gender inequality in earnings has received less attention.
The issue of race is in some respects parallel to that of gender but has its own unique features. Women and men grow up in the same families, while whites and nonwhites grow up in different families, and these differences involve socioeconomic factors as well as race or ethnicity per se. Furthermore, racial and/or ethnic segregation by educational major, job, or occupation has not been as extreme as gender segregation in recent years. However, just as research has shown that the effects of race on income declined over the postwar years (though the trend has stalled and perhaps reversed since about 1980), it also has shown that the direct effects of race on socioeconomic attainment have declined, at least through the late 1980s. African-Americans experience no disadvantage at all in educational attainment because of race per se (though they experience a disadvantage stemming from their lower average socioeconomic origins). Specifically race-based intergene-rational factors still may affect the levels of black-white inequality in the next generation, but they probably operate through the quality (as opposed to quantity) of schooling, and these effects are not well understood.
Comparative analyses of mobility that go beyond the mobility table approach described above are complicated by substantive differences in institutional structures across nations and differences in the measurement of key variables. Nonetheless, progress is being made. In perhaps the most notable application of the original Blau and Duncan model to comparative analysis, Treiman and Yip used the ratio of the net effect of education on occupational attainment to the direct effect of the father’s socioeconomic background (this might be conceptualized as a ratio of ”achievement” to ”ascription”) to compare the process of attainment in different countries. This ratio varies from a relatively high level in industrialized societies (particularly in Scandinavia) to a low level in less industrialized societies (with India having the lowest value in their study). Scholars also are focusing on comparative studies of topics such as the transition from school to work, job mobility, earnings mobility, sex segregation, and family dynamics. Many of these studies are being carried out with newly available panel data on demographic and socioeconomic outcomes that are being collected in many industrialized societies. These new sources of data are complex, and it will take several years before a broad-based comparative literature that uses them becomes available. The direction and pace of research, however, are encouraging. Social mobility is likely to retain its vitality as well as its centrality in sociology for the foreseeable future.
