Some Major Findings Regarding Social Mobility in the U.S.
In the extant literature, the common if not perennial fear among students of American social mobility is that it is on the decline.26 For the most part, however, this conclusion has not been evident in the major modern studies of intergenerational occupational mobility for the U.S. Featherman and Hauser's (1978) meticulous study of the OCG-I and OCG-II data sets finds increased mobility and reduced ascription between 1962 and 1973 for a variety of models of various socioeconomic outcomes. Those authors state they have "detected two complementary trends: declining status ascription and increasing universalistic status allocation____If anything, the weight of the evidence of change suggests that the acquisition of schooling, jobs, and earnings has become less constrained by social background" (Featherman and Hauser 1978: 481).
Another important study (Hout 1988) investigates intergenerational occupational mobility from 1972 to 1985. This analysis combines features of association models as well as an earlier model of (Sobel, Hout, and Duncan 1985) that incorporates parameters for structural mobility. An additional innovation is that Hout's (1988) model also includes occupation-specific measures of status, autonomy, and training. In doing so, he allows for various sources of heterogeneity within broad occupational categories. Especially relevant is the role of education and training, and Hout finds that origin-destination association varies by educational level. In particular, the origin-destination association is nil among the college educated.27 Thus, as the proportion of college-educated workers increases, an increase in circulation mobility is evident. Although structural mobility decreased during this time period, circulation mobility increased. As Hout observes (1988: 1358), "the more college graduates in the work force, the weaker the association between origin status and destination status for the population as a whole. Overall mobility remains unchanged because a decline in structural mobility offsets the increased openness of the class structure.''
DiPrete and Grusky (1990), using data for approximately the same period as Hout (1988), reach similar conclusions, although their methodology differs. They find little evidence of an increase in the effect of ascription on occupational attainment, although
26 A decline in social mobility is also hypothesized by Herrnstein and Murray (1994), although in their view this decline is generated primarily as a consequence of the increasing stratification of cognitive skills.
27 As Hout (1988) notes, the finding that college education erases the origin-destination association may help to explain the paradox of higher-than-average attainment (i.e., beyond what would be predicted by social origins) of certain ethnic minorities in the U.S. such as Asian Americans and Eastern Europeans, as these groups have a higher than average proportion of college graduates.
some of their results suggest a slowdown in the growth of universalistic practices. Overall, however, the findings of DiPrete and Grusky seem generally consistent with those of Hout, showing considerable social fluidity during the latter 1970s and the 1980s despite some slowdown in structural mobility. DiPrete and Grusky (1990) argue that part of the continued universalism during this period is political, deriving from the enforcement of equal-opportunity employment legislation.
Hauser et al. (2000) provide results for occupational attainment during the early 1990s using the Wisconsin Longitudinal Survey. This thorough and informative test of a status attainment model that includes a variety of social background variables indicates the continuing importance of schooling as key determinant of occupational attainment, even after controlling for a mental ability test score. The direct effects of social background variables (net of schooling) are small or negligible, yielding no obvious evidence of a direct increase in the role of ascription in occupational attainment during the early 1990s.
Some Major Findings From Cross-National Studies of Social Mobility
A great deal of the research on social mobility has been cross-national. Hout (2003) succinctly summarizes much of our current knowledge derived from this research. Several important empirical generalizations emerge from his review. Of particular significance is what has come to be known as the Treiman constant: ''occupations are ranked in the same order in most nations and over time'' (Hout 2003: 2). In other words, in terms of average levels of prestige and socioeconomic rewards, occupations tend to be ordered similarly across most nations and across time within nations. This result was first reported and analyzed in Treiman (1977) and was subsequently extended by Ganzeboom and Treiman (1996). The Treiman constant is important because it implies that occupation is a major dimension of social inequality that may be directly compared across nations and across time periods for the purpose of studying patterns of social mobility. This finding thus underlies comparative studies of social mobility.
The latter topic is sufficiently complex to have generated a variety of debates and viewpoints over the years. However, a general consensus seems to be emerging recently regarding several basic conclusions. The first is that there is a great deal of variation in levels and patterns of structural mobility across nations and and across time periods (Erikson and Goldthorpe 1992). These variations in structural mobility (which are also sometimes referred to as variations in absolute mobility rates) reflect the generally historical and idiosyncratic features of the economic development and class structures of different nations. As stated by Goldthorpe (2000: 232) ''most mobility researchers came to accept the view, either implicitly if not explicitly, that variation in absolute rates cannot in fact usefully be regarded as systematic, and that explanations of such variation, whether over time or cross-nationally, will need to be provided far more in specific historical, than in general theoretical terms.''
A second major conclusion from this literature is that relative mobility rates— which may be construed as indicative of circulation mobility—show a great deal of similarity across industrial nations, particularly in regard to the fundamental pattern of association between origins and destinations. In other words, industrial nations do not differ much in terms of their levels of inherent social fluidity or degrees of openness.
Although differences do exist between countries, they are small compared to the cross-national similarities. In fact, "such variations in pattern as do occur are not of any major quantitative importance" (Goldthorpe 2000: 234). Hout (2003: 7-8) states that this conclusion generally applies both to "vertical" and "class" models of mobility.
Although somewhat less well conceptualized and studied, another basic conclusion from comparative studies is that education plays a critical role in the status attainment process in industrialized nations. As stated by Hout (2003: 9) ''education is the main factor in both upward mobility and the reproduction of status from generation to generation.'' For the U.S., this conclusion is evident in numerous studies such as Blau and Duncan (1967), Featherman and Hauser (1978), Jencks et al. (1979), and DiPrete and Grusky (1990). Studies of status attainment in other industrialized nations, however, also report similar findings (Hope 1985; Hout 1989; Ishida 1993; Treiman and Ganzeboom 1990). Because of the important role of education in determining occupational attainment, upward mobility is facilitated to the extent that persons from lower-status origins obtain high levels of education while status reproduction results when persons obtain levels of education that are consistent with their class origins (Bielby 1981: 6-10; Hout 2003: 9-12).
An additional important conclusion from comparative studies of social stratification is the persistence of class differentials in educational attainment in most industrialized nations. "Class differentials in educational attainment, considered net of all effects of expansion per se, have tended to display a high degree of stability: that is, while all children of all class backgrounds have alike participated in the process of expansion, the association between class origins and the relative chances of children staying on in education, taking more academic courses, or entering higher education has, in most societies, been rather little altered'' (Goldthorpe 2000: 182). The main source of this finding is Shavit and Blossfeld (1993), who investigate educational attainment in 13 countries including Czechoslovakia, Germany, Hungary, Israel, Italy, Japan, Netherlands, Poland, Sweden, Switzerland, Taiwan, United Kingdom, and the U.S. Previous studies have also reported similar findings for France (Garnier and Raffalovitch 1984) and the Philippines (Smith and Cheung 1986). The stability in class differentials in educational attainments seems somewhat surprising given the considerable cross-national variation in educational systems and the general increase in average levels of educational attainment across cohorts in all industrialized nations. However, this finding is consistent with (if not implied by) the previous two general conclusions about the cross-national similarity in relative mobility rates and the important role of education in status attainment in industrialized nations.28
Other major conclusions from comparative studies of social stratification pertain to gender. The first in this regard is that, when measured in terms of years of schooling completed, gender differentials in educational attainment have notably declined throughout industrialized nations since World War II (Shavit and Blossfeld 1993; Breen and Goldthorpe 1997). Indeed, in some nations, recent cohorts of women actually attend schooling for more years than do men. As mentioned by Hout (2003: 12), this trend may derive from ''rising returns to market work among women, the educational
28 The two countries where class differentials in educational attainment do appear to be significantly attenuated are the Netherlands and Sweden (De Graaf and Ganzeboom 1993; Jonsson 1993), and these two countries are also characterized by somewhat higher levels of circulation mobility (Ganzeboom and De Graaf 1984; Erikson and Goldthorpe 1987).
and occupational aspirations of post-feminist women, the family resources of smaller families, and the interests of privileged parents who have daughters but not sons.''
Another general result from comparative studies of social stratification is that "occupational distributions are gendered'' (Hout 2003: 3). In other words, throughout industrialized nations there are significant gender differentials in occupational attainment. In particular, women are typically much more likely to be employed in lower-level white-collar occupations. The most systematic cross-national study of this phenomenon is Charles and Grusky (1995). Although this study finds notable gender segregation in the occupational distributions of each of the countries considered, nations nonetheless differ significantly in terms of their specific patterns and in the changes of these patterns over time. Future research on this topic may need to investigate more thoroughly the role of gender differentials in educational type and specialty that continue to persist in many nations (despite declines in gender differentials in total years of schooing completed).
In sum, comparative studies have yielded an impressive body of knowledge about cross-national commonalities in the basic facts and processes involved in social stratification. In our review, we have been careful to limit these generalizations to those pertaining to industrialized nations because almost all of these studies are based exclusively on data from such countries.29 The extent to which these generalizations may also be applicable to developing nations thus remains a topic for future research. The main impediment to extending these studies to developing nations has been the availability of appropriate data.
Was this article helpful?