Which of the following statements is true? "The chances of social mobility in the UK match those in the United States and Japan." "British society is no more open now than it was at the time of the first world war." "Education plays a similar role in linking class origins to class destinations in both Germany and Britain."
The answer is that they are all true. These remarkable findings have emerged during the past 20 years from an intensive programme of research organised by the International Sociological Association. The researchers, scattered across several continents, form an academic network based on a shared interest in class mobility within industrialised societies. Until recently their discussions featured a good deal of disagreement-mainly about methods and techniques. Now, however, there are signs of an emerging consensus about results that will overturn much of what we thought we knew about social class in the UK and elsewhere.
defining terms
Social mobility can be looked at in two ways. First, it is possible to think in terms of career mobility, viewing movement between social classes as a work-history-from, say, first job to present job. Second, someone's current class may be considered in relation to that of his or her parents.
How is class itself defined? The classical definitions based on property ownership are no longer very useful. But the same can be said of some of the contemporary status and occupation-based definitions. Most international mobility researchers have settled for the categories devised by John Goldthorpe in the 1970s, which are based on relative advantage in employment conditions. The various categories in this occupational schema may not always correspond with our common sense and cultural understandings of class, but they have been subjected to intensive scrutiny by researchers in several countries and found to be effective for cross-national comparisons. This alone has been an important achievement of the research programme.
The top class is the salariat; the three intermediate classes are small proprietors/self-employed, routine clerical, and farmers; and the three lower (or working) classes are skilled and unskilled manual workers and agricultural labourers.
People in salariat occupations exchange commitment to the job against a high level of trust on the part of their employers. Professionals and managers are required to exercise delegated authority or specialised knowledge on behalf of their organisation, in return for which they enjoy relatively high incomes, career advancement chances, and a good deal of autonomy at work.
The intermediate classes stand between the poles of the service relationship and the pure labour contract. Organisations are often ambivalent about whether the routine clerical class are "staff" or "labour." The groups at the bottom of the scale, skilled manual and unskilled manual wage-workers, supply discrete amounts of labour in a relatively short-term and specific exchange of money for effort. Such working-class jobs are also subject to more intensive supervision from above.
HOW MUCH MOBILITY?
The substantive argument has focused on an astonishingly simple proposition about the chances of mobility, first advanced by David Featherman, Frank Lancaster Jones and Robert Hauser in the mid-1970s. They argued that the overall pattern of mobility chances-or what has since become known as the degree of social fluidity-"is basically the same" across industrial societies-at least those with a market economy and a nuclear family system.
This proposition has been supported by the most recent research. European-mainly British, Swedish and German-researchers involved in the Casmin (Comparative Analysis of Social Mobility in Industrial Nations) project during the 1980s conducted an exhaustive study of social mobility patterns in 15 industrial societies. They concluded that "the total amount of the association between class origins and destinations that is cross-nationally variable is... only very small relative to the amount that is cross-nationally common."
This is not to say that there are no variations whatever in social fluidity across societies (see Tables 1 and 2). But according to the Casmin data, the degree of "common social fluidity" is such that something like 95 per cent of all of the association between class of origin and class destinations was found to be identical across the nations investigated. Even those who were initially sceptical-notably some prominent Dutch researchers-now accept that there is a basic similarity between nations in the patterns of social mobility.
Researchers have also demonstrated an impressive stability in the pattern of mobility chances over time. The Casmin team examined class mobility among different birth-cohorts within countries, but were unable to find any tendency towards greater openness in class structures over the past 75 years. This pattern is also widely reported in other studies. Perhaps surprisingly, not only has there been no significant reduction in the class inequalities of mobility chances in the UK, but the same is also true of the apparently more open societies of the US, Australia and Japan. In much of the industrial world, the unequal pattern of social mobility chances has remained basically the same throughout most of this century-despite welfare states, free education, and redistributive tax regimes.
more mobile, no more open
This finding may seem implausible to those of us who grew up in the affluent 1950s, swinging 1960s, or even the entrepreneurial 1980s-but it has now been reproduced so often, across so many countries and data-sets, that it can no longer be in dispute.
It seems that we may have mistaken changes in the shape of the class structure for changes in social fluidity or the degree of openness. In most advanced societies, the postwar years saw rapid economic expansion, and a transformation in the occupational division of labour. With the shift from manufacturing to services, the proportion of the workforce involved in manual labouring declined, and there was a growth of professional, managerial and administrative positions. But more "room at the top" has not been accompanied by greater equality in the opportunities to get there. More salariat jobs are now done by the children of non-salariat parents than ever before-but this is a function of the expansion of salariat jobs. Adjusting for the overall increase in salariat jobs the relative job chances of a salariat child and a non-salariat child are virtually unchanged. In short, the growth of skilled white-collar work has increased opportunities for mobility generally, but the distribution of those opportunities to rise into the salariat has stayed the same.
These results carry profound implications for social policy. Despite half a century of social reform aimed-at least in part-to reduce the influence of class on processes of social selection, the degree of openness or inequality of opportunity in advanced societies has remained unchanged. Only Sweden, it seems, has moved towards diminishing inequality in class mobility chances-and then only marginally-perhaps because social policy in that country (including the tax and welfare system) has actually succeeded in reducing the inequalities of conditions between classes. (Recent changes in Sweden show that even in the model social democratic state there may be no majority for reducing inequality of conditions.)
What this programme of research seems therefore to have discovered is the "bedrock" of class inequalities in industrial societies. The Swedish exception suggests that governments which pursue equality of opportunity as an objective of social policy will find this bedrock difficult to penetrate unless they also take sustained action to reduce inequalities of class position, because an important aspect of class advantage is precisely the ability it gives people to confer further advantage on their children, either directly via the transmission of wealth, or indirectly through the medium of enhanced opportunities for educational achievement.
NATIONAL VARIATIONS
Despite the similarities in mobility between industrial countries, some differences can be detected. One way of measuring these differences is shown in Table 1. These results have been calculated from data gathered by the 1991 International Social Justice Project. For each country studied we report salariat versus working class "odds ratios" by gender. Odds ratios are a simple means of measuring comparative mobility chances. Here, they show the chances of someone from a salariat family finding himself or herself in such a position rather than in the working class, as compared to the corresponding chances of someone with a working-class background.
These data suggest that, in the UK for example, boys and girls from salariat homes are five or six times more likely to achieve salariat employment (and to avoid working-class destinations) than are their peers from working-class backgrounds. But the same is broadly true for a wide variety of countries, both democratic capitalist and former state socialist. The US fares rather badly when compared with some former communist countries. Mobility between the salariat and working class in the UK-supposedly class-ridden and backward-looking-turns out to be more or less the same as in progressive west Germany.
Similarly, using some of the now standard modelling techniques developed by mobility researchers, we can order these nations in terms of overall fluidity, as in Table 2. We are measuring minor differences when set against the background of a "basic similarity" in fluidity, but they illustrate that it is by no means the case that the most advanced capitalist societies are uniformly more open than their post-communist neighbours.
EFFECT OF EDUCATION
What are the implications of all this for social justice? It is entirely possible that unequal and unchanging class mobility rates simply reflect the fact that talent is unequally distributed, with the most gifted people tending to rise to the highest social positions and passing on their (perhaps genetic) endowments to their children. To decide whether these industrial societies are indeed just or unjust, analyses must include data not only on people's origins and destinations, but also on their "merits." There is no other way to tell whether the indisputably differing chances of class mobility reflect differing actual chances or differences in people's ability to take advantage of what are essentially equal opportunities.
Our own research into this issue in the UK suggests that children from salariat backgrounds are more likely to arrive at salariat destinations than are children from the working class-even when those children have attained the same level of education. At low levels of attainment (up to CSE grade 5), 23 per cent of those from salariat origins arrive at salariat destinations, as compared with only 7 per cent of men and women from working-class backgrounds. Similarly, among people holding medium-level qualifications (such as O-levels and A-levels), 39 per cent of those from salariat backgrounds themselves achieve salariat destinations, with only 20 per cent downwardly mobile into the working class. The corresponding figures among individuals with working-class origins holding the same educational qualifications are 17 and 36 per cent respectively. Among those with degrees or equivalent professional qualifications, 86 per cent from salariat backgrounds arrived at salariat destinations, as compared to 76 per cent of working-class children.
Conclusion
Sociological research into social mobility over the past quarter of a century has transformed our understanding of class processes and debunked a series of powerful political myths. The reality seems to be that, although all industrialised societies do not have identical patterns of social mobility, the distribution of mobility chances within such societies is not only "basically similar" but also "basically unchanging."
Social mobility - plus ça change...
November 20, 1995
