Does the UK have 7 social classes?
Last month there was a lot of chatter about an article entitled ‘A New Model of Social Class: Findings from the BBC’s Great British Class Survey Experiment’ that claimed to have identified seven distinct classes in the UK. Though not a sociologist, I thought I’d stick my oar into the debate by having a look at the paper.
The identification of seven classes was the result of a large survey conducted with the help of the BBC that gauged the levels of different kinds of ‘capital’ possessed by individuals, including not just assets and income but cultural capital (traditional and popular) and social capital. The authors then used a pretty swish statistical model to identify the most significant clusters of individuals possessing similar amounts of the various types of capital as one another. They thus identified their seven classes, which include an underclass and an elite as well as several distinct middling classes with various mixes of the different types of capital. This is an objective not subjective model of class, it doesn’t matter whether or not individuals identify as belonging to one class or another. The model receives a degree of validation by the fact that different occupations/professions predominate amongst their independently-defined classes (although ‘social capital’ is defined in terms of the occupations of people an individual knows, so the logic seems a bit circular (endogeneity?)).
Unlike the dominant approaches to class in sociology, the methodology is one of ‘bottom up’ formation of categories, deriving its classes from the empirical data, rather than defining classes analytically in a ‘top down’ fashion. That makes the paper pretty interesting. The paper isn’t atheoretical though, the methodology is based on the assumption that individuals can (imperfectly) convert different types of capital into one another. I’m not really sure, however, that ‘new cultural capital’ (knowledge of/participation in popular culture) is as important as assets and income for determining individual life-chances as the paper seems to assume. Once, yes, a classical education and knowledge of high culture might have enabled social and professional advancement. Sadly, I’m not sure that knowledge of 80s indiepop, The Wire or Marvel comics confers such an advantage (although fondness for BSG, GoT and amusing cat pictures does seem to help in the world of IR blogging). In addition, why isn’t human/educational capital included as a variable? It’s omission doesn’t seem to make sense as far as I can see.
The major issue with the article is that the class-categories it comes up with are an odd amalgam of socio-economic classes and generational cohorts. This is because the ‘classes’ identified by the survey are in part defined by their income and their assets. But for many individuals in professions and occupations with some kind of career ladder incomes start pretty low, but rise as individuals climb the ladder. In addition, in the usual life-course within industrialised societies, individuals can be expected to build up assets (such as property and pension pots) as their working life progresses. We should expect most young people to have relatively few assets and fairly similar (low) incomes. As individuals grow older their trajectories diverge, with gaps between the incomes of people in different professions/occupations widening. Individuals will then tend to build up asset portfolios at different rates. So in the schema offered by the authors individuals will fall into very different classes based on their age-group. This indeed does seem to have occurred. The ‘traditional working class’ for example is in part defined by its low incomes but reasonably high level of asset ownership (uh oh, something’s gone wrong if we are defining the w/c by asset-ownership), and the average age of its members is significantly higher than the average. ‘Emergent service workers’ comprise low-asset and low-income individuals with high ’emerging cultural capital’ , i.e. young people of all occupations who are just starting out on their careers and whose trajectories haven’t diverged yet.
The categories are muddled because the survey is a cross-section of the whole population. The usual methodology for research into class is to conduct a cross-sectional survey of an age-cohort and then follow-up the survey at regular intervals. This enables comparison of the trajectories or life-courses of individuals occupying different economic classes and/or status groups. So sociologists are normally more concerned with, for example, lifetime earnings rather than income at a fixed point in time. The methodology adopted by the authors instead compares individuals at very different points along the life-course. Now, maybe the article does tell us something about the broad class-status-generational groups that exist in contemporary Britain. Their survey also highlights the existence of a distinct ‘elite’ class (I might do another post on this) and the decomposition of the old working class that has occurred between generations. But I think most sociologists won’t want to collapse together different sociological categories like this and will stick to the well-established and well-validated ‘top-down’ schemas of scholars like Goldthorpe and Wright.