State of the art

‘Socio-economic classification’ (SEC) is a generic, non-theoretical term which may be given to any measure designed to identify the position of individuals within the social structure, however that may be conceived and measured (whether categorially or continuously). As the term suggests, SECs play a key role in the study of the intersection of the social and economic spheres. Individuals in similar socio-economic situations occupy as a consequence common positions in the social structure. They possess similar resources as a result of their socio-economic positions and thus similar possibilities and constraints in terms of opportunities, behaviour and ‘life chances’ – for example in relation to educational attainment, health, material rewards and social mobility. These are all factors vital to our understanding of the development and consequences of a knowledge based society.

It is not surprising, therefore, that official socio-economic classifications (SECs) of national populations are widespread and have a long history. Currently nine of the EU member states have their own SECs. However, the lack of valid and reliable SECs for cross-national research has been a continuing obstacle to comparative European policy and scientific analysis. This was recognised by Eurostat as an issue to be addressed in the harmonisation of European statistics (Østby et al 2000 and Everaers 1998). To this end it has funded two projects: the first examined the extent, nature and use of SECs in member states (Grais 1999) and the second, by way of an Expert Group (Rose et al 2001), made proposals for an ESeC that could be operationalized using commonly collected data. This proposal is designed to build on these projects.

In terms of state of the art, Grais notes two contrasting approaches to the derivation of SECs in EU member states, the ‘theoretical’ and the ‘intuitive/empirical’. The latter are more common. Only the UK (for the new National Statistics Socio-economic Classification – see Rose and Pevalin 2003 and 2002), the Netherlands and Sweden (and also Norway) have adopted a ‘theoretical’ approach (all of which are based on or closely related to the Erikson-Goldthorpe-Portocarero (EGP) schema – see Erikson and Goldthorpe 1992). We may also note that the French Catégories Socioprofessionnelles (CSP) is also ‘theoretical’, although of a more inductive type. We agree with Brauns (1999), however, when she argues that the basic principles of CSP and EGP are relatively similar.

Our proposed theoretical approach to the development of a European socio-economic classification will also follow that of the EGP schema. This sees individuals as occupying a limited number of common positions in the social structure in terms of social power. As we have noted, individuals possess certain resources and consequently face a range of possibilities and constraints in terms of their behaviour. Those who share similar resources, and thus similar structural positions, will share similar possibilities and constraints in terms of ‘life-chances’. Therefore, they may also be hypothesised to act in similar ways.

In this approach, the structural base of social power provides ‘a link between the organisation of society and the position and behaviour of individuals’ (Breen & Rottman 1995: 455 emphasis added). While there are many bases of social power – age, race, gender, social status – it is generally agreed among sociologists that the most important in modern market-economy societies is that of social class, i.e social power based on market or economic power. These class positions exist independently of the individuals who occupy these positions. They are ‘empty places’ (Sørensen 1991). Empirical research then addresses the issue of how structural position, as objectively defined in this manner, affects life-chances.

The EGP approach to class analysis (see Erikson and Goldthorpe, 1992; and see also Goldthorpe 2000) first distinguishes three basic social positions, those of employers, the self-employed and employees. Employers are then further distinguished in terms of the number of employees in their establishments (i.e ‘large’ and ‘small’ employers). Similarly, employees are differentiated in terms of the nature of their employment relations as determined by their employment contracts. Three types of contract are identified, the service relationship, the labour contract and mixes of these two. The service relationship is typical of managerial and professional employment. It involves not only rewards for work done but also prospective elements, e.g favourable pension arrangements and similar perquisites and greater career opportunities and job security. The labour contract is a more specific exchange of money for effort and is typically found in manual (and increasingly in very routine non-manual in sales and services) work. Intermediate employees (e.g clerical workers) typically have a combination of elements of both basic contractual forms.

Empirical research has demonstrated the validity and utility of this model of employment regulation (Goldthorpe 2000: Ch.10; Rose and Pevalin 2003: Parts II and III). Despite the claims of writers such as Beck (2000) and other globalisation theorists to the effect that class inequalities have diminished, and by implication that the analytic value of class models such as ours has been undermined, the empirical evidence against this view remains clear. The validity and utility of conceptualising class in terms of employment relations continues to be affirmed (see, for example, Goldthorpe and McKnight 2002).

The means by which this theoretical model may be operationalized to create the ESeC was discussed earlier. Importantly, from the viewpoint of creating ESeC (and this is made obvious from the appendices of the Grais report) are the implied similarities of the various national socio-economic classifications in terms of categories and meanings. Similar variables enter into national SECs as those employed in the EGP schema – occupation, activity status, status in employment, enterprise size, agricultural and non-agricultural sectors – and the categories of the theoretically based SECs are generally interpreted to have meaning as ‘social’ units. Nearly all are ‘nested’ classifications; i.e they have more and less aggregated versions, thus providing flexibility for analysts. Some also have ‘reduced’ forms that offer a means for dealing with the varying quality of data available to analysts (see, for example, Rose and Pevalin, 2003: 18-19). Thus, there are many common basic features to national SECs that may be exploited in the derivation of an ESeC. Details differ on issues such as size cut-off for enterprises, definition of household reference persons and the precise treatment of the economically inactive. However, the principles of national SECs, whether implicit or explicit, appear to be similar.

Finally, it should be noted that academics have produced comparative social class measures for use in their own cross-national research. Most importantly in this context is the development of the EGP schema for the Comparative Analysis of Social Mobility in Industrial Societies (CASMIN) project (see Erikson and Goldthorpe 1992). Two of the partners in this proposal, MZES and SOFI, were involved in developing this schema. As we have indicated, the same theoretical model will be used for ESeC, but it will be operationalized using EU harmonised variables rather than national ones.