The Polarisation of Work and the Distribution of Income in Britain

Recent studies have highlighted a polarisation in the distribution of work during the 1980s, and some have conjectured that this polarisation has been a principal cause of the rise in income inequality. Using data from the British Household Panel Survey, we document the pattern of work polarisation during the first half of the 1990s and show that polarisation was not mainly responsible for income inequality growth over this period and argue that it was unlikely to have done so during most of the 1980s either. It played a supporting role rather than a leading one. JEL classification numbers: D31, J21. Running title: Work polarisation and the distribution of income. Acknowledgements This note derives from an Institute of Labour Research programme of research on “Labour Market Dynamics in a Changing Environment” funded by the Leverhulme Trust. We also acknowledge core funding support from the UK Economic and Social Research Council and the University of Essex. Thanks to Centre and Institute colleagues for comments and advice. Address for correspondence: Stephen P. Jenkins, ESRC Research Centre on Micro-Social Change, University of Essex, Wivenhoe Park, Colchester CO4 3SQ, United Kingdom. Email: [email protected]. The Polarisation of Work and the Distribution of Income in Britain by Marco G. Ercolani and Stephen P. Jenkins Non-technical summary Recent studies, notably by Gregg and Wadsworth (1995, 1996a, 1996b), have highlighted a polarisation in the distribution of work in Britain: a decline in numbers of ‘traditional’ singleearner households has coincided with a rise in both the fraction of ‘work rich’ households (in which all adults work) and the fraction of ‘work poor’ households (within which no adults work). It has been conjectured that this polarisation has been a principal cause of the rise in income inequality, and we subject this hypothesis to scrutiny. Using data from the British Household Panel Survey, we document the pattern of work polarisation during the first half of the 1990s and show, using shift-share counter-factual simulations, that polarisation was not mainly responsible for income inequality growth over this period, and argue that it was unlikely to have been so for most of the 1980s either. It has played a supporting role rather than a leading role. Our results should not be interpreted as saying that the changing distribution of work is irrelevant for understanding the changing distribution of income. In fact we tend to the opposite conclusion: it is precisely by comparing the experience of different groups within the population characterised in terms of attachment to work (rather than by age, say) that a coherent explanation of recent income inequality trends can be constructed. Our point is to reiterate that the focus should be on the changes within work status groups rather than the changes in subgroup sizes themselves.