Social policy and taxation have commonly been regarded in Britain as quite separate aspects of government policy. Changes in taxation appear to be decided largely independently of the aims of social policy, and reforms of the social security system are often proposed with no regard to their fiscal consequences. Despite the fact that Chancellors of the Exchequer have recently arrogated the right to announce increases in National Insurance benefits, there is little evidence of co-ordination between the Treasury and the Department of Health and Social Security over income maintenance. Yet there is clearly a close relationship between these two arms of government policy, and it is important that any proposal for reform should consider taxation and social policy in conjunction.

This paper, based on an exploratory study of households in an English south coast town, argues that main income source is a significant factor to be taken into account when distinguishing between households. The research was triggered by information drawn from the 1981 Census showing that less than half the population of Great Britain experienced a direct relationship between current paid employment and income. It suggests that generalisations based on an assumption that household income drawn mainly from paid employment is the norm, marginalise more than half the population. The study draws attention to some of the issues highlighted when the focus of research is on sources of income. These are presented as supplementing, rather than negating, views from other perspectives. This exploratory study highlights additional insights having implications for social policy, while acknowledging that each merits more detailed attention.

First, we employ a methodology that decomposes the change in poverty into changes due to the means and variances of market and transfer incomes and to other factors which affect the shape of the income distribution. We then briefly describe the Reagan economic programme. Finally, we estimate a regression model that reflects changes in these moments of the income distribution to changes in the unemployment rates and to the Reagan programme. The regression estimates, together with the poverty rate decomposition, allow us to show that the administration's cuts in income-transfer programmes account for about 40 per cent of the recent rise in poverty and that increased unemployment accounts for about a third of the increase in poverty.