Since the election of the Thatcher Government in 1979, public expenditure on the welfare state has undergone a series of changes. The nature of change has, however, often been obscured by the use of inadequate and misleading statistical data. This paper presents a systematic examination of public expenditure on education, health care, housing and social security over the period 1979/80 to 1984/85. It shows that, contrary to much political rhetoric, the major changes in the welfare state have not always arisen from reductions in programme expenditures, but from changes in the composition of expenditure. It is the latter that has often increased economic inequality and can be legitimately referred to as ‘restructuring’.

‘Tax expenditures’ are public revenue losses which result from special allowances and reliefs given to various categories of taxpayer for reasons of economic and social policy. In 1983/4 tax expenditures in the personal income tax system cost nearly £11 billion which was equal to 35 per cent of revenue from personal income tax or 9 per cent of total public expenditure. This paper assesses their significance in the context of public expenditure and tax policy. It identifies those allowances and reliefs in the personal income tax system which may be regarded as tax expenditures, evaluates them and compares their cost with direct expenditures in similar areas. Many tax expenditures are inequitable and inefficient; and they are difficult for governments to control. If they were reduced some public expenditures could be protected from cuts, or the general burden of income tax could be reduced.

Social segregation in public housing has usually been explained in institutional terms by reference to the policies and practices of the local authority. This paper draws on data from a household survey of new council tenants in Glasgow to demonstrate that the outcome of the rationing process cannot be attributed to factors originating at the level of the institution alone. The research examines stages of the allocation process, in particular the behaviour of applicants in expressing preferences and their propensity to reject offers, as well as institutional factors. It is suggested that household choice behaviour, which is related to income, is important in determining social patterns in council housing.

Rowntree is widely regarded as having originated in 1901 the ‘scientific’ definition of poverty as the minimum income level required for physical subsistence. He is quoted as defining ‘secondary poverty’ above this income level as mismanagement. Critics of this approach confuse Rowntree's use of concepts with his discussion of causes, and they overlook Rowntree's own explanation that his concept of poverty was relativistic life-style, and that his distinction between primary and secondary poverty was a heuristic device to convince individualists that the life-style of the poor was at least in part caused by low income and not by improvidence. Townsend's major life work in defining and measuring poverty as relative deprivation is usually presented as overthrowing Rowntree's paradigm. The paper shows that Rowntree's early views and methods have been widely misunderstood by later authors, and it argues that the evidence necessitates a reconsideration of Rowntree's position, which would show Townsend's achievement as a paradigmatic shift not from absolutist to relativistic models of poverty but from relativistic models based on standards prescribed by expert observers to relativistic models based on standards derived from the whole population by social surveys.