The national insurance system of contributory work-related benefits was designed for a ‘core’ workforce in continuous full-time employment. It provides incomplete coverage for the rising numbers of people in the peripheral workforce: part-time jobs, temporary work and self-employment. National estimates are presented for workers whose earnings are below the national insurance threshold, and who are thus excluded from social security benefits such as unemployment, sickness and retirement benefits—roughly 2 million in the period 1985–87. It is estimated that another million workers at least are within the NI net, but outside the income tax net. In addition, there are up to 2 million non-working people who have trivial earnings that leave them outside the NI and tax nets. Homeworkers are found in both these groups. The paper discusses the implications of these findings for claims about the size of the black economy and the characteristics of people participating in the black economy, and for the future development of the social insurance system.

Between 1955–7 welfare expenditure in Britain came under serious attack. The main protagonist was the Treasury and its chosen implement a five-year review of the social services, to be presided over by a ministerial Social Services Committee. The attack rebounded, for the Committee provided the opportunity for the consolidation of the defence of welfare expenditure and for a frontal attack on Treasury assumptions. This neglected episode in Conservative government social policy places in historical context the early defeat of monetarism (with Thorneycroft's resignation in 1958) and provides the background to the establishment of the Plowden Committee and of the Public Expenditure Survey Committee. It also raises questions about the degree of post-war consensus and the failure to make the constructive development of the welfare state an objective of ‘conviction’ politics.

The abolition in 1982 of the Earnings-Related Supplement (ERS) to unemployment benefit, which had been introduced in 1966, left the UK with no element of income support for the unemployed linked to previous earnings. The ERS scheme represents an important case study of economic and social policy but it has been little researched hitherto. The paper examines the history of ERS, showing how the original legislation and subsequent development produced a benefit that bore little relation to schemes in other countries. Unpublished administrative data on the receipt of ERS are used to help document the scheme's failings, and the paper closes by placing ERS within the context of the development of unemployment insurance in Britain.

This paper examines the historical context of the policy decision of the (then) DHSS in July 1974 to establish Regional Secure Units with an initial provision for 1,000 places. A brief examination of the history of the detention of the criminally insane and the setting up of the county asylums is followed by an examination of the various problems faced by the authorities concerned with the care of the criminally insane and the mentally ill in general in the 1960s. The paper examines the different streams of influence and power that converged upon this solution: government, special hospitals, public inquiries, unlocking of hospital wards, criminal law, DHSS and the Home Office, judges, voluntary bodies, prisons, psychiatrists and the official government reports known as the Glancy and the Butler Reports. The paper seeks to explain the policy decision to build regional secure units as a dynamic outcome arising from the confluence of opportunities, participants and solutions: a policy formation model put forward by March and Olsen (1976).

Legislation in 1985 and 1986 has attempted to give new impetus to the growth of occupational and personal pension schemes. This article demonstrates that, because of the interaction between occupational pensions and means-tested social security provision, many of today's pensioners receive little or no financial benefit from their occupational pension. The evidence presented is consistent with the thesis that the pensioners who are worst affected by the ‘pension trap’ include those who, as workers, were low-paid and experienced interrupted employment. As a consequence they reached retirement with small state and occupational pensions, limited savings and no house which they owned. Many are women. The penetration of occupational and personal pensions is currently lowest among the same groups of workers and it could be that the Government's policy to extend the coverage of private pensions will result in an increase in the severity of the pension trap. Some measures for reducing the impact of the pension trap are discussed.