For more than 50 years the United States and Canada maintained illegal-drug policies that followed the same course: a long period of punitive prohibition followed by moderation and an emphasis on drug abuse as a public health problem. Then in the 1980s, the USA reverted to a punitive model while Canada increased its commitment to a health-based approach. Why this divergence after following the same path for so long? In this paper I argue that one factor was Canada's adoption of national health insurance, which guaranteed universal access to health care, including addiction treatment. As the country's most popular policy it was protected against budget cuts during a period of welfare-state retrenchment in the 1980s. In the USA, on the other hand, public health insurance was limited to the elderly and the poor, and addiction treatment services were isolated and stigmatized. Thus the public health side of drug policy was poorly positioned to resist welfare cutbacks and ascendant criminal-justice interests. The experiences of the USA and Canada have implications for policy reformers and for the study of how institutional interests cross policy domains.

The dimension of time has been relatively neglected in studies of means-tested benefits. This article explores some of the issues which arise when time is considered explicitly. It is based on housing department records for 205 rent allowance claimants in Bristol which were monitored over a twenty-seven- to thirty-two-month period between November 1974 and June 1977. All the claimants had initially applied for rent allowances in response to a one-off promotion undertaken by the local authority and so they do not constitute a representative sample. Nevertheless mathematical controls indicate that the findings are fairly robust. Only a third of the claimants studied were continuously in receipt of rent allowances during the monitoring period and most claimants who dropped out of the scheme did so as a consequence of changes in their circumstances rather than because they failed to reapply. For most claimants, therefore, the rent allowance scheme provided temporary assistance rather than continuing financial support. The implications of the marked fluctuations in the financial circumstances of the observed claimants are discussed in relation to the administration of means-tested schemes.

This article sets out to explore the ways in which social policy is catered for, at university level, in various European countries. It starts with a review of the ‘Sigriswil (1972 and 1973) Colloquia’ papers on ‘The Study of Social Policy: Teaching and Research’. Having weighed up the evidence – or lack of evidence – emanating from these get-togethers, the article then goes on to explore university social policy teaching – and the extent to which this may or may not constitute a coherent specialist programme – in selected West German, French, Swedish and Swiss (Geneva) university arrangements. Concluding comments on this evidence highlight several key aspects of debate so far as present and probable trends in social policy and administration education at university level in Britain are concerned.

Within this Book Review the publisher and price of the book were incorrect. The publisher is The Netherlands Institute for Social Research and the price is $84.90. The publisher apologies for any inconvenience caused.