The possibility of using public sector housing to disperse black people across our cities in the interest of racial integration is again attracting attention. Based on an analysis of housing department records, this article examines the effects on Birmingham's black settlers of the operation of one particular policy of racial dispersal. Having described the events leading to the introduction and ultimately the termination of the policy, the article proceeds to investigate the locations in which Asian, West Indian and native white tenants were housed, their area preferences, and the housing categories from which their allocations were made, as well as their ‘points’ levels on allocation. The article concludes by arguing that, on these dimensions, the consequences of the policy for black people were almost entirely negative.

The recent debate about the establishment of a ‘British Brookings’ involved a number of fundamental issues which were not brought out. In fact the idea that the British policy-making process should be made more ‘rational’ through the development of what are sometimes called policy studies is not new. It has roots in the Heyworth Report on social studies, which recommended greater use of social research in policy-making, and in the Fulton Report on the civil service, which argued for more policy-planning. These two approaches may now be seen as basically the same, and the problem as one of changing the relationship between social science and (social) policy. However, past analyses of this relationship attribute difficulties to quite different causes and hence yield a variety of prescriptions for reform. It is argued here that the policy studies which are needed must avoid the disciplinary fragmentation of the social sciences as well as that of the current administrative structure, that they must encompass research both for policy and on policy, and that they must seek their own conceptual structure, and in addition that certain organizational requirements follow from this.

The BMA (British Medical Association)–TUC (Trades Union Congress) Joint Committee on Medical Questions, established in 1936, was an experiment in consensus. It was the first attempt to cross the barriers in medical politics. In the 1930s the strategy of ‘administration rather than advocacy’ prompted the co-operation of the TUC with the BMA on medical questions of mutual interest. The operation of the partnership is explored here in the formulation of a scheme for a national maternity service acceptable to both parties. However, the reality of co-operation at the level of specific proposals was an insistence by the BMA on its own point of view. The TUC had to decide whether a joint scheme was worth the price. Although a scheme was eventually produced, as an experiment in consensus the committee had failed. To some extent the war obscured this failure, but the consensus itself was wrecked by the intransigence of the BMA and the hostility of the Labour movement, which was angry at the extent of the concessions which the TUC had made for a national maternity scheme.

This article suggests a Gateway Effects theory of discrimination in employment which attempts to link discrimination at the entry port into a job with its effects on later employment experiences and behaviour. A longitudinal sample of black and white 1971 school-leavers from Bradford and Sheffield was used to propose this theory. An attempt is made to generate a theory about discrimination which explains the position of black British school-leavers and which also incorporates the concerns of existing alternative American-based labour market theories about discrimination. Several implications for policies about discrimination are deduced.