Do party manifestos matter to government policy? Does a genuine party mandate operate within the British political process? These questions are generally neglected in analyses of British politics, but they are crucial in assessing how far political parties transmit electoral preferences into government action. We try to answer them through a novel use of available data, using content analysis to code and classify policy emphases within the post-war election programmes of the Conservative, Labour and Liberal parties. Statistical analysis reveals that government party programmes are remarkably well reflected in post-election policy priorities, measured as percentages of central government spending in major policy areas. This gives strong support to traditional mandate theory within the context of the ‘Westminster model’ of party government. Anomalies, such as a strong relationship between Liberal emphases and expenditures in three key areas, and the more consistent relationship of expenditures with Conservative rather than Labour priorities, are also considered.
* State University of New York at Binghamton and University of Essex, respectively. Invaluable support and intellectual guidance has been provided by Hans-Dieter Klingemann and the Department for Research on Institutional and Social Change of the Science Centre, Berlin, which has also graciously hosted the authors during the time of analysis and the writing of this piece. This study is one of a series based on the work of the Manifesto Research Group of the European Consortium for Political Research, in which authors' names are alternated to indicate the joint nature of the research. Support for the comparative analysis of expenditures and election programmes has been given by the Nuffield Foundation (Soc 205/125), the Tercentenary Fund of the Bank of Sweden, the Stiftung Volkswagenwerk (11/38850) and the British Social and Economic Research Council (E/l00/23/0220/1) as well as by sabbatical leaves from the authors' home institutions. This support is gratefully acknowledged. We likewise appreciate Thomas Uthup's help in gathering the expenditure data. Considerable revision and improvement has been made as a result of helpful suggestions from anonymous readers of an earlier draft.