Government Popularity and Attitude Towards the Government's Record Revisited

A recent paper in this Journal (John Hudson, ‘The Relationship Between Government Popularity and Approval for the Government's Record in the United Kingdom’, XV (1985), 165–86) examined the relationship between government popularity as indicated by the voting intentions series of the opinion polls and approval for the government's record. Among other things it demonstrated that government popularity increased in the quarter prior to an election and declined in the quarter thereafter and that in the two years following the election of a new government the polarization of political allegiances declined (low approval at this time yielded higher levels of government popularity than would otherwise be the case and high levels of approval yielded somewhat lower levels of government support than in other periods). No evidence was found for any other form of electoral cycle of the type suggested in previous studies. The conclusions were based on ordinary least squares estimates and the use of the Cochrane-Orcutt technique, employed because of potential problems with serial correlation in residuals. Whilst this attempt to examine the relationship between these two variables is to be welcomed, especially the theoretical developments hitherto frequently lacking in this area, there are a number of points of concern.