Error-Correction as a Concept and as a Method: Time Series Analysis of Policy-Opinion Responsiveness

Abstract

Since first cutting my research teeth as a doctoral student on the subject of the politics of public celebrations (Jennings, 2004), I have been interested in the nature of the relationship between public opinion and the behaviour of elected government (see also Jennings, 2012). This chapter looks at the problem of interpreting dynamic data: Do elected politicians listen to the demands of the public in making and implementing their decisions? I am interested in questions not because I think that responsiveness to public opinion is intrinsically a good thing but because it seems to me to be at the core of the functioning of democratic systems, as well as being central to the electoral survival of government: where non-responsive or underperforming political parties and candidates tend to be punished at the ballot box by the public. In fact, the case that first stimulated my interest in such a question — the Millennium Dome — involved quite the reverse. In that instance I was struck by the puzzle of why elected officials who were preoccupied with re-election (the poll-obsessed Blair government, no less), and who were likely knowledgeable of the economic determinants of vote choices of the public, would risk political capital on an unpopular white elephant such as the Millennium Dome? There would surely have been no electoral punishment for opting for a far more modest celebration of the new millennium.