Speaker: Rob Johns (Essex University)

Abstract:
It has long been recognised that the first-order/second-order distinction is inadequate for regional elections. It it also now recognised that, at least in those regions with more formal powers and more distinct identities and party systems, regional elections can reasonably be seen as first order. What has not yet been tested is whether, in such cases, the order of regional and statewide elections is reversed. For those voters with a stronger sense of regional identity and a greater interest in regional politics, statewide elections might become second-order contests at which voting is driven by regional considerations. In this paper, I use the Scottish Election Study 2017 to test for ‘reverse order’ in that Westminster election. The results of the test are promising. First, it is possible to identify voters whose primary focus is on Scottish politics, even at a Westminster election. Second, in 2017, their party choice was better predicted by evaluations of the Scottish government and Scottish leaders than by the corresponding evaluations of at the UK level. In other words, the 2017 election had a distinct second-order flavour for those voters. The results argue for a further refinement of our understanding of multi-level elections.