Abstract
Scotland and Wales went to the polls on the same day in 2016, the former in a post- independence referendum environment and the latter shortly after receiving further devolved powers. Despite this horizontal simultaneity, the results in these elections displayed considerable diversity. Sub-state elections are often classified as second-order and as such are characterized by lower turnouts, anti-government voting and small party support. However regional elections in the United Kingdom sit uncomfortably in this definition. This article focuses on the experience of Scotland and Wales, exploring regional and national factors that help to explain regional electoral outcomes. It examines valence models, multi-level blame attribution, and relative national and regional party leader support. In a second stage, a multinomial logistic regression is undertaken, finding that the relative importance of regional and national factors differs between the two regions.