Explaining stability and change of territorial identities

A significant body of work examines the presence and strength of territorial political identities (either subnational, national or supranational). A common assumption of this literature is that the presence and strength of these political identities are invariant over time. Given the importance of political identity, it is surprising that this assumption has not been empirically tested. We address this omission by testing this assumption through considering the question of who is most likely to exhibit variation in the reporting of territorial identities and why. We posit that one source of instability in territorially based political identity is rooted in cognitive dissonance which emerges through the interaction of partisanship and electoral outcomes. We explore these questions using panel data from the British Election Study (1997–2001), the Canadian Election Study (2004–2008). Results reveal that the territorial identities of Labour and Liberal partisans, in Britain and Canada respectively, are compatible with expectations.