The issue ownership theory is mainly developed in the US. It holds that parties, in the eyes of the voters, are considered as best-placed to deal with a certain policy problem or issue. Issue-ownership has important electoral consequences; when voters think about the issue when they vote, chances increase they will vote for the party that owns the issue. We argue that, for the issue ownership approach to become useful in a multi-party context, it needs to be further developed theoretically. Our major claim is that issue ownership consists of two aspect that are associated but analytically and empirically different dimensions: a competence dimension and an associative dimension. The competence dimension refers to the classic idea that some parties are considered ‘better’ to solve a problem. The associative dimension, in contrast, refers to the spontaneous association voters make between an issue and a party. These are two different things and both matter foir the vote. We test these ideas on data from Belgium, a country with a fragmented party system. We both draw on a cross-sectional sample of voters as on an online experimental study manipulating Belgian voters’ competence and associative issue ownership.
