Democratic decentralisation and gender sensitive policies have offered a massive opportunity for political parties in Africa to mobilise support. As for instance, in Malawi in 2000, as many as 860 new political offices were up for elections as a consequen ce of democratic decentralisation. Democratic decentralisation and policies aimed at including women in decision-making structures represented an expansion of the political parties. Yet, while some parties gain from liberalisation other parties meet sever al challenges by opening up the political space. The paradox is thus, that if one compares the situation in Malawi, South Africa and Uganda, it is in the less competitive systems that reforms, such as democratic decentralisation and gender quotas, are sus tained. The explanation of this is easier to understand when one know that the incumbent parties in the dominant party systems in Uganda and South Africa originated as freedom movements, when the incumbent party in the competitive party system in Malawi w as launched as a consequence of a personal conflict between the former and current president. The aim of this analysis is therefore to investigate which interests parties in Malawi, South Africa and Uganda represents by grounding our understanding of the notion of representation in comprehensive in-depth studies. Existing research argues that the representation have to be understood in relation to the strong position of the president, the prevailing clientelistic structures in the society and the illibera l character of many Africa countries. What is unique with this project is that it explores how local actors understand the notion of representation and what it is that motivates them to participate in party politics. The project thus combines an analysis of the motivation and strategies of national elites, with bottom up perspectives on the introduction of institutional reforms and the function of political parties.