Most models of policy responsiveness expect parties in office to behave as rational goal-maximising actors who are likely to respond to electoral incentives. Although previous studies have identified circumstances under which elected executives might be less âpolicy responsiveâ in their actions, they provide little explanation as to why executives can be concurrently responsive and non-responsive on issues of importance to voters. This thesis reassesses the appropriateness of rational choice theories for explaining how elected executives are likely to respond to electoral incentives. It searches for evidence that elected executives in the UK rationally respond to electoral incentives by maximising not just their goal of office, but also their policy goals. The setting of the analysis in the UK allows the thesis to consider how governments respond to voters in a system where the strong fusion of powers means that the relationship between the government and voters is not complicated by the presence of multiple veto players. This thesis uses three different methods of analysis. Firstly, I examine the issue prioritisation of UK government agendas and find that Labour governments are more likely to prioritise those policy issues which are of greater relative importance to Labour Party supporters, but that there remain differences between the issue prioritisation of Labour and Conservative governments which are not explained by policy responsiveness nor by the policy goals of Labour and Conservative MPs. I then explore how the first Thatcher administration responded to the electoral incentives identified by private polling and find that they do appear to have rationally balanced policy and office goals, although the analysis cannot verify the extent to which private polling influenced the policy-making process. However, I then investigate the decision-making process in three cases where governing parties failed to respond to electoral incentives and uncover several patterns which challenge the ârational choiceâ expectations of much of the literature on policy responsiveness. I find that time pressure leads political actors to rely on heuristics and not on public opinion research when forming judgements about votersâ likely reactions, and that politiciansâ own policy goals significantly bias their judgements about votersâ likely reactions. This thesis therefore challenges the rational choice expectations of much of the literature on policy responsiveness by arguing that political decision-makers are unable to rationally anticipate the reactions of voters.