The transition from fuel cars to electric cars is a large-scale process involving many interactions between consumers and other stakeholders over decades. To explore how policies may interact with consumer behavior over such a long time period, we developed an agent-based social simulation model. In this model, detailed data of 1795 respondents have been used to parameterise an agent architecture that addresses different consumer needs and decision strategies. Numerical experiments indicate that effective policy requires a long-lasting implementation of a combination of monetary, structural and informational measures. The strongest effect on emission reduction requires an exclusive support for full battery electric cars and no support for hybrid cars.