This thesis has two aims. The first aim is to set out an argument for social insurance in the form of compulsory income insurance in the event of sickness or unemployment, and to explore two lines of arguments for social insurance policies that are commonly associated with an active welfare state that seeks to prevent or reduce reliance on social insurance. The second aim is to outline and defend an account of legitimacy that takes moral autonomy seriously by making legitimacy (...) partly dependent on our entrenched values and preferences. The first aim is relevant for articles I-VI. In article I it is argued that the extent to which behavioural responses to social insurance is seen as ethically problematic, it is primarily a problem that concerns the institution rather than the morality of the individual whose behaviour is influenced by social insurance. Thus, insofar as behavioural responses to social insurance are an ethical problem, it is a problem for political philosophy rather than individual ethics. In article II an argument for social insurance in the form of compulsory income insurance in the event of sickness or unemployment is presented, viz. the argument from autonomy. It is based on a concern for the protection of our identity according to what is called a “thick” conception of the person, which holds that our identities as separate persons are constituted by our central aims and commitments. It is also argued that contrary to what has been claimed by its opponents; social insurance needs not lead to the bad risks exploiting the good risks, or be head-on in conflict with individual freedom. Article III identifies normative issues that deserve attention in relation to in relation to a general introduction of prevention policies in social insurance and market insurance. It is argued that the importance of these issues suggests that arguments and distinctions drawn from moral and political philosophy should play a more prominent role both in the debate on the shift towards an active welfare state and the use of prevention policies in market insurance. Article IV is a response to comments from Professor David Buchanan initiated by article III. Article V explores what is called the argument from autonomy for reduced compensation rates in social insurance or making compensation from such insurance conditional on different kinds of requirements such as participation in rehabilitation or vocational training. It is argued that such policies are justified if they tend to ensure an adequate level of autonomy, where autonomy is understood in the sense of a “thick” conception of personal autonomy based on Norman Daniel’s extension of the principle of fair equality of opportunity. Article VI discusses the objection that arguments pertaining to the principle of fairness often are irrelevant since the principle of fairness is based on the acceptance of the relevant benefits. It is argued that this objection from non-acceptance fails because we can – and do – accept the benefits form such institutions on a practical level and this is enough to ground obligations pertaining fairness. The implications of this argument for policies associated with the active welfare state are explored, taking a reform of the Swedish sickness insurance as an example. The second aim is relevant for article VII. In article VII it is argued that an account of legitimacy should satisfy three conditions. The justification thesis and the legitimacy thesis are presented as accounts of justification and legitimacy respectively. It is argued that the proposed accounts satisfy these conditions. An account of political obligations is also given. (shrink)