The use of well-being research in assessing and creating public policy is gaining popularity. The UK's Office of National Statistics has developed its National Well-being Index to do exactly that, and several other nations have followed suit. This sway in political will, however, can make the use of well-being research in public policy seem less controversial than it is. Much of the rhetoric around the move toward well-being measures in evaluating policy has centered either on the idea that well-being is what ultimately matters or that it is at least something we all care about. In this essay, I will argue that such claims are illegitimate from the perspective of political liberalism. The former kind of claim is illegitimate insofar as liberal societies should not base policy on comprehensive religious, moral, or philosophical doctrines that many reasonable citizens may not accept. The latter kind of claim is illegitimate insofar as some people care significantly more than others about well-being or the ingredients of well-being. Thus, the use of well-being research to evaluate public policy cannot be justified on either of these grounds. Does this mean that there are no justifiable grounds for well-being-based policy? Not necessarily. In this essay, I explore the possibility that the psychological aspects of well-being can be viewed as a That is, the preservation and promotion of well-being can be justified as instrumentally valuable for most people. Well-being is instrumentally valuable for most people regardless of their particular intrinsic values. In John Rawls's terms, well-being can be viewed as an all-purpose good that people are assumed to want whatever their plans. The reason for this is that the psychological aspects of well-being, the kinds of things subjective well-being research typically measures, tend to be cognitively and motivationally necessary for agency. Without being able to appreciate one's life emotionally and cognitively, one cannot sufficiently pursue one's own conception of the good. Thus, well-being tends to be necessary for leading a good life regardless of one's conception of in what a good life consists. Even if it is the case, however, that well-being can be viewed as a primary good in this way, it is not necessarily the case that the state should promote it as a matter of justice. The state should promote only primary goods that share certain important features. First, a given primary good must be distributable and objectively comparable if the state is to promote it in a just manner. In addition, the state must be in the best position to promote a given primary good. That is, the primary good must require institutional support: public policies related to its existence and continuation. Moreover, the primary good must be non-fungible. That is, it must not be commonly obtainable through substitutes--things other than the goods and services provided by the state. I will argue that the psychological aspects of well-being share each of these features. Thus, according to political liberalism, governments can legitimately provide the social goods necessary for the preservation and promotion of the psychological aspects of well-being. However, I end with a word of caution. The extent to which public policy should promote well-being is not obvious. It may be that only the basic psychological aspects of well-being tend to be necessary for agency. Indeed, it may be that only the absence of the psychological aspects of ill-being (such as chronic depression) is a primary good. This may make a difference to the kind of policies that governments should enact to preserve and promote well-being. Well-being and Political Liberalism Much of the talk about the use of well-being research in public policy has focused on either (a) the general validity or particular implications of well-being research, or (b) the value of well-being and related concepts (such as happiness, life satisfaction, quality of life, and so on). …