This paper analyzes the choice between taxes and cap and trade systems (also referred to here as a permit system or a quantity restriction) as methods of controlling greenhouse gas emissions. It argues that in the domestic context, with proper design, the two instruments are essentially the same. Commonly discussed differences in the two instruments are due to unjustified assumptions about design. In the climate change context and within a single country there is sufficient design flexibility that these differences can be substantially eliminated. To the extent that there are remaining differences, there should be a modest preference for taxes, but the benefits of taxes are swamped by the benefits of good design; even though the very best tax might be better than the very best quantity restriction, the first order of business is getting the design right. In the international context, however, taxes dominate more strongly. The design flexibility available within a single country is reduced in the international context because of the problems of coordinating systems across countries and minimizing holdouts. Moreover, the incentives to cheat and the effects of cheating are not equivalent for the two instruments in the international setting. Because climate change will require a global system for emissions, these considerations mean we should favor taxes for controlling greenhouse gas emissions.