Since the signing of the Kyoto Protocol in December 1997, several studies have estimated the costs of reducing greenhouse gas emissions by the amount specified in the Protocol, as well as the price of the permits. A number of authors have recently shown that, following the US withdrawal and the Bonn and Marrakesh agreements, these abatement costs will be very low and the price of the permits could reach zero. However, these analyses usually take only the first commitment period (2008–2012) into account and do not explicitly consider the possibility of banking permits from one commitment period to the other (Art. 3.13 of the Protocol). The simple dynamic model that we develop here introduces this possibility. It allows one to analyze the impact of alternative future commitments for the US and the non-Annex B countries on world emissions, abatement costs, and the permit price. We find that, provided ambituous post-Kyoto commitments are negotiated: (1) in 2008–2012, the number of banked permits will largely exceed the amount of hot air and permit prices will be much higher than predicted by most other studies, (2) the banking provision significantly reduces world total costs but increases total costs for all permit-importing Annex B countries (i.e., all Annex B countries except those of eastern Europe) via a rise in the permit price in 2008–2017, and (3) the issue of market power on hot air is not likely to be relevant.