This paper discusses the proposal by the Garnaut Climate Change Review to auction (rather than grandfather in whole or in part) permits for an emissions trading scheme. While economic theory would suggest that how permits are allocated will have no effect on market outcomes in a frictionless world, the existence of real-world imperfections suggest that there is a risk that the auctioning of permits will adversely affect structure and conduct in the affected industries, particularly in the electricity sector. While these risks might be offset, to some extent, by the scope to use revenues from an auction to reduce distorting taxes, it is not clear from the Review’s documents to date that any auction revenues raised would indeed be put to effective use. Finally, the auction process itself, intended to be inclusive of a broad range of industries, will almost certainly be costly for all parties involved, and exposes participants to material financial risks if emissions quantities and therefore permit prices cannot be forecast with any degree of certainty. Those risks are a real cost to the economy, as are the resources participants would have to devote to seeking to manage them.