Purpose – The purpose of this paper is to investigate the validity of the cost of carry model by examining the time series properties of the deviation between future and spot prices in the European Union Emissions Trading Scheme (EU-ETS) over the time period 2005-2012. The paper utilizes a non-linear mean reverting adjustment mechanism, and discovers that although deviations of future from spot prices can exhibit a region of non-stationary behaviour, overall they are stationary indicating market efficiency in the trading of carbon permits. Design/methodology/approach – The methodology involves non-linear mean reverting unit root tests. Findings – The findings provide insights into the functioning of the EU-ETS market. They suggest that it is informationally efficient and does not permit arbitrage between spots and futures. Originality/value – The authors are the first study to examine efficiency in the EU-ETS by investigating the validity of the cost of carry model. The authors are also the only study to look at efficiency in both Phase I and Phase II of the scheme.