In energy efficiency networks, groups of firms exchange experiences on energy conservation in regular meetings over several years. The companies implement energy efficiency measures in order to reach commonly agreed energy savings and CO2 reduction goals. Energy efficiency networks exist in several countries, such as Germany, Sweden and China. Existing evaluations of such voluntary regional networks in Germany claim that participants improved energy efficiency at twice the speed of the industry average. Based on comprehensive data from the German manufacturing census, this paper examines whether participation in energy efficiency networks has a causal impact on energy conservation and CO2 emissions. I employ both a difference-in-differences estimator, using companies that joined energy efficiency networks at a later point in time as a control group, as well as a semiparametric matching estimator. I demonstrate that for the average participant there is no evidence of a statistically significant effect on energy productivity or CO2 emissions due to the network activities. However, there is some indication that exporters may have benefitted from the networks by reducing their CO2 emissions.