We estimate direct and indirect energy rebound effects for a wide variety of goods and services in Germany. To this end, we employ a linearized approximation of the popular Almost Ideal Demand System (AIDS) approach suggested by Deaton and Muellbauer (1980). Excluding measures of energy efficiency when estimating rebound can lead to biased results. We compensate for this shortcoming of previous research by explicitly accounting for energy efficiency in our estimations. Using data for Germany from 1970 to 2017, we find moderate direct and significant indirect rebound effects for different energy carriers across four model specifications. The size of the rebound effect proves to be sensitive to the expenditure shares used for the calculations, which in some cases leads to negative overall rebound estimates.