In this paper, we review the German practice of imputing the costs of owner-occupied housing by increasing the relative weight of actual rents in the CPI. As the structure of owner-occupied housing differs substantially from that of rental housing, this variant of the imputation method may cause a bias in the German CPI. For assessing the appropriateness of the German imputation method, we estimate alternative rental equivalent indices based on the GSOEP. We find some evidence of an understatement of the "true" rate of price increase, which is, however, not directly related to the imputation method.