In the last years several studies about the relationship between buildings' energy performance and residential property prices in Europe have been published. Nevertheless, there is a lack of studies regarding the Italian real estate market, probably due to the difficulties to access official data on property characteristics and transactions prices.Following the recast of European Commission's Energy Performance of Building Directive (2010/31/EU), since 2012, in Italy, is mandatory to provide information about energy performance on the real estate advertisements for the sale of a property, including energy label and CO2 emissions. Buildings are rated on a scale A to G, with A being the highest rating.The purpose of this research is to evaluate the impact of buildings' energy performance, as expressed by Energy Performance Certificates (EPCs), on the prices of residential properties in the city of Padua, representative of medium-size cities in Northern Italy.The study is based on a hedonic price model which aims to estimate the impact on house prices due to the improvement of energy performance. The study adopts a semi-logarithmic regression model. The analysis is performed on a dataset of nearly a thousand dwellings, listed on websites for property sales.The model considers as dependent variable the offer prices of dwellings, and includes indipendent variables that represent attributes such as location, typology, dimension, mainteinance condition and other features, as well as the energy label.The dwellings are all located in the city of Padua, with the exclusion of some areas of the city centre, in which the location variable could prevail on the others independent variables. The data collection period lasted from July to September 2013.Empirical findings show a positive and statistically significant relationship between buildings with better EPCs and house prices. Furthermore the relationship is stronger from class G to C, than from class C to A. The price premium for a higher energy rating is tested for robustness and consistency. Nevertheless, the study reveals a negative gap between the net present value of potential future savings and the estimated price premium.