One goal of the Swedish energy policy is to reduce the amount of electricity used for heating in the building sector. This means to reduce the primary energy used for heating which in this paper is analyzed in the context of various heating technologies and CO2 emissions. The analysis is applied to a region in Sweden (southern Sweden) for which detailed information on the energy infrastructure (the capital stock of the buildings and heating systems together with geographical variations in heat intensity) is available from a previous work [Johansson, P., Nylander, A., Johnsson, F., 2005. Electricity dependency and CO2 emissions from heating in the Swedish building sector-current trends in conflict with governmental policy? Energy policy] and which is large enough to be assumed representative for Sweden as a whole. The detailed mapping of the energy infrastructure allows a good estimate on the rate at which the energy system can be expected to be replaced with respect to economical lifetime of the capital stock (the year 2025 in this case). Two scenarios are investigated; a: target scenario for which energy savings are employed (e.g. improving climate shell in buildings) and oil and most of the electricity used for heating purposes are phased out and a second for which the current trend in the heating market continues. In the target scenario it is shown that although only applying commercially competitive heating technologies, it is possible to achieve a 47% reduction in primary energy use for heating with a 34% decrease in heat demand together with significant reduction in CO2 emissions. However, the scenario which continues the current trends on the heating market instead yields an increase (of about 10%) in primary energy use (reduction in conversion efficiency) of the heating system of the region over the period studied, in spite of a slight decrease in heat demand (9%, mainly due to energy efficiency measures) as well as in CO2 emissions. In light of the recently proposed introduction of energy performance certification of buildings in Sweden (and in EU), it can be concluded that such a certification system must take into account the primary energy use and local conditions. A politically acceptable way to do so would be to propose municipalities to develop "best strategies" on development of the heating market for various house types and districts within the municipality which can then be included as recommendations in the energy performance certification of the buildings. (c) 2006 Elsevier Ltd. All rights reserved.