An integrated mathematical model constituting of interlinked submodels on technology costs, progress and market penetration has been developed. The model was applied to a few new energy technologies to investigate the economic boundary conditions for a full market breakthrough and corresponding market impact on a 50 years time scale. The model shows that public subsidies amounting to slightly over 220 billion C in total worldwide would be necessary over the next 30-40 years to bring wind and photovoltaics to a cost breakthrough in the market and to reach a 20 and 5% share of all electricity at t = 50 years, respectively. These up-front learning investments would be partly amortized toward the end of the interval as the new technologies become cost competitive but could be fully paid off earlier if CO2 emission trading schemes emerge even with modest CO2 price levels. The findings are sensitive to changes in the parameter assumptions used. For example, a 2% uncertainty in the main parameters of the model could lead to a spread of tens of per cents in the future energy impact and subsidy needs, or when related to the above subsidy estimate, 155-325 billion C. This underlines the overall uncertainty in predicting future impacts and resource needs for new energy technologies. Copyright (c) 2006 John Wiley & Sons, Ltd.