Data-driven support and risk modeling for a successful heat transition in the building sector

Growing international interest in climate change and the ambitious climate goals of the Paris Climate Agreement requires policy decisions and actions to curb the adverse effects of human-made climate change. The buildings sector accounts for more than one-third of global greenhouse gas emissions and energy consumption, with space heating and water heating accounting for most of these, and offers great potential for progress toward climate goal achievement. Moreover, most of today's existing buildings were built before introducing more strict building codes than today and are therefore not sufficiently energy efficient. Due to the low number of new buildings compared to the existing building stock, extensive retrofitting is necessary, as the current building stock will continue to account for the largest share of energy consumption in buildings in the future. However, retrofits of these buildings are sparse, and the retrofit rate - the percentage of buildings that undergo retrofits in a year - is too low to meet climate goals. Therefore, this cumulative doctoral thesis examines two aspects for a successful heat transition in the building sector. The first aspect deals with the identification of general factors influencing energy efficiency and retrofitting practices on a regional level. It is not yet fully understood which local differences exist in building performance, energy efficiency, and retrofitting practices and how socio-economic factors influence these. Thus, this doctoral thesis follows the call to use the opportunities of advancing digitalization and data availability to examine this aspect. The findings indicate strong evidence for regional differences in building energy efficiency, confirm existing qualitative and small-scale studies regarding the influence of socio-economic factors and classify retrofitting-related CO2 taxes as reasonable and easy to implement. The second aspect shifts the focus from a regional level to individual retrofit decisions. It examines risk in general and inaccurate predictions of building energy performance in particular as barriers to individual retrofit decisions. The results show that promoting energy efficiency reduces the variance – and thus the risk - of future energy bills and opens up opportunities for more sustainable investment behavior. In addition, policy instruments such as energy efficiency insurance are more effective and cost-efficient than subsidies in mitigating the risk of environmentally friendlier investments. Regarding building energy performance prediction, data-driven approaches exceed the currently prescribed engineering method (in Germany) by almost 50% in prediction accuracy and provide insights into influencing factors. In summary, this doctoral thesis provides insights using data-driven and risk-modeling approaches for a better understanding of factors influencing energy efficiency and retrofitting on a regional level and risk in retrofit decisions and contributes managerial and policy implications that support a successful heat transition in the building sector.