Improved methods to evaluate realised energy savings

This thesis regards the calculation of realised energy savings at national and sectoral level, and the policy contribution to total savings. It is observed that the results of monitoring and evaluation studies on realised energy savings are hardly applied in energy saving policy. Causes are the lack of common formats and definitions, the use of inappropriate energy quantities, a too high aggregation level in the analysis, different methods as to the calculation of total versus policy savings, lack of comparability with other countries and with saving targets from scenario studies and finally too few focus on the needs of policy makers. In this thesis four new evaluation methods are presented that regard total and/or policy savings. The MONIT evaluation system is based on energy balances with energy carriers and sectors. By constructing a set of energy balances, using driving factors and starting from the base year balance, observed changes in energy consumption are split into 14 contributions. These comprise various saving effects, structure-effects due to shifts in socio-economic activities, substitution between fuels and effects of energy import or -export. MONIT also supplies CO2-emissions and -reductions coupled to energy consumption and savings. The second method regards a so-called top-down decomposition of changes in total energy use according to the Protocol Monitoring Energy savings (PME). Here energy savings are calculated separately for end-use, cogeneration and energy supply, as to meet policy demands of Dutch government. A special feature is the provision of uncertainty margins for the savings figures obtained. These prove to be quite high, e.g. 30% for savings at the national level. The third method regards a qualitative analysis of possible interaction between effects of various policy measures, using an interaction-matrix. Each matrix-cell contains information on the type and strength of interaction between two policy measures. For overlapping combinations the combined energy savings effect is smaller than the sum of both effects apart; for reinforcing combinations it’s the other way around. Results for Dutch households show that only 9% of all possible combinations show a strong interaction effect. The last method regards a simulation of historic energy developments for households in the Netherlands. A bottom-up model, applied for scenario studies, has been adapted to simulate observed trends in the past. The simulation results show the contribution of different policy measures and the effect of autonomous drivers, e.g. energy prices. The results also show that the combined effect of energy taxes, investment subsidies and regulation of gas use for space heating is 13-30% lower than the sum of the three measure effects apart. Another finding is that policy measures increasingly mitigate the effect of higher energy prices. It is shown that a combination of these methods can solve almost all evaluation problems. The results acquired are used to develop a new evaluation system for the Netherlands that is suited as well to meet the increasing international evaluation demands