Household demand for electricity and natural gas

As an integral part of the project Søren Leth-Petersen has prepared a Ph.D. dissertation. The dissertation contains five chapters. (1) Demand for Space Heating in Apartment Blocks: Measuring Effects of Policy Measures Aiming at Reducing Energy Consumption; (2) Micro Econometric Modelling of Household Energy Use: Testing for Dependence between Demand for Electricity and Natural Gas; (3) Imputing Consumption from Income Tax Registers; (4) Residential Demand for Heating in Liberalised Electricity Markets. Evidence from Denmark and Norway; (5) Habit Formation in the Consumption of Energy by Households. (1) contains an analysis of consumption of energy in apartment blocks. The paper quantifies the effect of a number of policy initiatives aiming at reducing consumption of energy. Results among other things indicate that green taxes have had a limited short-run effect. Moreover, a consultancy scheme introduced to inform about energy savings potentials is found to have a limited effect. The second paper contains an analysis of the data set constructed from administrative register data. The analysis is limited to households living in single family owner-occupied houses, and the analysis focuses on observations from one year, 1996. The analysis shows the fairly surprising result that demand for heating is unaffected by the presence of children. On the other hand, evidence also points towards children impacting the consumption level of electricity. The analysis also shows that consumption of both gas and electricity is increasing with income. The model has also been estimated on the Danish Expenditure Survey, cf. (4), which is a random sample drawn from the Danish population, and the results from (2) are confirmed. The results from these two studies are of great importance among other things for policy makers evaluating the distributional impacts of green taxes on electricity and gas. The analysis presented in (2) also shows that households do not substitute between gas and electricity in the short run, i.e. for given technology. This is in contrast to Norway where households often are equipped with heating equipment that uses several different energy camers, cf. (4), enabling them to substitute between the use of different energy carriers in the short run. This implies that Norwegian households are more able to exploit short-run price variations in the electricity price that emerge following the liberalisation of the electricity market. One of the important problems of using administrative register data for consumption analysis is that no information on total expenditure (and hence savings for a given level of income) is available. In (2) this is solved by imputing total expenditure from the information available in the income tax registers. (3) shows that this methods is indeed valid for the major part of the data. This is likely to be of importance also to future research on consumption behaviour. The final paper (5) investigates if demand for energy is habit forening. Habits are often referred to by policy makers when initiating informational campaigns aiming at reducing consumption of energy. Habit formation implies among other things that consumption is increasing over the lifecycle of a household. The analysis shows that the traditional economic definition of habit formation does not give a sufficient description of the data. The evidence suggests that households use rule-of-thumb behaviour when making decisions about consumption of energy