2 Methodology
Our data are drawn from a sample of 2530 single-family home owners, surveyed
in 2005 as part of the German Residential Energy Consumption Survey. Four
different retrofit measures (and their combinations) are surveyed: roof insula-
tion, facade insulation, windows replacement, and heating-equipment replace-
ment. These measures, along with the option not to undertake a retrofit, yield
a total of 16 different combinations from which the household chooses. In total,
64% of the households retrofitted their homes between 1995 and 2004.
While the decision concerning renovation is essentially driven by two deter-
minants, investment cost and the savings from reduced energy usage, the house-
hold’s choice is difficult to anticipate because of several uncertainties. First,
varying expectations of future energy prices will result in varying expectations
of the profitability of renovation options. Further, a household may face infor-
mation deficits as well as high costs of information acquisition about existing
retrofitting alternatives. Even when the alternatives are known, the calculation
of energy savings is likely to be beyond the capabilities of the layperson. Finally,
there might exist other hidden costs and benefits that determine the household’s
decision process. Examples of costs include the noise and dirt that accompany
some retrofit measures, while benefits may include higher social standing from
spill-over effects within a neighborhood (Ioannides 2002). As a consequence of
these considerations, there might exist preference heterogeneity concerning the
attributes of a retrofit, leading in turn to heteroegeneity with respect to the
household’s expected net benefits and hence WTP for energy-saving measures.
We accommodate such heterogeneity by employing econometric models that af-
ford broad coverage of the determinants - both observable and unobservable - of
the individual household’s utility from alternative retrofitting options.