Table 4: U-values and efficiency factors
Home Constructed Between | |||||
< 1975 |
1975 -1990 |
1991 -2001 |
2002 -2005 |
Required | |
U(Roof) |
1.5 |
0.5 |
0.4 |
0.3 |
0.3 |
U(Facade) |
1.5 |
1 |
0.5 |
0.35 |
0.35 |
U(Window) |
3.5 |
3.5 |
2 |
1.7 |
1.3 |
Efficiency Factor |
<1987 |
1987 -2001 |
2002 -2005 |
Required Standard | |
ep (Non-Electric) |
1.19 |
1.11 |
1.05 |
1.05 | |
ep (Electricity) |
1.05 |
1.05 |
1.05 |
1.05 |
Note: U-values are measured in W∕(m2*K). Source: Ecofys (2004), IWU (1997).
principle information sources. The first of these is the BKI, or Construction-Cost
Information Center of German Architects, which publishes unit-cost figures for
various types of retrofit measures based on samples of retrofitted buildings (BKI
2006). Because these figures are national averages that aggregate material and
labor costs, we supplement this information with regional wage data for various
classes of craftsman obtained from a labor-survey conducted by the FDZ (2006).12
We normalize both average-unit cost and wage data so that they are measured
in prices of the year 2000. The final step in calculating investment cost involves
constructing the ratio of local wages to the national average, which serves as a
regional weighting scheme to be multiplied by the average construction cost from
the BKI. This figure is in turn multiplied by an additional weight capturing the
12This survey contains average wages for various classes of craftsman, and, as with the house-
hold data, is measured at the scale of a municipality, of which there are approximately 13,490
in Germany. For a given craftsman class, there is an average of 200 municipalities from across
Germany for which wage data is available. To ensure overlapping coverage with the house-
hold data, we use GIS to spatially interpolate wages between the centroids of the represented
municipality using an inverse-distance weighted algorithm (Childs 2004). In this way, location-
specific wage information from the different craftsman classes can be assigned to each household
location in the dataset.
28