Page 454 - Invited Paper Session (IPS) - Volume 1

IPS176 Thomas A. K.
under consideration. The signals regarding the change in momentum are
uniform in almost 80 per cent.

4. Discussion and Conclusion
In the paper, experimental price indices for German commercial real estate
are presented and compared. The virtue of the compilation here is that the
CPPIs, reflecting different definitions of commercial real estate, are coherent
in terms of source data for prices and weighting schemes. In particular, the
appraisals of transaction prices for office and retail structures as well as rental
housing in 127 towns and cities are taken from bulwiengesa and the weighting
schemes are derived from data on the floor spaces of these object types in
2011. With respect to the various definitions in use, the main conclusion is that
while the CPPIs partly exhibit trends of somewhat different strength, they are
very similar in terms of their general time series profiles referring, for instance,
to the location of turning points and signals with regard to the direction of
change or the change in momentum.
The experimental price indices help users analyse commercial property
markets. However, a number of weaknesses have to be mentioned. First, the
price data does not cover Germany as a whole but also 127 towns and cities.
The reduced regional coverage is perhaps of minor relevance for office and
retail because, in these market segments, transactions are concentrated on
towns and cities. For rental housing, it is definitely a shortcoming because
market activity is also present in rural areas and the trends in house and
apartment prices are shown to systematically alter between the more
populated and the rural areas of Germany. Second, with office, retail and rental
housing, the most important object types are covered. However, at least to
some part, the industrial and logistics building stock as well as hospitals and
other commercial buildings may also be marketable and, thus, to be included
in CPPI measurement. Third, the appraisals collected by bulwiengesa are a
second-best solution. Of course, actual transaction prices would be preferred.
7
However, disaggregate transaction-based price data is not available. Fourth,
prices are measured at annual periodicity. However, analysts desire price
indices at quarterly frequency.
With floor-space data, a stock-based weighting scheme is implemented.
With regard to definitions of commercial real estate which require distinguish
properties according to use and legal status of the owner, it is only stock
weights upon which price indices can be based. The 2011 census provides an
optimal source for the floor spaces of rental housing. As the census does not
cover non-residential real estate, there is scope for improvement in stock

7 The vdp CPPIs are compiled on the basis of granular transaction prices. The source data,
however, is exclusive to those who construct the price indices.
443 | I S I W S C 2 0 1 9