The overriding aim of this thesis is to establish how integrated input-output data
frameworks in monetary, physical and time units can contribute to a be~ter understanding
of the environmental pressures generated by a given final demand including the
underlying economic, social and demographic driving forces. The thesis mainly focuses
on environmental' input-output analysis and related methods and evaluates the·
opportunities provided by recent data developments at the Federal Statistical Office. In
particular, physical input-output tables and social accounting extensions published as part
. of the 'socio-economic reporting system' are used for improving the specification and
conceptualisation of production technology and lifestyles.
The first part of the thesis contributes to the recent discussion on monetary and .
physical input-output analysis. In particular, it looks at how· the representation of
production technology can be improved through the availability of information from
physical input~output tables (PlOT) to allow for a more robust allocation of
environmental pressures to final consumption/demand. The conceptual discussion
,- highlights a whole range of misperceptions in the debate associated with the' construction
of the German PlOT and highlights the shared conceptual basis between monetary inputoutput
tables (MIOT) and PlOTs to the extent to which product flows are concerned.
However, a detailed empirical comparison of production structures iIi monetary
and physical units using the graph theoretical toolkit provided by qualitative input-output
. analysis ~ighlights fundamental differences in their representation of technologies due to
the particular scope of monetary and physical measurement: 45% of all intermediate
product flows in MIOT and PlOT are fundamentally different in that they have a positive
record in one table and a zero record in the other.
As expected, most of these are 'weightless' immaterial service flows. However,
the thesis highlights that such fundamental differences in the production structures
associated with intermediate service flows are not only relevant in tertiary sectors, but are
prominent throughout the economy: in fact, for some manufacturing sectors of capital
goods with a high service component immaterial service flows can make up to 90% of all
intermediate' outputs, highlighting the importance of an endogenisation of capital
investment for an adequate attribution of environmental pressures to final demands.
Remaining differences are explained by unpriced, material flows in environmental service sectors (recycling, waste treatment), where PlOTs provide a more comprehensive
coverage. The first part of this thesis concludes by highlighting that production
technology in environmental input-output models will usmilly be most appropriately
specified in hybrid units. An outline of some of the main avenues for future research is
provided.
The second part of the thesis uses detailed SAM-type extensions to better
understand the environmental pressures associated with lifestyles in their socio-.
demographic context. Initially, an expenditure based lifestyle definition is deployed. to.
analyse the social and demographic driving forces behind changes in GHG emissions
associated with consumption patterns of 45 lifestyle groups in Germany between 1990
and 2002. A structural decomposition analysis confirms previous studies in that most
technologically induced reductions in GHG emissions have been 'eaten-up' by additional
emissiotls from growth in final consumption. However, results highlight that important·
demographic trends are at work at the same time counteracting GHG emission savings.
These pressures need to be considered in climate change policy formation, if climate
change targets are to be delivered,
Results from the environmental input-output model are further analysed using a
panel regression approach in order to highlight the influence of individual social,
economic and demographic detetminants of GHG emissions. The time-specific effects
capture the slowing progress in GHG emission reductions after the re-unification in
Germany. Group specific effects highlight the dominance of household size and the
be10ngingness to a particular social group for differences in GHG emissions from
consUmption patterns of different lifestyle groups.
The analysis is concluded by highlighting the importance of adding social and
. demographic information into standard environmental input-output frameworks to better
understand global environmen~al pressures generated by the consumption patterns of
different lifestyle groups. However, the top-down classification of lifestyles as commonly
applied in national accounting based on only a few socio-demographic descriptors such as
income, occupancy and household size is seen to limit the analysis. Of at least equal
importance with people's social and demographic characteristics are the local conditions
within which they are acting: general neighbourhood characteristics, the accessibility of
private and public services and building properties (size, type, age, insulation etc.). Geodemographic
lifestyle classifications, as commonly applied by marketing practitioners, are proposed as a spatially-specific alternative raising hopes to overcome the 'one size
fits all'-type policy recommendations which are commonly derived from environmental
input-output models.
Finally, the commonly applied expenditure based lifestyle definition is
fundamentally challenged. It is argued that a lifestyle definition should be based on what'
people do rather than on what they spend. Following the economic household production
function literature, this activity focus in the empirical description of lifestyles can be
achieved through the introduction of time-use ~ata. The usefulness of the approach is.
'demonstrated in an empirical example using data from the input-output tables in time
units provided by the Federal Statistical Office of Gennany.
In the Appendix of this thesis, an initial analysis of the social' and economic
determinants of CO2 emissions based on geo-demographic lifestyle data is provided.
Furthennore, different ways of dealing with environmental pressures from imported
products based on single region and multi-regional input-output models are discussed and
a methodology for estimating Ecological Footprints based on input-output analysis is
proposed.