Page 25 - Materials Chemistry, Second Edition

6 LIFE CYCLE ASSESSMENT HANDBOOK

don't tell what the ultimate outcome of a shift to cycling in commuting will be.
The income not spent on cars will be spent on something else, anything. The
shift to cycling is also linked to a different spatial infrastructure, with different
retail systems, different housing requirements, etc. Though one may be con-
fident that this is all to the environmental good - there may be good reasons
to believe so - that assessment is not just based on LCA. The analysis of the
u
overall system effect can easily be set p in a way that cycling really is bad. If
the income not spent on cars is assumed to be spent to a substantial degree on
flight based holidays, the net environmental outcome of more cycling might
well be negative. When reckoning with such behavioral mechanisms, the
choice of mechanism will determine the outcome, quite haphazardly at the
moment. So the question is if a strategy for analysis can be set p to include
u
the most relevant mechanisms in an equitable way. The move towards conse-
quential LCA is a possible step, but not the only one.
A core question is if dynamic, non-linear mechanisms can be incorporated
in the comparative static or steady state framework of LCA, as consequential
LCA. Or, should the micro level LCA technology system better be placed in
a broader modeling system reckoning with income effects, dynamic market
mechanisms, structural effects and constraints, and what more might be rel-
evant? The modeling required definitely does not fit in the linear homogenous
system of LCA based on matrix inversion for easy solutions. It seems wise to
first investigate divergent cases with an open mind as to most relevant causali-
ties, and to look into options for structured modeling later. Then a choice for
micro-type consequential LCA might be substantiated, or not, or only for some
applications.

1.6 Macro Level LCA for Policy Support

The use of LCA in public policy has been coming up, with an LCA-type of
analysis being used. The domain of application of LCA has been that of spe-
cific product choices. However, the link to broader policy issues, never absent,
seems on the rise. Biofuel, see below, is a major example, with unresolved dis-
cussion in the EU. The general feature of policy applications is that they should
show how a change considered would work out, requiring an ex ante analy-
sis of consequences of policy options, or an ex post analysis showing how a
policy has worked out. In both cases we need to know 'how the world would
have been different/ The functional unit with an arbitrary volume then is to be
replaced by an analysis covering the total volumes. Policies tend to be set up
in order to reach specified goals, not marginal effects of an unknown volume.
Using traditional arbitrary-unit LCA for policy support assumes a correspon-
dence between micro level LCA outcomes and macro level consequences for
the choice at hand. This assumption should be substantiated. It also relates
to the average versus marginal discussion, with causalities most easily estab-
lished at a marginal level, but overall effects then requiring integration over
all marginal changes, as increments. For substantiating the consequences of