Quantitative analysis of “greener” innovation at national systems level is
not always possible due to the lack of reliable, comprehensive and adequate
data sets. In the Swedish case, there are promising data sets that can be
combined, even if one always can debate the adequacy of existing variable
definitions. Official data collected by the central statistics authority
include several interesting variables on all (!) private and public
organizations in Sweden and all (!) employees compiled into time series for
a number of years. These can be merged with other data sets on the
environmental goods and services sector and energy consumption data in
industry and therefore allow for a detailed “demographic” or “population
ecology” analysis of environmentally oriented innovation since at least 2003
in industry and the green product sector. In this paper, these databases are
described in some detail. Problems of definitions and measurement are
particularly discussed. Some initial descriptive statistics are presented.
Further, it is suggested to consider some models inspired by demography and
population ecology and also include multi-level (in this case national,
organizational and individual level) models. In particular it is suggested
that diffusion models could be applied, including models in which diffusion
processes at organizational level interact with individual level flows of
ideas through networks. It is suggested to apply such multi-level regression
analysis in order to estimate the power of both organizational and
individual level factors in the “greening” of the Swedish innovation system.