Benchmarking Sustainable Development: A Synthetic Meta-Index
Approach Laurens Cherchye (Catholic University of Leuven, Belgium)
Timo Kuosmanen (Wageningen University, The Netherlands) WIDER
International Conference on Inequality, Poverty and Human
Well-being, Helsinki, 30-31 May 2003 Slide 2 What is Sustainable
Development? The Brundlant Commission Report (1987) gives the
standard definition SD="Meeting the needs of the present without
compromising the ability of future generations to meet their own
needs Aspects of SD: Economic, Social/Political, Environmental
Slide 3 Environmental Sustainability Index 2002 An Initiative of
the Global Leaders for Tomorrow Environment Task Force, World
Economic Forum Slide 4 From WEF (2002): Strengths of ESI + Measures
Environmental Sustainability + Permits cross-country comparisons +
Method is transparent, reproducible + Enhances capacity to
benchmark performance, guide policy, deepen understanding
Weaknesses - Assumes particular set of weights (!) - Suffers from
gaps in available data - Lacks time series data which limits
ability to identify policy drivers Slide 5 Data Envelopment
Analysis (DEA) weighting Slide 6 Related studies Zaim, Fre,
Grosskopf (2001) Soc. Ind. Res. DEA based achievement and
improvement indexes Mahlberg & Obersteiner (2001), IIASA report
Re-measuring HDI by DEA Slide 7 Points of departure Broader scope
of SD that accounts for economic, social-political, and
environmental aspects Meta-level approach (index of indices)
Emphasis on developing a weighting mechanism which neither
specifies weights a priori nor allows for any weights that show the
country in positive light (weight-restricted DEA). Slide 8
Weight-restricted DEA Slide 9 DEA-based SD index: definition Slide
10 3 types of weight restrictions: Relative weight between 2 SD
indicators (h,i) for a given country j Relative weight between SD
categories (k,l) for a given country j Relative weight of the given
indicator i between 2 countries (j,k) Slide 11 Components of our
meta-index: Slide 12 Weights Slide 13 MISD rankings, High-income
countries (>$9266/cap.) Slide 14 MISD rankings,
Upper-middle-income countries Slide 15 MISD rankings,
Lower-middle-income countries Slide 16 MISD rankings, Low-income
countries Slide 17 MISD versus GDP/capita r = 0.315 Slide 18
Conclusions DEA appears a promising tool for weighting multiple
dimensions of SD to identify benchmarks. Normative judgement on
min/max bounds for weights seems less controversial than choice of
any specific set of weights. Weight flexibility can be restricted
across SD outputs, but also across output categories and countries.
Slide 19 Challenges for future research Constructing a superior SD
index directly from measures and indicators, rather than using
aggregated indices. Dynamic index: Measuring a rate of change in
the stock variables, instead of mixing up stocks and flows. A
Malmquist index approach. Slide 20 Full paper available Download in
pdf form from my homepage:
http://www.sls.wau.nl/enr/staff/kuosmanen/ Or send e-mail to:
[email protected]