January 28, 2004
Study Dissects SRI Performance by Different Social and Environmental Factors
by William Baue
A Pictet study correlates financial performance with specific aspects of socially responsible
investment, such as corporate governance and stakeholder relations.
What is in a name? When it comes to the title of a quantitative study on socially responsible
investment (SRI) addressing a worldwide audience, lots. Case in point: Decomposing SRI
Performance by Christoph Butz, a sustainable investment specialist on the quantitative team of
Geneva-based private bank Pictet & Cie.
Ask your Microsoft Word thesaurus for synonyms of the term "decompose," and it comes up
with words that connote deterioration, such as "rot" or "decay," descriptors investors hardly want
associated with portfolio performance. However, the term also means "separate into constituent
parts," the study's intended connotation, as suggested by its subtitle, Extracting Value through
"This could finally explain why our study was sometimes heavily
misinterpreted by journalists who thought we wanted to shoot down SRI as a whole!" exclaimed Mr.
Butz, who is not a native English speaker and worked with a translator in writing this study.
In fact, the study assumes that SRI is not monolithic but rather multifaceted. It accordingly
breaks down SRI into environmental and social value chains, and further dissects each into numerous
subcategories according to policy and actual practice. The study analyzes the financial
performance of 288 listed European companies from January 1999 through July 2003 in each
"What it's trying to do--identify which social and environmental factors
have contributed positively or negatively to performance--is an important next step," said Lloyd
Kurtz. Mr. Kurtz is senior vice president and research analyst at San Francisco investment firm Harris Bretall Sullivan & Smith, and administers
the Moskowitz Prize for the
Social Investment Forum (SIF), an annual
award for the best quantitative studies on SRI. He also manages the SRIstudies.org website, an online annotated bibliography of
quantitative studies of SRI.
"We need to move from the generalities ('SRI Good' versus
'SRI Bad') to more specific questions, such as, 'what are the performance implications of specific
corporate governance policies?' Mr. Kurtz told SocialFunds.com.
In fact, the Pictet study
does examine the effect on financial performance of corporate governance, including such factors as
limitation of shareowner rights, remuneration, and fraudulent practices.
coefficient on the corporate governance factor is at odds with longer-term studies, Gompers being
the most important, that show positive returns to shareholder-friendly governance policies," stated
Mr. Kurtz, referring to a 2003 study authored by Paul Gompers and Joy
Ishii of Harvard and Andrew Metrick of Wharton.
"This finding was a surprise to us, too,"
said Mr. Butz. "However it cannot be denied that good compliance with formal corporate governance
criteria has penalized performance for the 288 companies over the last 4.5 years--examining on a
different sample for a different period, different results are indeed possible."
broadest factor analysis, the study finds that environmental factors slightly underperform the
benchmark (the MCSI Europe Index) while social factors more significantly outperform the benchmark.
"The conclusion that environmental policies don't have much impact is at odds with
Dowell, Hart, and Yeung's 2001 Moskowitz Prize winning study, which
demonstrated a statistically significant association," said Mr. Kurtz.
"Dowell, Hart, and
Yeung studied a sample of 89 companies in the extractive industries and headquartered in the US,"
said Mr. Butz. "For the broader market we studied, the results are, unfortunately, different."
The study looks even closer and analyzes specific factors within the broader social and
environmental categories, for example looking at relations with specific stakeholder groups.
"Breaking things down into stakeholder groups is conceptually elegant," said Mr. Kurtz.
The study assesses corporate performance taking into consideration companies' relationships
with each of five stakeholder groups. Isolating clients and the general public results in
benchmark outperformance, while isolating out employees resulted in neutral performance.
Unexpectedly, separating out shareholders and suppliers both resulted in benchmark
However, the study noted that isolating these elements, while useful,
also masks their necessary interrelationship. In other words, the study admitted that dissecting
specific SRI factors and correlating them to performance must also be counterbalanced by the
realization that these discrete elements interact dynamically in the real world and cannot be fully