March
13, 2009
Can SRI Ratings Predict Corporate
Behaviour?
by Ron Robins*
So far there are only a few studies in this area and
at best they point to weak linkages between an
individual company’s SRI (socially responsible
investing) ratings—as assigned by ratings agencies—and
its subsequent corporate behaviour.
The first study I will review is,
Imitate or Differentiate? Evaluating the validity of
corporate social responsibility ratings, by
Aaron K. Chatterji (Duke University), and David I.
Levine (University of California), published in February
2008. This study analyzed data from top SRI ratings
firms which included KLD Research & Analytics, Inc.,
Calvert Group, FTSE4Good, Dow Jones Sustainability
Indexes, and Innovest Strategic Advisors. The focus was
on the raters’ ability to predict which companies they
followed would later get involved in scandals.
Regrettably, the results found that raters could not
predict which companies would become embroiled in them.
Firms with both high and low ratings were equally likely
to be entangled in scandals, according to the
researchers. Furthermore, neither a “… narrow focus
on governance or a broad measure of social
responsibility (including charitable giving, environment
impact, product safety, etc.) seems to distinguish firms
that will have major scandals from those who will not.”
However, in another study we see some glimmers of hope
with SRI raters in predicting corporate behaviour
relating to the environment. The study is titled,
How Well do Social Ratings Actually Measure Corporate
Social Responsibility? Researchers again
included Aaron K. Chatterji and David I. Levine, but now
also included Michael W. Toffel (Harvard Business
School). Focusing on KLD’s SRI screening, they concluded
that, “… firms with more KLD ‘concerns’ have
slightly, but statistically significantly, more
pollution and regulatory compliance violations in later
years. [However], KLD environmental strengths [ratings],
in contrast, do not accurately predict pollution levels
or compliance violations.” The researchers say that
KLD might be able to significantly improve upon their
predictive ability in the environmental area with
adjustments to the way they gather and use data.
Of course, besides looking at corporate scandals and
environmental concerns, there are many other variables
that could be studied. Hopefully, in time such research
will be completed and provide us with a better
understanding and modelling methodology to improve the
predictive power of SRI corporate ratings by rating
organizations.
However, there is recent evidence that ratings can
encourage companies with poor ratings to perform better.
In the 2008 study,
How Firms Respond to Being Rated, again by Aaron
K. Chatterji and Michael W. Toffel, and covering 600
firms in the USA rated by a prominent social rating
agency, they say that, “While negative ratings may
‘shame’ firms that are performing poorly, the threat of
regulatory action and the presence of ‘low hanging
fruit’ are important drivers of how firms respond to
information-based incentives [like SRI ratings].”
Congratulations to Aaron Chatterji and his colleagues in
pioneering this important area of research. Most likely
many others will join them in such research and in time
create a set of variables that have greater predictive
power in determining corporate outcomes and activities.
My next editorial reviews the ability of SRI raters’
ability in foretelling an individual company’s stock
market performance. Similarly, the possibility of SRI
raters in predicting the outcomes of groups of stocks
and portfolios will be examined as well. And these
results are more promising.
*
Ron Robins, MBA, is founder, Investing for the
Soul (http://investingforthesoul.com/),
Huntsville, Ontario, Canada. He advocates, teaches, and writes
on the subject of ethical investing. To contact him,
e-mail
to Ron Robins or call 705-635-3034.
© Ron
Robins, 2009. Provided full credit, which
includes title, my name, and link to this post is given,
anyone may print or re-produce this article in part,
or in full, to any relevant web page. |
