Khan, the MSCI KLD dataset, also used

Khan, Serafeim, Yoon (2016) analyse and compare the impact of material
versus immaterial sustainability investments on future performance. This is a
new approach since the materiality classification data on sustainability issues
provided by the Sustainability Accounting Standards Board (SASB) was newly
available at the time of writing their paper. The paper therefore adds an
additional layer and quality of information to previous work on the
relationship of SRI and stock performance.

With regards to sustainability data, the authors use the MSCI KLD
dataset, also used in most other similar studies as it includes a large number
of U.S. companies over a long period of time. It covers the following issues:
(1) Community, (2) Corporate Governance, (3) Diversity, (4) Employee Relations,
(5) Product, (6) Environment, and (7) Human Rights. The dataset is designed as
a binary system of strengths and concerns, representing policies, procedures,
and outcomes that enable a firm to have a positive or negative impact on the
focal issue.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

To distinguish between material and immaterial sustainability issues,
the authors refer to SASB guidance. Issues are considered being material by
SASB if there is evidence of wide interest from a variety of user groups and
evidence of financial impact. As of February 2014, SASB had produced guidance
for the following six sectors (again comprising of a total of 45 industries):
healthcare, financials, technology and communications, nonrenewable resources,
transportation, and services.

In a next step, they classify each KLD data item as material or
immaterial following SASB guidance for each of the 45 industries in the sample
and construct on this basis a materiality (immateriality) score that measures
performance on material (immaterial) sustainability issues for each firm-year.

The reduced number of items identified to be material for the different sectors
is small compared to the total number of 124 KLD data items (range from 13 for
the healthcare sector to 32 for the services sector). This indicates that this
classification helps to reduce complexity of the analysis and to focus on the
relevant items only.

Methodology applied

In order to analyse the impact of sustainability investments on future
performance, Khan, Serafeim, Yoon (2016) then orthogonalize changes in this
materiality score with respect to changes in firm size, market-to-book ratio
(MTB), profitability (ROA), financial leverage, R&D and advertising
spendings, institutional ownership, and sector fixed effects. They then form
portfolios of firms in the top and bottom quintile of the unexplained portion
of the sustainability index change and estimate Fama and French (1993)
calendar-time regressions to test for one-year-ahead abnormal stock return
performance of the portfolio.

By adopting this approach, they aim to control for standard risk factors
and then test whether a portfolio long and short scoring high or low in the
focal characteristic, sustainability scores in this case, yields alpha. The
alpha indicates the future stock returns associated with the relevant firm
characteristic and unexplained by the firm’s exposure to conventional risk

According to Khan, Serafeim, Yoon (2016), this research design allows to
examine the correlation between changes in sustainability investments and
changes in stock prices and to capture unexpected performance that cannot be
attributed to the five systematic risk factors of the model. As a result, if
ESG data are informative about a firm’s future performance that is not
attributed to its correlation with market, size, value or growth
characteristics, momentum, and liquidity, then this leads to a significant
alpha. Using this approach, the authors aim to attribute the future performance
of a portfolio to material sustainability investments more confidently.


points, controversies

A general need
for further research on sustainability activities and their impact on
performance serves as starting
point for the analysis of Khan,
Serafeim, Yoon 2016. They aim to contribute with their paper to the
controversial discussion on the relation between sustainability or sustainability
ratings and firm performance, as previous studies didn’t result in a clear picture.

The main viewpoints
in this discussion are:

1) Sustainability investments
are efficient from the perspective of shareholders’, as they lead for example to
obtaining better resources, higher-quality employees or better marketing of
products and services. They may also mitigate the likelihood of negative
regulatory, legislative, or fiscal action while protecting and enhancing
corporate reputation.

2) Sustainability
investments are not efficient from the perspective of shareholders’, as they disproportionately
raise a firm’s costs and therefor result in a competitive disadvantage. Reasons
for making such inefficient investments could be managers capturing private
benefits or political beliefs.

Overall, Khan,
Serafeim, Yoon 2016 respond to this controversy by differentiating material
from immaterial investments based on the SASB classification previously not available
and analysing their impact on firm performance. They argue, that such classification is conducive to empirical testing.
First, as the research process of the SASB to distinguish material from
immaterial issues may be captured by special interests, for example by NGOs that
seek to steer the output in preferred directions. Second, because this
classification is new and needs to be validated for its use by future

A controversial issue
from a methodical point of view is that also alternative interpretations of
alphas than the one adopted by the authors exist in literature. The
interpretation of the materiality alpha used by Khan,
Serafeim, Yoon 2016 is however that
‘since materiality classifications were not previously available, investors
could not react to them as soon as ESG performance data became available (the
sustainability data did not distinguish between material and immaterial
investments). As such, the price change (or alpha) was realized over a longer
horizon as the materiality investments began to pay off through observable
metrics, such as higher accounting returns, or as investors better understood
the financial implications of sustainability investments through their own

Finally, a critical
point concerning the used materiality data is that SASB adopts an investor
viewpoint only. As a result, a topic might be classified as immaterial from an
investor standpoint although it could be important for other stakeholders. As sustainability
investments very often affect financial performance indirectly via investments primarily
targeting other areas, e.g. customer satisfaction, employee engagement, regulatory
risk etc., an overlap between materiality classifications for different
stakeholders can be expected.


Open questions and future research

Khan, Serafeim, Yoon 2016
suggest that ESG performance measures that take into account materiality
guidance like the one provided by SASB are more likely to clarify the relation
between sustainability investments and financial performance. One focal area of
future research should therefore be to further explore different materiality
classifications and how they affect future firm performance.

As they find a robust relation between
investments on material sustainability issues and future financial performance,
the authors also ask for further examination of the structural relations that
lead to this association. Some of the questions yet to be answered include the
influence of investments in material issues on specifc areas such as customer
loyalty and satisfaction, employee engagement, brand and reputation, or access
to finance. Other areas of interest for future research include how firms decide
on which area they focus with their investments, as well as why and how firms
choose to make different types of disclosures around those investments.
Finally, the authors believe that it would be helpful to extend their work
using ESG data from several providers, since past research has shown that such
ratings can differ substantially for the same firm.