Morningstar - Q4 2021 - 45

EXHIBIT 1
ESG Risk Rating by Sector Average
ratings generally reflect where we
would expect to find the most and least
sustainable companies.
Sector
u Real Estate
t Consumer Discretionary
i Communication Services
a Information Technology
y Financials
s Consumer Staples
p Industrials
d Healthcare
r Materials
f Utilities
o Energy
Source: Morningstar. Data as of 07/31/2021.
Average ESG
Risk Rating
19.68
20.51
21.25
22.66
29.15
29.17
29.33
29.47
36.74
37.96
44.26
in Xiong (2021). To summarize, the regression
found that the monthly return of stocks increased
(or decreased) by 0.015% when the ESG Risk
Rating was improved (or lowered) by 1 over the
time period. In other words, the lower the
ESG Risk Rating, the higher the future return.
this question, I formed five portfolios based on the
rating. I ranked the stocks from lowest to highest
ESG Risk Rating every month and sorted them into
quintiles. Quintile 1 contained the lowest ratings
(the most sustainable stocks); quintile 5 had the
highest. I then averaged each quintile's price/book
ratios and equal-weighted market caps over the
entire period from August 2009 to November 2020.
EXHIBIT 2 shows the average characteristics of the
ESG Risk Rating quintiles. Stocks with a lower
ESG Risk Rating tended to have higher price/book
ratios and vice versa. This suggests that stocks
with a low ESG Risk Rating are slightly growthoriented.
However, the relationship between ESG
Risk Rating and market cap was mixed at best.
The Performance of ESG Risk Ratings
A critical question is whether the ESG Risk Rating
can forecast both returns and downside protection.
I studied the predictive power of the rating on
stock performance using two methods: (1) a time
series of cross-sectional regressions and
(2) building quintile portfolios sorted on the rating
and evaluating their future performance. The
technical details of the findings using the
first method (cross-sectional regressions) are listed
Next, I investigated the rating's economic
performance on a portfolio setting. I returned to
our ESG Risk Rating quintiles to form portfolios.
EXHIBIT 3 shows the performance measures
for the five portfolios; returns are market-cap
weighted. The relationship between performance
and risk rating was not perfect: The geometric
mean slowly increased through quintile 3 and then
decreased in quintiles 4 and 5. The first three
quintiles behaved similarly using the other
performance measures, and all three performed
better than quintiles 4 and 5. Comparing quintile 1
with quintile 5, the annual arithmetic return was
6.81 percentage points higher (16.32% versus
9.51%), the standard deviation was 3.71 percentage
points lower (14.51% versus 18.22%), and the
Sharpe ratio was 0.59 percentage point higher
(1.08 versus 0.49).
Downside, or tail, risks are measured by
skewness, kurtosis, and maximum drawdown. All
three measures were more monotonic than other
performance measures. Together, negative
skewness and large kurtosis corresponded to large
tail losses, and the combined two show that
quintile 5 suffered the worst loss and quintile
1 the least. Max drawdown was the smallest for
quintile 1 (negative 18.86%) and largest for
quintile 5 (negative 37.56%), a whopping 18.70percentage-point
difference. Note that the
maximum drawdown for each portfolio occurred
during the COVID-19 pandemic, from January
2020 to March 2020.
EXHIBIT 4 shows the growth of $1 for the five
portfolios. Over the 11-year period, quintile
5, the least sustainable portfolio, underperformed
the other four, particularly quintiles 1, 2, and 3.
The Impact of the Energy Sector
It is well known that energy stocks have performed
poorly for most of the past decade. How much
did the sector account for the underperformance
of the portfolio with the highest ESG Risk Ratings?
To find out, I used the same two methods as
above, with some modifications, to single out the
impact of the energy sector.
Like before, the technical results of the crosssectional
regression method are detailed in
Xiong (2021). To summarize, I chose utilities as
the base sector. I found that the information
technology sector had the most positive coefficient,
while energy had the most negative coefficient.
On average, stocks in the energy sector
underperformed the base utilities sector by 0.68%
monthly. In contrast, stocks in the information
technology sector outperformed by 0.32% monthly.
The coefficient on ESG risk was cut by more
than half compared with our first study and was
no longer statistically significant, indicating
that sectors played an important role in explaining
returns. In other words, the energy sector's
underperformance was responsible for the main
part of the underperformance of the quintile
with the highest ESG Risk Ratings.
To form the five ESG-risk-sorted portfolios,
I excluded energy stocks. The performance of the
quintiles excluding energy stocks is shown
in EXHIBIT 5 . The difference in arithmetic mean
between the lowest (quintile 1) and highest
(quintile 5) ESG risk quintiles shrank to 2.61
percentage points-62% less than the performance
gap shown in EXHIBIT 3. It was still positive
but not statistically significant. The difference in
max drawdown was now 7.87 percentage
points, reduced from a difference of 18.70
percentage points in EXHIBIT 3 . Skewness and
kurtosis measures tell the same story-
that the downside or tail risk protection was
smaller but still meaningful.
Overall, the results of both studies are similar
in the sense that the direction of most
outperformance measures is the same, but the
magnitude or significance is less because
sectors such as energy play an important role
in performance differences.
Explaining the Trends
It makes sense that stocks with a low ESG
Risk Rating provided better downside protection
than high-rated stocks. A low rating signals
morningstar.com/lp/magazine
45
http://www.morningstar.com/lp/magazine
Morningstar - Q4 2021

Table of Contents for the Digital Edition of Morningstar - Q4 2021
