Popularity
Page 12
Table 6.3 presents various performance summary statistics for the cap-weighted four quartiles. The last two rows show the alphas against the Carhart four factors. The results are generally consistent with Table 6.2 , but the t -statistics are less significant.
Overall, based on brand, both the equally weighted composites and cap-weighted composites are consistent with the popularity thesis and are in contrast to the risk–return paradigm.
Figure 6.1. Growth of $1 for Equally Weighted Quartiles Based on Interbrand’s BV Rankings, April 2000–August 2017 (log scale)
______________________________
Table 6.3. Summary Statistics of Cap-Weighted Quartile Returns Based on Interbrand’s BV Rankings, April 2000–August 2017
Measure
Q4
(lowest BV)
Q3
Q2
Q1
(highest BV)
Geometric mean (%)
7.27
5.00
3.12
3.79
Arithmetic mean (%)
8.30
7.02
4.48
5.40
Standard deviation (%)
13.85
19.38
16.18
17.50
Sharpe ratio
0.479
0.277
0.178
0.216
Skewness
–0.626
–0.624
–0.352
–0.401
Carhart’s alpha (%)
2.58
0.66
–1.12
1.02
t -stat. of alpha
1.52
0.31
–0.61
0.58
Competitive Advantage. Morningstar’s economic moat idea is an estimate of a company’s sustainable competitive advantage—a characteristic that investors nearly uniformly agree is good. To calculate economic moats, Morningstar’s equity analysts evaluate and estimate a company’s sustainable competitive advantage on the basis of five criteria: network effect, intangible assets, cost advantage, switching costs, and efficient scale. Based on the equity analysts’ assessment of each company with respect to one or more of these criteria, each company is given a rating of wide moat, narrow moat, or no moat. Morningstar started assigning moat ratings in 2002. The economic moat idea measures the quality of a business and has nothing to do with whether or not the security in question is priced fairly.
The period of our analysis of economic moat and returns is July 2002 through August 2017. The number of companies that had economic moat ratings in our sample varied over time—with a minimum of 427 (July 2002), a maximum of 1,611 (November 2008), and an average of 1,039 companies. Figure 6.2 shows the distribution of economic moat ratings through time. Note that the number of stocks for each of the three moat ratings is not the same.
Starting in July 2002, we placed each stock in the universe of moat-rated stocks into three groups, based on their moat ratings. We updated the three ratings groups each month on the basis of the most recent publicly available moat ratings. We believe that investors prefer companies they consider to have a sustainable competitive advantage. Thus, wide-moat companies represent the most popular stocks and no-moat companies represent the least popular stocks. Table 6.4 presents various performance summary statistics for the equally weighted portfolios corresponding to the three moat ratings.
In Table 6.4 , as we move from no-moat companies with, on average, the lowest sustainable competitive advantage (least popular companies) to the wide-moat companies with, on average, the highest sustainable competitive advantage (most unpopular companies), we find that the no-moat companies produced the most superior arithmetic and geometric mean returns. This additional return came, however, with more risk.
The last two rows of Table 6.4 show Jensen’s annualized alpha and its t -statistic. We used the equally weighted returns for all stocks across the three moat ratings as the market return and ran simple regressions to get Jensen’s annualized alpha. The annualized alphas are –1.59% for the no-moat companies and 0.95% for the wide-moat companies, which is the opposite of what one might expect given that no-moat companies had a much higher geometric return than wide-moat companies. Neither of the alphas is statistically significant at the 5% level.
Figure 6.2. Distribution of Economic Moat Ratings, July 2002–August 2017
Figure 6.3 shows the growth of a $1 investment for each of the three moat ratings. For this historical period, in up markets, the lower the sustainable competitive advantage, the better the returns. During down markets, however, such as the 2008 financial crisis and the more minor downturns in 2011 and 2015, the greater the sustainable competitive advantage, the milder the downturn.
______________________________
Table 6.4. Summary Statistics for Equally Weighted Portfolios Based on Morningstar’s Economic Moat Ratings, July 2002–August 2017
Measure
No Moat
Narrow Moat
Wide Moat
Geometric mean (%)
15.40
12.08
11.15
Arithmetic mean (%)
18.57
13.52
12.13
Standard deviation (%)
23.59
16.08
13.26
Sharpe ratio
0.729
0.758
0.815
Skewness
0.169
–0.400
–0.565
Jensen’s alpha (%)
–1.59
–0.28
0.95
t -stat. of alpha
–1.11
–0.52
1.21
Figure 6.3. Growth of $1 for the Three Equally Weighted Portfolios Based on Morningstar Economic Moat Ratings, July 2002–August 2017 (log scale)
Table 6.5 presents various performance summary statistics for the cap-weighted portfolios based on the three moat ratings. The last two rows show the alphas against the Carhart four factors. In general, the results in Table 6.5 are consistent with those in Table 6.4 .
Overall, based on competitive advantage, both the equally weighted composites and cap-weighted composites are consistent with both the popularity and the risk–return paradigms. The composites with no competitive advantage (unpopular stocks) produced the highest geometric returns but did so with the highest amount of risk.
Reputation. First introduced in 1999, the Harris Poll reputation quotient (Harris Poll 2015 ) has used a consistent methodology to measure the “reputation” of the most visible US companies for the last 15 years. The Harris Poll study involves two phases. It starts with a “nomination phase” in which the general public is asked to identify the US companies with the best and worst reputations. In this phase, 100 companies with the highest or best reputations are identified. Prior to 2015, only the 60 best companies were identified. For the 2015 study, 4,034 interviews were conducted to identify the top 100 companies. In the second phase, the “rating phase,” on the basis of a 20-minute interview (with approximately 250 adults participating in the interviews), each of the companies is rated on 20 attributes (classified into six categories). We treated the Harris Poll reputation quotient as a proxy for a dimension of popularity.
Table 6.5. Summary Statistics for the Cap-Weighted Portfolios Based on Morningstar’s Economic Moat Ratings, July 2002–August 2017
Measure
No Moat
Narrow Moat
Wide Moat
Geometric mean (%)
9.94
9.88
8.37
Arithmetic mean (%)
12.13
11.14
9.20
Standard deviation (%)
19.76
15.10
12.40
Sharpe ratio
0.548
0.652
0.639
Skewness
–0.745
–0.563
Carhart’s alpha (%)
–0.47
0.71
0.75
t -stat. of alpha
–0.39
1.13
0.85
The Harris Poll reputation quotient (RQ) is most similar in spirit to Fortune ’s most admired companies, although rather than polling the general public, the Fortune survey polls senior executives, directors, and securities analysts when identifying the most admired companies. The Harris Poll RQ may also seem similar to Interbrand’s BV rankings, but the Harris Poll is survey based and focuses on reputation, whereas the Interbrand rankings are valuation based and focus on estimated brand value. Thus, the two ranking organizations are measuring two distinct characteristics of a company.
In contrast to Interbrand BV ranks, in which multiple brands from a single company may appear on their list, the Harris Poll collapses subsidiaries and brands into a single parent company. As in previous analyses, we manually linked each publicly traded company to an appropriate stock ticker and removed privately held companies. We linked the Harris Poll RQ to tickers on the NYSE and NASDAQ and to ADRs.
Table 6.6 shows the top-ranked RQ companies, their rank, and their RQs for the first year of the poll (2000) and the 2017 poll. The constituents and the rank orders have changed substantially over time.
Mirroring the process that we applied to the Interbrand brand values, we used RQ ranks to form quartiles. The updated annual RQ ranks are typically released in February; thus, we delayed quartile formation until April of each year. Again, Quartile 1 contains the 25% of companies with the highest RQs (the most popular companies) and Quartile 4 contains the 25% of companies with the lowest RQs (the least popular companies).
Table 6.6. Harris Poll RQ Top 40 for 2000 and Top 50 for 2017
2000
2017
Rank
Company Name
RQ Score
Rank
Company Name
RQ Score
1
Johnson & Johnson
83.4
1
Amazon.com
86.3
2
Coca-Cola Co.
81.6
2
Wegman’s Food Markets
85.4
3
Hewlett-Packard Co.
81.2
3
SpaceX
83.1
4
Ben & Jerry’s Ice Cream
81.0
4
Publix Supermarkets
82.8
5
Intel Corp.
81.0
5
Johnson & Johnson
82.6
6
Walmart
80.5
6
Apple
82.1
7
Xerox
79.9
7
UPS
82.1
8
Home Depot
79.7
8
Walt Disney Co.
82.0
9
Gateway
78.8
9
Google
82.0
10
Walt Disney Co.
78.7
10
Tesla Motors
81.7
11
Dell
78.4
11
3M Co.
81.5
12
General Electric
78.1
12
USAA
81.4
13
Anheuser-Busch
78.0
13
Alphabet Inc.
81.3
14
Lucent Technologies
78.0
14
Coca-Cola Co.
81.2
15
Microsoft
77.9
15
General Mills
81.2
16
Amazon
77.8
16
Costco
81.1
17
IBM
77.6
17
Clorox Co.
81.1
18
Sony
77.4
18
Under Armour
80.7
19
Yahoo!
76.9
19
Toyota Motor Corp.
80.2
20
FedEx Corp.
75.7
20
L.L. Bean
80.1
21
AT&T
75.7
21
Netflix
79.9
22
Procter & Gamble Co.
71.9
22
SC Johnson
79.7
23
Nike
71.3
23
Lowe’s
79.7
24
McDonald’s
71.2
24
Microsoft
79.3
25
Southwest Airlines
70.6
25
Kroger Co.
79.2
26
AOL
69.2
26
Berkshire Hathaway
79.2
27
Fiat Chrysler Automobiles
69.1
27
PayPal
79.0
28
Toyota Motor Corp.
68.6
28
FedEx Corp.
79.0
29
Sears Holdings Corp.
67.6
29
Kimberly-Clark Corp.
78.9
30
Boeing Co.
67.3
30
LG Corp.
78.8
31
Texaco
67.3
31
Boeing Co.
78.7
32
Ford Motor Co.
66.9
32
Meijer’s
78.7
33
General Motors
63.1
33
Southwest Airlines
78.6
34
Apple
62.1
34
Chick-fil-A
78.5
35
MCI/WorldCom
62
35
BMW
78.2
36
ExxonMobil
61.6
36
Vanguard Group
78.2
37
Kmart
60.5
37
Nestle
78.1
38
Bank of America
58.8
38
Whirlpool Corp.
78.0
39
Amway
58.1
39
General Electric
77.9
40
Philip Morris Companies
57.2
40
Prosper
77.9
41
Hewlett-Packard Co.
77.8
42
Honda Motor Co.
77.8
43
Subaru
77.6
44
Kellogg Co.
77.6
45
Hobby Lobby
77.5
46
Nike
77.5
47
Visa
77
.3
48
Whole Foods Market
77.2
49
Mondelez International
77.1
50
Walgreens
77.1
Table 6.7 and Figure 6.4 present, respectively, summary statistics and the growth of a $1 investment in each RQ quartile.
In Table 6.7 , we see as we move from Q4 containing the stocks with the lowest RQs (most unpopular stocks) through the various quartiles to Q1 containing the stocks with the highest RQs (most popular stocks) that the lower-RQ quartiles outperformed the higher-RQ quartiles on the basis of arithmetic return. Q4 with the lowest RQs had a significantly better geometric return and Sharpe ratio than the other quartiles, but Q4 also had the highest standard deviation.
Table 6.7. Summary Statistics of Equally Weighted Quartile Portfolios Based on Harris Poll RQ Rankings, April 2000–August 2017
Measure
Q4
(lowest RQ value)
Q3
Q2
Q1
(highest RQ value)
Geometric mean (%)
12.61
7.14
5.66
7.02
Arithmetic mean (%)
14.73
8.97
7.50
8.41
Standard deviation (%)
19.51
18.56
18.56
16.10
Sharpe ratio
0.665
0.393
0.315
0.418
Skewness
–0.038
0.122
–0.413
–0.344
Jensen’s alpha (%)
4.01
–1.09
–2.56
–0.25
t -stat. of alpha
1.88