Professional Documents
Culture Documents
a r t i c l e
i n f o
Article history:
Received 11 March 2010
Accepted 21 June 2010
Available online 6 July 2010
JEL classication:
G11
G12
G14
G19
a b s t r a c t
Empirical academic studies have consistently found that value stocks outperform glamour stocks and
the market as a whole. This article extends prevailing research on existing value anomalies. It evaluates
simple value strategies for the European stock market (compared to many other studies that test market
data on a country-by-country basis) as well as sophisticated multi-dimensional value strategies that also
include capital return variables (Consistent Earner Strategy) and momentum factors (Recognized Value
Strategy), the latter reconciling intermediate horizon momentum and long-term reversals of behavioral
nance theories. It can be shown that these enhanced value strategies can produce superior returns
compared to returns of the whole market or simple value strategies without capturing higher risks
applying traditional risk measures.
2010 The Board of Trustees of the University of Illinois. Published by Elsevier B.V. All rights reserved.
Keywords:
Behavioural Finance
Market Anomalies
Value Investing
1. Introduction
In their 1934 book, Security Analysis1 , Benjamin Graham and
David Dodd argued that out-of-favor stocks are sometimes underpriced in the marketplace, and that investors cognizant of this
phenomenon could capture strong returns. This philosophy is now
widely known as value investing. Although value investing has
taken many forms since its inception, it generally involves buying shares which appear underpriced based on some form(s) of
fundamental analysis. Value shares typically feature low price-tobook, price-to-earnings, or price-to-cash ow ratios, while glamour
stocks generally are characterized by valuation metrics at the opposite end of the spectrum.
As early as 1977, academic studies have used share price and
earning per share data to classify stocks into the value or glamour
categories and compare historical performance. Stocks with low
price-to-earnings multiples (often called value stocks) appear
to provide higher rates of return than stocks with high priceto-earnings ratios as rst shown by Nicholson (1960) and later
conrmed by Ball (1978), Basu (1977, 1983), and Fama and MacBeth
(1973).2 De Bondt and Thaler (1985) obtain a similar result for their
contrarian strategy based on buying stocks with low past returns
because of the behavioral hypothesis of investor overreaction. A
stocks price-to-book value ratio has also been found to be a useful predictor of future returns. Fama and French (1992) concluded
that size and price-to-book value together provide considerable
explanatory power for future returns in U.S. markets.
These results raised questions about the efciency of the market if one accepts the capital asset pricing model, as Lakonishok,
Schleifer and Vishny pointed out. In 1994, they published Contrarian Investment, Extrapolation, and Risk3 . Using data from
1968 to 1994, they grouped U.S. stocks into value and glamour
segments based on price-to-book, price-to-cash ow, and price-toearnings ratios, as well as sales growth. The researchers concluded
that, for a broad range of denitions of value and glamour,
value stocks consistently outperformed glamour stocks by wide
margins.
In their 1998 study, Value versus Growth: The International
Evidence, Fama and French tackled the question of whether value
stocks tended to outperform glamour stocks in markets outside
the U.S. The researchers found that, from 1975 to 1995, value
2
3
1062-9769/$ see front matter 2010 The Board of Trustees of the University of Illinois. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.qref.2010.06.005
528
4
Simliar results had been shown by Lakonishok, Hamao, and Chan (1991) for
Japan.
5
We widely follow Lakonishok et al. (1994) in the structure of our analysis.
2. Methodology
The sample period covered in this study starts on July 1, 19946
and ends June 30, 2009.7 For portfolio strategies that are tested over
2-year (3-year) performance horizons the last reformation date is
July 1, 2007 (July 1, 2006). As market proxy for the European stock
market the EuroStoxx index has been selected. Results are also veried for the EuroStoxx50 in order to verify that results still hold
if only large capitalization equities are examined. Results for stock
returns of indices containing only large rms are less contaminated
by signicant look-ahead or survivorship bias.8,9
Based on the index we form our model portfolios using as a rst
step one-dimensional (single) accounting ratios, such as dividend
yield (DY), price-to-book10 (P/B) and price-to-earnings11 (P/E). In
addition, corporate return (RoE) and momentum (Levy2712 , Relative Strength 3 months) ratios13 are computed for a Consistent
Earner Strategy (trying to mimic investment styles of successful
and well-known value investors who focus on outstanding companies at a sensible price)14 and a Recognized Value Strategy (trying
to further improve performance by timing reversals better based
on the stock momentum life cycle hypothesis).15 Then ratios and
historical performance data are used to sort individual stocks into
portfolios.16 Based on the investment strategy chosen, deciles are
formed for which performance is measured for 13-year time horizons. Within each of our portfolios, we equally weight all stocks
and compute returns using a buy-and-hold strategy for years t + 1,
t + 2 and t + 3 relative to the time of formation. If a stock is delisted
from the stock exchange during a year, we continue with the same
portfolio using the return of that stock at the time it was last traded
6
Results for starting dates on January 1, April 1 and October 1 were also tested
and results are comparable to conclusions drawn from yearly starting dates on July
1.
7
If the 30th is not a weekday, then the last trading day of the month is used. Years
in tables and graphs refer to a time period from July 1 that year until June 30 of the
subsequent year. Formation and reformation occur based on publicly available price
and accounting data from the previous year (t1). Results for current year estimates
as accessible at formation and reformation dates were comparable. Reformation at
the beginning of the second quarter was chosen in order to ensure that fundamental
company information for the entire previous year published in annual reports, SEC
lings or by the nancial media was available to investors and incorporated into
valuation ratios. 1994 was chosen as the rst formation year because the EuroStoxx
was created in 1999 and index constituents were recalculated back to this time.
8
Look-ahead and survivorship bias are common types of sample selection biases.
The rst is created by the use of information or data in a study or simulation that
would not have been known or available during the period being analyzed. This
will usually lead to inaccurate results in the study or simulation. To avoid this bias
we calculated ratios based on data available at the time of portfolio formation and
reformation, not from revisions published thereafter. The second bias occurs, for
example, when backtesting an investment strategy on a large group of stocks. Then it
may be convenient to look for securities that have data for the entire sample period.
However, eliminating a stock that stopped trading, or shortly left the market, would
input a bias in data samples. To avoid this problem we used historical constituent
lists for the EuroStoxx when we constructed our quantile portfolios.
9
Banz and Breen (1986), Kothari, Shanken, and Sloan (1992). La Porta (1993) also
points out that the selection bias is less serious among larger rms.
10
Rosenberg, Reid and Lanstein (1984) and Lee and Swaminathan (2000) published the rst cited study that evaluated the performance of the book-to- market
strategy.
11
Basu (1977) rst looked at the relationship between common stocks and their
price-to-earnings ratios. He found that a low P/E ratio portfolio earned 6% more per
year than a high P/E portfolio in the 14-year sample.
12
Levy, R. (1967).
13
RoE = Return on Equity, Levy27 = a stocks price divided by its price 27 weeks
earlier, Relative Strength 3 months = performance of a stock compared to the index
during the last 3 months (MO3m).
14
Greenblatt (2006).
15
Oyefeso (2004), Lee, C., & Swaminathan, B. (2000). Price momentum & trading
value. Journal of Finance.
16
As Lakonishok et al. (1994) we consider only positive ratios.
529
Table 1
Yearly average decile returns for portfolio strategies based on one-dimensional classications by various measures of valuea .
Panel A: DY (t0)
Return overview: Yearly
Value premium
Glamour
R1
R2
R3
ANN
CR3
AR
Value
10
101
5.07%
6.81%
7.20%
6.36%
20.30%
6.36%
9.89%
11.19%
5.56%
8.85%
28.97%
8.88%
8.01%
5.69%
7.19%
6.96%
22.36%
6.96%
10.42%
11.52%
9.56%
10.50%
34.91%
10.50%
9.07%
9.71%
8.14%
8.97%
29.41%
8.98%
9.42%
12.56%
9.39%
10.45%
34.74%
10.46%
11.16%
13.02%
10.41%
11.52%
38.71%
11.53%
12.86%
11.49%
7.67%
10.65%
35.48%
10.67%
8.28%
11.59%
15.75%
11.83%
39.85%
11.87%
13.66%
12.46%
15.48%
13.86%
47.61%
13.87%
8.60%
5.64%
8.28%
7.50%
27.31%
7.51%
Value premium
Glamour
R1
R2
R3
ANN
CR3
AR
Value
10
101
7.23%
7.64%
5.09%
6.65%
21.29%
6.65%
7.32%
8.70%
7.71%
7.91%
25.65%
7.91%
8.04%
7.70%
6.20%
7.31%
23.57%
7.31%
9.88%
10.44%
8.45%
9.59%
31.60%
9.59%
10.39%
9.22%
7.64%
9.08%
29.78%
9.08%
9.45%
9.88%
11.18%
10.17%
33.72%
10.17%
8.60%
11.40%
10.81%
10.26%
34.05%
10.27%
9.22%
12.88%
10.39%
10.82%
36.09%
10.83%
9.01%
9.87%
12.95%
10.60%
35.29%
10.61%
12.64%
18.99%
15.75%
15.76%
55.14%
15.79%
5.40%
11.35%
10.67%
9.12%
33.84%
9.14%
Value premium
Glamour
R1
R2
R3
ANN
CR3
AR
Value
10
101
1.11%
7.51%
6.60%
5.04%
15.88%
5.07%
5.87%
7.20%
9.26%
7.44%
24.01%
7.45%
9.46%
8.92%
5.78%
8.04%
26.11%
8.05%
11.03%
11.16%
8.72%
10.30%
34.18%
10.30%
10.00%
11.58%
7.69%
9.75%
32.18%
9.76%
9.49%
13.38%
5.94%
9.56%
31.51%
9.60%
12.74%
9.31%
11.22%
11.08%
37.06%
11.09%
9.04%
9.42%
10.12%
9.53%
31.39%
9.53%
11.12%
11.97%
12.95%
12.01%
40.54%
12.02%
13.76%
13.87%
13.39%
13.68%
46.89%
13.68%
12.66%
6.35%
6.79%
8.64%
31.01%
8.60%
Panels AC present the average percentage decile returns for one-dimensional value strategies formed in ascending order based on P/E and P/B; descending order based on
dividend yield (DY). The value portfolio refers to the decile portfolio containing stocks ranking lowest on P/E or P/B, or highest on dividend yield (DY). The glamour portfolio
contains stocks with precisely the opposite set of rankings. Portfolio reformation occurs yearly at the beginning of July during the period from 1994/95 to 2008/09. The
right-most column contains the value premium based on the performance difference between decile 10 and 1.
Panel A: percentage decile returns as described above for a one-dimensional value strategy based on dividend yield (DY).
Panel B: percentage decile returns as described above for a one-dimensional value strategy based on P/E.
Panel C: percentage decile returns as described above for a one-dimensional value strategy based on P/B.
a
R1 = 1st year return, R2 = 2nd year return, R3 = 3rd year return, ANN = annualized 3-year return, CR3 = compounded 3 year return, AR = average 3 year return, (t0) = refer
to annotation 15.
17
The reasons for stocks to be delisted from the stock exchange during a reformation period were mostly acquisitions by other corporations. The last traded
price therefore was taken to approximate real performance of that stock in a given
period. Return on cash proceeds during the remaining period was not included in
the performance calculation.
18
Performance data for year 2 and year 3 (assuming no reformation) is also indicated.
40%, and (1) top 30%) based on each of two variables. The sorts are
12 pairs of variables: DY and P/E, DY and P/B, P/E and P/B, RoE and
DY, RoE and P/E, RoE and P/B, DY and Levy27, DY and MO3m, P/E and
Levy27, P/E and MO3m, P/B and Levy27, P/B and MO3m. Depending
on the two variables being used for classication, the value portfolio either refers to the portfolio containing stocks ranked in the top
group (1) on both variables from among P/E, P/B (sorted in ascending order), or else the portfolio in the top group on one of those
variables or/and in the top group (1) on reversely sorted DY, RoE,
Levy 27, and MO3m. The glamour portfolio19 contains stocks with
precisely the opposite set of rankings. Portfolio formation and reformation occur yearly at the beginning of July during the period
from 1994/95 to 2008/09. For each quartile, performance is measured using the same procedure as for one-dimensional (single)
accounting ratios.
19
In the case of momentum variables, it is usually inappropriate to talk of value
and glamour criteria but for simplicity we refer to the value quantile for the top
parameter characteristics and to glamour for the bottom.
530
Graph 1. Yearly and 3-year annualized decile returns. Graph 1 shows the yearly percentage decile portfolio returns for years 13 and annualized for the entire 3-year period
for a one-dimensional value strategy based on dividend yield (DY). Portfolio reformation occurs yearly at the beginning of July during the period from 1994/95 to 2008/09.
20
Ratios in this article are based on actual data available at formation and reformation. Ratios in graphs and tables are therefore (for better differentiation) assigned
a (t0). Results for ratios based on estimated data were also tested. Outcomes are
comparable.
21
Though RoE (Return in Equity), Levy27 (stock price divided by its price 27 weeks
before) and Relative Strength 3 month (MO3m) constitute no value variable they
are used in combination with value variables for our two-dimensional Consistent
Earner Strategy and Recognized Value Strategy strategies. For their returns as standalone variable, refer to the appendix section.
22
Results of the t-statistic for the test of the hypothesis that the differences in
returns between value and glamour are equal to zero are presented in Table 3, page
14.
23
For simple return and momentum variables, return differences fall in a range
between 2.50% and 11.44%. Results are presented in the appendix section. Though it
was not part of the research question formulated in this study it nevertheless seems
worthwhile to mention that return differences for single capital and momentum
variables are not statistically signicant (page 14).
24
Comparable results are obtained for the EuroStoxx50 index. Lakonishok et al.
(1994) divide the U.S. market into the 50% largest/smallest corporations instead
and likewise conclude (after eliminating the size-effect) that results still hold when
analyzing only large corporations.
25
For decile performance graphs of different one and two-dimensional ratios refer
to the appendix section. Outcomes are comparable.
531
Table 2
Yearly average quantile returns for portfolio strategies based on two-dimensional classications by various value measures.
Return overview: Yearly
Value premium
Glamour
3/3
DY&P/E
DY&P/B
P/E&P/B
DY&RoE
P/E&RoE
P/B&RoE
DY&Levy27
DY&M03m
P/E&Levy27
P/E&M0m3
P/B&Levy27
P/B&M03m
Value
3/2
Multi value
6.50%
9.79%
8.91%
5.89%
6.77%
5.21%
Consistent
4.96%
8.68%
2.95%
12.18%
3.29%
5.44%
Recognized value
2.42%
6.69%
2.51%
8.03%
0.50%
8.25%
0.84%
8.25%
1.81%
5.29%
1.50%
7.32%
2/3
3/1
2/2
1/3
2/1
1/2
1/1
Total
1/13/3
8.33%
10.74%
10.09%
13.20%
8.00%
5.64%
10.81%
9.84%
10.60%
3.22%
10.64%
12.39%
10.37%
10.62%
12.21%
12.97%
14.08%
11.18%
12.38%
11.85%
10.98%
9.88%
9.89%
9.52%
5.89%
2.94%
4.20%
9.64%
7.75%
6.57%
10.62%
12.22%
9.28%
9.96%
10.85%
10.37%
11.10%
10.60%
9.73%
11.21%
10.72%
10.83%
12.12%
9.25%
11.59%
11.39%
11.46%
11.50%
9.89%
9.50%
9.19%
6.43%
8.51%
14.78%
5.85%
7.50%
5.92%
4.32%
6.40%
6.75%
13.12%
12.24%
7.16%
8.27%
8.96%
9.26%
8.91%
9.48%
8.96%
10.45%
8.86%
7.79%
8.19%
9.57%
10.39%
11.78%
13.43%
14.72%
14.79%
12.83%
11.39%
10.66%
10.88%
11.23%
11.72%
10.67%
15.59%
14.57%
13.99%
13.63%
18.86%
18.54%
16.54%
14.97%
12.88%
13.71%
9.79%
9.78%
9.33%
9.35%
9.22%
9.20%
16.44%
16.03%
17.05%
14.14%
11.08%
12.21%
Presents the yearly average percentage quantile returns for two-dimensional value strategies classied into nine groups of stocks by independently sorting in ascending/descending order into three arrays ((1) bottom 30%, (2) middle 40%, and (1) top 30%) based on each of two variables. The sorts are 12 pairs of variables: DY and P/E, DY
and P/B, P/E and P/B, RoE and DY, RoE and P/E, RoE and P/B, DY and Levy27, DY and MO3m, P/E and Levy27, P/E and MO3m, P/B and Levy27, P/B and MO3m. Depending on the
two variables being used for classication, the value portfolio either refers to the portfolio containing stocks ranked in the bottom group (1) on both variables from among
P/E, P/B (sorted in ascending order), or else the portfolio in the bottom group on one of those variables or/and in the bottom group (1) on reversely sorted dividend yield
(DY), capital return (RoE), Levy27, and MO3m. The glamour portfolio contains stocks with precisely the opposite set of rankings. Portfolios reformation occurs yearly at the
beginning of July during the period from 1994/95 to 2008/09. The right-most column contains the value premium based on the performance difference between group 1/1
and 3/3.
For two-dimensional value strategies based on two value variables (Multi Value Strategy) the yearly average return differences
(quantile 1/1 minus quantile 3/3) presented above fall in a range
between 2.94% and 5.89% depending on the variable combination
chosen. These strategies do not improve investment performance
compared to simple value strategies. Furthermore statistical significance is reduced26 (in fact, investment results get worse and are
statistically insignicant).27
Strategies based on combinations of value and capital return
variables28 (Consistent Earner Strategy) seem to result in investment returns comparable to single variable value strategies ranging
from 6.43% to 14.78%. Statistical signicance however improves.29
The Consistent Earner Strategy mimics investment styles of wellknow investors like Warren Buffett or Joel Greenblatt who further
developed the value investing concept by focusing on nding an
outstanding company at a sensible price or buying cheap and
good companies with competitive advantages indicated by a high
return on capital rather than generic companies at a bargain price
as originally promoted by Graham and Dodd.
Strategies based on combinations of value and momentum
variables (Recognized Value Strategy) do improve both investment performance and signicance compared to single variable
value strategies. Investment returns are in a range from 11.08%
to 17.05% annually. The Recognized Value Strategy is based on the
stock momentum life cycle hypothesis30 stating that stocks move
alternately through periods of relative glamour and neglect
attempting to reconcile intermediate horizon momentum and long
horizon-reversals of behavioral nance theories.
26
4. Risk evaluation
Two competing theories have been proposed to explain why
various investment strategies have produced higher returns in
the past. The capital asset pricing model (CAPM) relating risk and
expected return is grounded on the theory that rational investors
demand higher returns for higher risks. Serving as a common model
for the pricing of risky securities in the nancial industry, it takes
into account an assets sensitivity to non-diversiable risk (also
known as systematic risk or market risk), often referred to by the
measure beta (), as well as the expected return of the market
and the expected return of a theoretical risk-free asset. In contrast,
behavioral nance theories recognize a psychological element in
nancial decision-making, thus challenging traditional models that
assume investors will always weigh risk/return factors rationally
and act without bias.31 For example, the human tendency to avoid
admitting error, called fear of regret by psychologists, can cause
an investor to hold a losing stock too long or sell a winner too soon.
Similarly, investment choices are inuenced positively or negatively by attitudes toward wealth, by investor attention, mimicry
(herding instinct), etc. The premise of behavioral nance is that
taking psychological factors into account can enhance the effectiveness of investment strategies and explain the success of contrarian
and value strategies (described as anomalies or inefciencies in
standard nancial literature).
In this section it will be examined whether superior returns by
value portfolios constructed based on one- and two-dimensional
variables (Multiple Value, Consistent Earner, Recognized Value
strategies) formed using the EuroStoxx benchmark as the sample index (EuroStoxx Value Anomaly) can be explained by a higher
exposure to systematic risk. In the subsequent analysis we widely
follow the methodology used by Lakonishok et al. (1994).
Value stocks would be fundamentally riskier than glamour
stocks if they underperform glamour stocks in periods of severe
market declines in which the marginal utility of wealth is high,
31
532
Graph 2. Yearly return differences (value minus glamour): EuroStoxx. Graph 2 shows the yearly percentage portfolio return differences (value quantile (1/1) minus glamour
quantile (3/3)) based on the EuroStoxx index for a two-dimensional value classication scheme based on DY (t0) and P/E (t0). In addition, years in which the European market
index (EuroStoxx) declined are marked by N, years in which it rose are indicated by a +. Yearly data in the 1994/952008/09 period refers to the 12 months between the
beginning of July and the end of June the following year. Portfolio reformation occurs yearly.
32
Common portfolio measures are also indicated, however only to provide further
useful information.
33
Refer to the appendix section for a complete graphical overview of value difference evolutions based on all variable combinations.
34
The Sharp Ratio (SR) is a risk-adjusted measure developed by William F. Sharpe,
calculated using standard deviation and excess return to determine reward per unit
of risk. The higher the Sharpe ratio, the better a portfolios historical risk-adjusted
performance.
tively (35.0%, 43.3%, 30.0%). Results for single capital return and
momentum variables are comparable.35
Multi Value Strategy: Two-dimensional value strategies based
on simple value variable combinations (e.g. DY & P/E, DY & P/B, P/E
& P/B) generated negative performance differences over a 1-year
time horizon in 22, 22, and 23 instances respectively (36.7%, 36.7%,
38.3%). Performance differences for these value variable combinations however are not statistically signicant.
Consistent Earner Strategy: Two-dimensional value strategies
based on simple value and capital return variable combinations
(e.g. RoE & DY, RoE & P/E, RoE & P/B) generated negative performance differences over a 1-year time horizon in 23, 20, and 20
instances respectively (38.3%, 33.3%, 33.3%). Statistical signicance
increases compared to single value variables.
Recognized Value Strategy: Two-dimensional value strategies
based on simple value and momentum variable combinations (e.g.
DY & Levy27, DY & MO3m, P/E & Levy27, P/E & MO3m, P/B & Levy27,
P/B & MO3m) generated negative performance differences over a
1-year time horizon in 21, 19, 19, 25, 18, and 18 instances respectively (35.0%, 31.7%, 31.7%, 41.7%, 30.0%, 30.0%). Return differences
and statistical signicance both increase compared to single value
variables.
In most cases presented above, the magnitude of underperformance of value versus glamour was mostly small relative to the
mean outperformance, and return differences were negative during market declines in even fewer instances. Thus we can conclude
that downside risk is relatively low.
We proceed and examine the performance of one- and twodimensional strategies in extreme down markets on a monthly
basis thereby comparing the performance difference (value minus
glamour) in the worst months for the stock market as a whole.
Table 4, Panels AD, lists the performance of one- and twodimensional strategies based on DY and on combinations of DY
35
Note that value differences for single value variables are statistically signicant
at the 5.0% condence interval (exception: P/B). Results for single variables based
on capital return and momentum ratios are generally not statistically signicant.
533
Table 3
Quarterly average portfolio return differences (value minus glamour) based on various value portfolio strategies for one- and two-dimensional classication schemes
(EuroStoxx).
Q3 94
Q4 94
Q1 95
Q2 95
Q3 95
Q4 95
Q1 96
Q2 96
Q3 96
Q4 96
Q1 97
Q2 97
Q3 97
Q4 97
Q1 98
Q2 98
Q3 98
Q4 98
Q1 99
Q2 99
Q3 99
Q4 99
Q1 00
Q2 00
Q3 00
Q4 00
Q1 01
Q2 01
Q3 01
Q4 01
Q1 02
Q2 02
Q3 02
Q4 02
Q1 03
Q2 03
Q3 03
Q4 03
Q1 04
Q2 04
Q3 04
Q4 04
Q1 05
Q2 05
Q3 05
Q4 05
Q1 06
Q2 06
Q3 06
Q4 06
Q1 07
Q2 07
Q3 07
Q4 07
Q1 08
Q2 08
Q3 08
Q4 08
Q1 09
Q2 09
DY (tO)
PB (tO)
PE (tO)
RoE (tO)
Levy27
M03m
DY&PE
DY&PB
PE&PB
+
+
N
+
+
+
+
+
+
+
+
+
+
N
+
+
N
+
+
+
N
+
+
N
N
N
N
+
N
+
N
N
N
+
N
+
N
+
+
+
N
+
+
+
+
+
+
N
+
+
+
+
N
+
N
N
N
N
N
+
4.22%
3.11%
8.86%
10.63%
6.15%
11.56%
1.65%
3.05%
0.32%
0.25%
2.88%
1.70%
2.53%
0.16%
10.92%
14.15%
1.69%
8.11%
14.73%
8.14%
7.09%
18.40%
6.69%
7.47%
7.88%
27.42%
17.79%
7.12%
8.88%
1.28%
4.19%
11.89%
8.13%
2.11%
10.57%
5.13%
5.66%
2.52%
5.68%
4.46%
5.29%
7.02%
0.48%
2.74%
1.08%
1.50%
1.32%
3.21%
3.44%
0.56%
0.14%
0.43%
0.77%
0.18%
1.04%
2.15%
5.35%
1.51%
7.37%
6.64%
7.40%
0.09%
12.32%
2.42%
10.58%
5.01%
16.11%
0.58%
3.58%
6.65%
35.86%
3.30%
16.38%
0.56%
13.30%
8.86%
11.30%
4.27%
5.33%
1.85%
0.78%
32.52%
9.90%
10.72%
4.79%
36.22%
26.65%
3.30%
8.01%
9.22%
3.60%
13.80%
0.67%
7.56%
2.92%
0.45%
12.90%
1.14%
3.40%
3.94%
0.38%
3.90%
7.78%
0.63%
6.74%
1.49%
10.41%
1.51%
6.36%
2.57%
3.32%
0.66%
5.95%
6.67%
2.75%
4.71%
1.44%
10.37%
15.04%
9.53%
5.40%
4.70%
3.63%
9.94%
4.00%
9.23%
3.25%
4.94%
0.05%
2.72%
8.70%
10.96%
1.72%
1.53%
6.03%
4.04%
8.97%
7.42%
12.45%
6.72%
10.13%
17.44%
11.61%
5.12%
6.92%
25.93%
25.98%
8.23%
7.58%
7.72%
14.64%
16.84%
2.57%
3.79%
4.06%
3.36%
10.86%
4.43%
3.98%
2.27%
6.94%
1.98%
0.10%
7.11%
3.90%
4.71%
4.65%
4.40%
4.22%
2.46%
2.30%
9.94%
3.40%
7.57%
4.37%
3.41%
0.86%
4.79%
6.27%
10.08%
6.25%
5.61%
0.67%
5.98%
8.90%
9.43%
12.70%
5.64%
8.38%
0.96%
29.91%
11.35%
8.90%
3.68%
5.56%
4.29%
0.98%
4.71%
0.04%
0.31%
3.00%
13.48%
2.91%
2.67%
4.58%
3.85%
5.68%
3.83%
9.14%
2.26%
9.72%
8.28%
5.11%
10.72%
7.85%
7.21%
4.89%
1.17%
1.18%
5.25%
12.77%
1.68%
1.24%
6.56%
4.70%
5.33%
7.29%
6.45%
1.57%
1.07%
3.00%
2.28%
3.74%
8.53%
2.53%
2.47%
4.73%
15.61%
9.71%
8.18%
1.50%
2.88%
1.07%
0.08%
5.19%
3.45%
12.16%
6.58%
0.35%
8.62%
6.26%
2.77%
10.95%
4.15%
12.35%
2.74%
5.92%
16.19%
3.22%
0.53%
41.98%
1.31%
11.49%
11.11%
3.75%
6.57%
15.84%
2.19%
18.30%
33.65%
5.55%
21.09%
28.21%
16.23%
0.21%
18.58%
0.94%
8.49%
6.17%
0.30%
10.47%
0.62%
1.47%
4.89%
3.28%
3.50%
7.95%
0.91%
2.71%
4.49%
0.92%
2.67%
5.11%
7.86%
0.39%
13.28%
12.80%
0.11%
14.24%
5.51%
1.53%
2.39%
1.15%
1.15%
5.23%
6.88%
2.53%
2.02%
2.20%
1.45%
5.93%
6.43%
3.20%
2.78%
8.57%
4.12%
5.77%
2.28%
6.08%
8.46%
45.66%
1.89%
16.24%
13.66%
7.95%
30.22%
7.21%
9.59%
15.42%
20.98%
4.77%
14.46%
34.00%
20.66%
2.23%
22.71%
1.29%
4.10%
1.98%
3.08%
10.02%
2.66%
3.00%
2.11%
8.04%
2.44%
3.30%
4.23%
0.29%
0.01%
4.88%
4.88%
5.78%
9.18%
5.22%
16.21%
12.29%
7.00%
7.47%
9.73%
1.75%
0.02%
8.38%
8.77%
2.52%
7.53%
3.03%
0.50%
0.22%
0.58%
1.90%
2.08%
4.04%
2.21%
3.35%
4.53%
8.39%
12.20%
11.52%
3.11%
5.15%
15.85%
4.92%
5.25%
5.51%
19.06%
15.63%
4.14%
6.47%
2.40%
6.18%
11.29%
4.17%
1.44%
0.87%
3.44%
3.84%
0.72%
4.60%
2.52%
6.09%
3.61%
1.11%
4.09%
0.88%
4.53%
1.49%
2.76%
1.45%
0.32%
0.34%
1.75%
1.47%
1.84%
1.13%
3.68%
2.37%
5.91%
9.16%
10.98%
7.33%
1.82%
5.85%
2.26%
5.60%
8.33%
12.44%
4.85%
1.76%
1.30%
2.62%
1.08%
10.44%
5.21%
8.38%
10.53%
8.30%
12.03%
9.34%
1.85%
0.61%
20.65%
3.01%
8.20%
1.39%
20.60%
15.13%
2.70%
5.20%
2.96%
4.15%
9.84%
7.37%
0.28%
2.77%
0.17%
3.17%
1.17%
3.46%
1.09%
2.70%
4.73%
2.17%
0.98%
0.04%
3.31%
5.29%
0.60%
2.68%
2.44%
1.89%
0.44%
4.51%
5.54%
1.95%
9.06%
4.21%
7.19%
13.18%
9.38%
5.94%
2.49%
6.07%
0.75%
0.11%
3.78%
0.21%
1.94%
4.39%
4.01%
6.69%
1.16%
11.45%
5.03%
5.75%
6.93%
13.76%
10.73%
8.59%
1.82%
2.75%
30.90%
12.15%
10.71%
3.03%
25.50%
18.92%
4.23%
4.30%
0.21%
6.34%
12.55%
2.68%
5.66%
1.32%
5.66%
4.88%
0.31%
3.14%
3.70%
4.05%
4.30%
2.87%
2.87%
1.27%
7.84%
5.09%
0.82%
1.13%
0.70%
1.11%
3.14%
7.87%
8.20%
0.04%
5.58%
0.52%
8.23%
10.56%
12.55%
Mean
t-statistic
p-value
st. dev.
Number
st. dev. (ann.)
SR
2.20%
2.2875
0.0258
7.43%
60
14.87%
0.59
1.30%
0.9034
0.3700
11.15%
60
22.30%
0.23
3.21%,
3.1401
0.0026
7.92%
60
15.85%
0.81
1.63%
1.6815
0.0979
7.50%
60
15.00%
0.43
2.64%
1.8406
0.0707
11.09%
60
22.19%
0.48
2.46%
1.6572
0.1028
11.50%
60
23.00%
0.43
1.45%
1.8494
0.0694
6.05%
60
12.11%
0.48
0.56%
0.6185
0.5386
7.07%
60
14.15%
0.16
0.95%
0.8990
0.3723
8.22%
60
16.45%
0.23
534
Table 3 (Continued)
Value
premiums
1/13/3
Q3 94
Q4 94
Q1 95
Q2 95
Q3 95
Q4 95
Q1 96
Q2 96
Q3 96
Q4 96
Q1 97
Q2 97
Q3 97
Q4 97
Q1 98
Q2 98
Q3 98
Q4 98
Q1 99
Q2 99
Q3 99
Q4 99
Q1 00
Q2 00
Q3 00
Q4 00
Q1 01
Q2 01
Q3 01
Q4 01
Q1 02
Q2 02
Q3 02
Q4 02
Q1 03
Q2 03
Q3 03
Q4 03
Q1 04
Q2 04
Q3 04
Q4 04
Q1 05
Q2 05
Q3 05
Q4 05
Q1 06
Q2 06
Q3 06
Q4 06
Q1 07
Q2 07
Q3 07
Q4 07
Q1 08
Q2 08
Q3 08
Q4 08
Q1 09
Q2 09
+
+
N
+
+
+
+
+
+
+
+
+
+
N
+
+
N
+
+
+
N
+
+
N
N
N
N
+
N
+
N
N
N
+
N
+
N
+
+
+
N
+
+
+
+
+
+
N
+
+
+
+
N
+
N
N
N
N
N
+
Mean
t-statistic
p-value
st. dev.
Number
st. dev. (ann.)
SR
DY (tO) &
Levy27
11.85%
0.48%
10.59%
12.70%
2.33%
21.29%
7.43%
5.62%
2.91%
11.17%
2.87%
4.75%
4.47%
0.38%
26.67%
9.28%
0.66%
14.37%
18.95%
0.79%
1.64%
3.96%
1.30%
9.76%
6.54%
24.74%
14.88%
9.53%
16.31%
18.90%
9.53%
25.04%
26.31%
9.27%
9.98%
12.87%
2.17%
0.27%
2.32%
3.64%
10.01%
6.85%
2.88%
6.96%
2.79%
5.46%
5.24%
4.31%
0.68%
7.26%
7.63%
6.04%
1.34%
2.94%
7.17%
5.49%
12.87%
4.37%
10.00%
9.97%
3.60%
2.7318
0.0083
10.20%
60
20.39%
0.71
DY (tO) &
M03m
0.69%
7.50%
2.73%
10.10%
1.65%
14.61%
2.66%
4.84%
7.28%
13.35%
7.85%
13.05%
6.66%
0.20%
16.22%
13.24%
7.69%
3.92%
14.72%
1.15%
0.57%
5.77%
11.57%
9.79%
2.68%
27.31%
20.30%
8.96%
14.99%
22.13%
5.67%
18.76%
29.98%
13.45%
9.51%
13.07%
0.19%
9.85%
1.26%
1.75%
6.66%
5.28%
4.35%
2.26%
0.27%
4.76%
5.41%
2.65%
0.54%
0.47%
1.75%
2.09%
0.31%
1.87%
0.76%
8.22%
1.99%
13.97%
8.89%
21.10%
3.45%
2.6653
0.0099
10.04%
60
20.07%
0.69
PB (tO) &
Levy27
11.99%
1.64%
18.27%
10.56%
1.98%
16.99%
-10.71%
10.20%
9.71%
5.78%
0.89%
26.72%
18.03%
2.98%
29.21%
5.83%
15.47%
21.38%
1.16%
0.20%
5.60%
0.66%
9.99%
0.33%
11.17%
35.09%
10.56%
2.40%
13.16%
18.33%
4.74%
20.03%
18.30%
10.42%
7.04%
8.67%
7.15%
1.11%
0.78%
0.61%
13.13%
6.07%
3.06%
5.55%
4.89%
1.97%
9.91%
6.46%
3.63%
8.77%
4.86%
5.61%
2.45%
5.82%
2.01%
7.93%
7.27%
12.61%
3.91%
1.01%
2.54%
1.7398
0.0871
11.30%
60
22.60%
0.45
PB (tO) &
M03m
4.82%
3.51%
7.96%
0.26%
3.26%
4.53%
9.53%
4.58%
3.00%
7.72%
52.23%
7.75%
16.98%
6.86%
26.69%
5.80%
12.76%
4.73%
0.80%
1.24%
7.77%
4.61%
11.53%
4.72%
2.56%
35.95%
18.68%
0.26%
9.82%
13.59%
3.60%
17.53%
26.14%
14.57%
4.67%
12.45%
9.70%
4.96%
2.40%
2.58%
8.20%
3.90%
6.13%
4.46%
4.02%
3.20%
5.16%
0.92%
3.59%
5.08%
4.50%
0.61%
1.83%
3.01%
2.51%
7.82%
12.16%
15.35%
2.74%
2.40%
2.65%
1.7271
0.0894
11.91%
60
23.81%
0.45
PE (tO) &
Levy27
4.44%
0.39%
6.84%
1.67%
5.66%
16.35%
3.47%
7.76%
1.62%
7.08%
4.65%
3.70%
11.60%
8.67%
25.44%
1.93%
13.22%
17.45%
5.37%
9.09%
0.96%
1.88%
2.91%
3.28%
11.86%
12.51%
16.07%
8.23%
13.83%
13.28%
10.03%
22.49%
17.09%
13.54%
5.76%
7.31%
8.72%
1.34%
8.01%
8.29%
9.87%
4.72%
5.88%
11.07%
6.20%
7.48%
12.34%
2.14%
0.50%
10.27%
8.25%
6.33%
4.78%
7.10%
2.98%
2.68%
20.98%
3.08%
8.09%
5.46%
3.88%
3.3196
0.0015
9.06%
60
18.13%
0.86
PE (t0)
& M03m
7.83%
7.93%
9.71%
0.26%
9.02%
0.64%
9.60%
2.86%
6.10%
7.39%
2.50%
13.83%
6.40%
4.18%
20.01%
4.09%
9.19%
9.10%
8.96%
1.55%
5.22%
7.20%
7.79%
5.40%
1.40%
28.28%
18.72%
6.50%
9.39%
7.57%
9.04%
18.99%
20.57%
14.92%
5.18%
10.79%
13.43%
0.31%
2.06%
1.84%
7.56%
1.91%
1.41%
6.53%
0.45%
4.80%
5.01%
0.89%
3.05%
5.94%
2.26%
2.03%
0.43%
8.27%
0.82%
12.75%
14.62%
9.23%
7.84%
3.90%
3.24%
2.8646
0.0058
8.76%
60
17.52%
0.74
Quarterly percentage portfolio return differences (value minus glamour) based on portfolios formed from EuroStoxx constituents using different one- and two-dimensional
value classication schemes. Portfolios reformation occurs yearly at the beginning of July during the period from 1994/95 to 2008/09. In addition, years in which the market
indexes declined are marked by N, years in which it rose are indicated by a +. The t-statistic for the test of the hypothesis that the differences in returns between value
and glamour are equal to zero, the corresponding p-value and the quarterly standard deviation of returns are indicated in the bottom lines. The return differences standard
deviation (annualized) and the Sharp Ratio (SR) are indicated for convenience as further useful information.
535
Table 4
Average monthly returns in different market environments: worst 25 down months, next 50 down months, positive 80 up months, best 25 up months.
Panel A: Dividend yield
W25
N53
P80
B25
10
Eq.-w.
101
t-statistic
p-Value
0.0954
0.0247
0.0209
0.0874
0.0891
0.0202
0.0278
0.0803
0.0939
0.0198
0.0251
0.0840
0.0851
0.0150
0.0254
0.0790
0.0852
0.0200
0.0247
0.0813
0.0848
0.0188
0.0256
0.0788
0.0793
0.0119
0.0261
0.0682
0.0842
0.0108
0.0271
0.0766
0.0844
0.0159
0.0240
0.0716
0.0720
0.0127
0.0279
0.0721
0.0852
0.0168
0.0255
0.0778
0.0234
0.0120
0.0070
0.0153
3.0424
2.0129
2.3613
1.5309
0.0056
0.0506
0.0205
0.1389
W25
N50
P80
B25
3/3
3/2
2/3
3/1
2/2
1/3
2/1
1/2
1/1
Eq.-w.
Index
1/13/3
t-statistic
p-Value
0.0946
0.0268
0.0194
0.0972
0.0900
0.0209
0.0269
0.0791
0.0937
0.0202
0.0253
0.0909
0.0921
0.0203
0.0287
0.0831
0.0822
0.0141
0.0262
0.0695
0.0784
0.0164
0.0237
0.0815
0.0773
0.0164
0.0255
0.0779
0.0811
0.0132
0.0288
0.0691
0.0784
0.0105
0.0261
0.0687
0.0855
0.0173
0.0257
0.0786
0.0000
0.0000
0.0000
0.0000
0.0162
0.0163
0.0067
0.0285
2.6067
3.3483
2.4413
3.6663
0.0155
0.0017
0.0167
0.0012
W25
N50
P80
B25
3/3
3/2
2/3
3/1
2/2
1/3
2/1
1/2
1/1
Eq.-w.
1/13/3
t-statistic
p-value
0.1071
0.0309
0.0209
0.1026
0.0872
0.0184
0.0224
0.0782
0.1021
0.0182
0.0225
0.0887
0.0837
0.0155
0.0293
0.0775
0.0804
0.0177
0.0254
0.0729
0.0959
0.0180
0.0265
0.0817
0.0743
0.0109
0.0285
0.0723
0.0727
0.0117
0.0256
0.0677
0.0692
0.0038
0.0290
0.0769
0.0853
0.0164
0.0257
0.0781
0.0379
0.0270
0.0080
0.0257
2.5519
3.3124
2.1662
1.7922
0.0175
0.0019
0.0331
0.0857
W25
N50
P80
B25
3/3
3/2
2/3
3/1
2/2
1/3
2/1
1/2
1/1
Eq.-w.
1/13/3
t-statistic
p-Value
0.1112
0.0300
0.0221
0.1042
0.0895
0.0193
0.0261
0.0785
0.1005
0.0182
0.0257
0.0874
0.0773
0.0157
0.0268
0.0778
0.0833
0.0142
0.0256
0.0730
0.0972
0.0155
0.0276
0.0843
0.0734
0.0151
0.0274
0.0746
0.0758
0.0113
0.0257
0.0657
0.0581
0.0072
0.0289
0.0719
0.0853
0.0161
0.0261
0.0779
0.0531
0.0229
0.0068
0.0323
4.0470
3.5922
1.7503
1.9776
0.0005
0.0008
0.0837
0.0596
Panels AD present the average percentage decile portfolio returns, the average total performance, the return differences (value minus glamour) on a monthly basis
representatively for one-dimensional value strategies based on DY; for two-dimensional strategies based on DY and RoE, DY and Levy27, DY and MO3m. The performance
of our portfolios are divided into four states of general market environments; the 25 worst stock return months in the sample based on the equally weighted index, the
remaining 50 negative months other than the 25 worst, the 80 positive up months other than the 25 best, and the 25 best up months in the sample. The average difference
in returns between value and glamour for each state is also reported along with t-statistics for the test that the differences of returns are equal to zero and its corresponding
p-value.
and RoE, Levy27 and MO3m during four states of the global market; the 25 worst stock return months36 in the sample based on the
equally weighted index, the remaining 50 negative months other
than the 25 worst, the 80 positive months other than the 25 best,
and the 25 best months in the sample. The average difference in
returns between the value and glamour portfolio for each state is
also reported along with t-statistics for the test that the difference
of returns is equal to zero and its corresponding p-value.37
The results in this table are ambiguous (due to low t-statistics).
While for most classication schemes, the value portfolio outperformed glamour in the markets worst 25 months and in the next
50 negative months, results in all cases are only statistically significant for the category Next negative 50 months.38 For example,
using DY, the value portfolio lost an average of 7.2% of its value
in the worst 25 months, whereas glamour lost 9.5% of its value.
Similarly, the value portfolio outperformed glamour in the next
worst 50 months in which the index declined. It lost 1.5% in these
months while glamour experienced a 2.5% decline. Similar results
can be observed for other one- and two-dimensional value strategies as well (refer to the appendix section). The value strategy did
generally better when the market fell. In the 80 positive months
other than the best 25, one- and two-dimensional value strategies
36
39
These results are consisted with earlier observations. For a complete list of all
variables examined refer to the appendix section.
40
Refer also to the appendix section for variable combinations other than with
dividend yield.
536
Table 5
Traditional risk measures: beta, standard deviation.
Panel A: Dividend yield
Beta ann.
St. dev. (ann)
10
Eq.-w.
1.0836
0.2098
1.0481
0.1994
1.0962
0.2050
0.9987
0.1860
1.0077
0.1870
1.0012
0.1862
0.8911
0.1671
0.9784
0.1835
0.9907
0.1852
0.9265
0.1801
0.9079
0.1796
Beta ann.
St. dev. (ann)
3/3
3/2
2/3
3/1
2/2
1/3
2/1
1/2
1/1
Eq.-w.
1.1358
0.2187
1.0360
0.1947
1.1008
0.2059
1.0642
0.2068
0.9211
0.1723
0.9837
0.1926
0.9364
0.1819
0.9544
0.1793
0.9199
0.1776
0.9122
0.1742
3/3
3/2
2/3
3/1
2/2
1/3
2/1
1/2
1/1
Eq.-w.
1.2331
0.2477
1.0090
0.1899
1.1042
0.2119
0.9885
0.1927
0.9497
0.1773
1.1074
0.2105
0.9047
0.1735
0.8805
0.1668
0.9046
0.2030
0.9093
0.1914
3/3
3/2
2/3
3/1
2/2
1/3
2/1
1/2
1/1
Eq.-w.
1.2773
0.2550
1.0326
0.1930
1.1206
0.2141
0.9496
0.1872
0.9590
0.1774
1.1341
0.2181
0.8966
0.1745
0.8920
0.1691
0.7953
0.1743
0.9088
0.1920
Beta ann.
St. dev. (ann)
Beta ann.
St. dev. (ann)
Panel A shows the beta with respect to the European market index EuroStoxx, the decile return standard deviations and the return value premiums standard deviations
(annualized = ann.) for various one-dimensional value strategies (DY) based on quarterly performance data. Portfolio reformation occurs yearly at the beginning of July
during the period from 1994/95 to 2008/09. Panels BD show the beta with respect to the European market index EuroStoxx, the quantile return standard deviations and
the return value premiums standard deviations (annualized = ann.) for various two-dimensional value strategies (DY and RoE, DY and Levy27, DY and MO3m) based on
quarterly performance data. Portfolio reformation occurs yearly at the beginning of July during the period from 1994/95 to 2008/09.
41
This conclusion is not valid for value strategies based on P/B where betas for the
value quantile portfolios seem to be substantially higher than for glamour portfolios.
42
It seems however that glamour and value strategies generally exhibit somewhat higher standard deviations than other portfolio quantiles or the total equally
weighted index.
43
44
Momentum criteria here included representatively Levy27 and relative strength
3 month (MO3m).
45
Lee & Swaminathan (2000). Price momentum & trading value. Journal of Finance.
537