Staying the Course: The Impact of Investment Style Consistency

on Mutual Fund Performance

Keith C. Brown*
Department of Finance
McCombs School of Business
University of Texas
Austin, Texas 78712
(512) 471-6520
E-mail: kcbrown@mail.utexas.edu

W. V. Harlow
Fidelity Investments
82 Devonshire Street
Boston, Massachusetts 02109
(617) 563-2673
E-mail: van.harlow@fmr.com

First Draft: November 29, 2001

This Draft: March 8, 2002

* Corresponding Author. We are grateful to William Goetzmann, Russ Wermers,

Sheridan Titman, Dave Chapman, Andres Almazan, Mark Carhart, Robert Litterman,
Bob Jones, as well as the participants at the University of Texas Finance seminar, the
Goldman Sachs Asset Management seminar, and the 2001 Columbine/Instinet Investment
Conference in Colorado Springs, Colorado. We also thank Xuehai En and Andras
Marosi for computational assistance and research support. The opinions and analyses
presented herein are those of the authors and do not necessarily represent the views of
Fidelity Investments.
 Staying the Course: The Impact of Investment Style Consistency
on Mutual Fund Performance

ABSTRACT

While a mutual fund’s investment style influences the returns it generates, little is known
about how a manager’s execution of the style decision might affect performance. Using
multivariate techniques for measuring the consistency of a portfolio’s investment
mandate, we demonstrate that more style-consistent funds tend to produce higher total
and relative returns than less consistent funds, after controlling for past performance and
portfolio turnover. These findings are robust across fund investment style classifications,
the return measurement period, and the model used to calculate expected returns. We
document a positive relationship between measures of fund style consistency and the
persistence of its future performance, net of momentum and past performance effects.
We conclude that the decision to maintain a consistent investment style is an important
aspect of the portfolio management process.

The return performance of public investment companies has for decades been one of the
most widely studied topics in all of finance. Given the tremendous volume of products
and assets under management represented by the mutual fund industry, there is little
wonder why this should be the case. For instance, a recent survey by the Investment
Company Institute, a trade association of mutual fund companies, revealed that by year-
end 2000 there were almost 8,200 funds with net asset values totaling approximately
$6.97 trillion held in more than 240 million shareholder accounts. Moreover, roughly $4
trillion of these assets were controlled by the almost 4,400 separate investment
companies composed exclusively of equity investments.1 Without question, this scale
and scope of activity makes fertile ground for financial economic research.
Sharpe (1966) and Jensen (1968) were among the first to examine the performance
of mutual fund managers. Those studies, both of which compare individual fund
performance to that of the overall stock market, reach the conclusion that the average
fund manager does not possess superior skill and what positive performance that did exist
does not persist. Carlson (1970), on the other hand, argues that conclusions about
relative fund performance depend on which market benchmark is used in the comparison.
In particular, he shows that most fund groups outperformed the Dow Jones Industrial

1
These statistics are listed in the data glossary of the 2001 Mutual Fund Fact Book, an annual publication
of the Institute. It is also interesting to note that the number of equity mutual funds that existed at the end
of 2000 (i.e., 4,395) exceeded the number of companies with shares traded on the New York Stock
Exchange (i.e., 2,862).
Average, but were unable to match the returns posted by the Standard & Poor’s 500 and
NYSE Composite indexes. To the extent that these various indicators of market
performance represent portfolios containing securities with substantively different
characteristics, Carlson’s work stands as one of the first suggestions that investment style
can have a significant impact on how investment performance is measured.2
Of course, investment style can also have a direct impact on how fund returns are
produced in the first place. Since the pioneering analysis of Basu (1977) and Banz
(1981), portfolio managers have been well aware of the benefits of forming portfolios of
stocks that emphasize various firm-related attributes (such as price-earnings ratios and
market capitalization, respectively). The work of Fama and French (1992, 1993), who
espouse a multi-factor asset pricing model that supplements the standard market risk
premium with factors correlated to firm size and book-to-market ratios, has served to
deepen the interest in the role that these attributes play in explaining the cross-section of
equity returns.3 In fact, the pervasiveness of these findings has been such that it is now
commonplace to define investment portfolios along just two dimensions: (i) firm size and
(ii) value-growth characteristics, with the former defined by the market value of the
company’s outstanding equity and the latter often defined by the relative price-earnings
and price-book ratios of the fund’s holdings.4
There is ample evidence that a fund’s investment style has become deeply ingrained
in how the fund itself is identified and the returns it ultimately produces. Most notably,
Morningstar, Inc., a leading provider of independent mutual fund investment information,
routinely classifies funds into the cells of a 3 x 3 grid defined by firm size (small-, mid-,
and large-cap) and fundamental attributes (value, blend, and growth) for the purpose of
performance evaluation. Further, several recent studies have demonstrated that a
portfolio’s chosen investment style appears to materially affect the ex post wealth of the

2
Subsequent studies by Lehman and Modest (1987) and Brown and Brown (1987) confirm the result that
different benchmarks can produce substantial differences in the conclusions about fund performance.
3
Two recent studies have added an interesting twist to this debate. First, Loughran (1997) documents that
the book-to-market factor itself exhibits strong seasonal and size-based components. Second, Daniel and
Titman (1997) argue that abnormal returns produced by portfolios consisting of small capitalization and
high book-to-market stocks are due to those characteristics directly rather than their loadings in a Fama-
French-type factor model.
4
For instance, the S&P/BARRA growth and value indexes are formed by sorting the S&P 500 companies
by their price-book ratios while the Salomon growth and value indexes sort stocks on several additional
variables including dividend yields and price-earnings ratios; see Sorenson and Lazzara (1995).

2
investor. For example, Capaul, Rawley, and Sharpe (1993), Lakonishok, Shleifer, and
Vishny (1994), and Fama and French (1998) all show that portfolios of value stocks
outperform portfolios of growth stocks on a long-term, risk-adjusted basis and that this
“value premium” is a pervasive feature of global capital markets.
In this study, we consider an aspect of the mutual fund performance debate that has
received little attention in the literature. Specifically, using two measures of explanatory
power commonly employed in practice, we investigate the impact that the temporal
consistency of a manager’s investment style has on both absolute and relative fund
performance, as well as the persistence of that performance. The underlying premise of
this investigation is that a manager’s decision to maintain a portfolio that is highly
correlated with its designated investment mandate should be positively related to the
returns he or she produces. There are potentially three reasons for this hypothesized
relationship. First, it is likely that more style-consistent funds exhibit less portfolio
turnover and, hence, have lower transaction costs than funds that allow their style to drift.
Second, regardless of relative turnover, managers who commit to a more consistent
investment style are less likely to make asset allocation and security selection errors than
those who attempt to “time” their style decisions in the sense of Barberis and Shleifer
(2000). Third, it is also likely that managers with consistent styles are easier for those
outside the fund to evaluate accurately. Therefore, since better managers will want to be
evaluated more precisely, maintaining a style-consistent portfolio is one way that they
can signal their superior skill to potential investors.
Using a survivorship bias-free universe of mutual funds classified by Morningstar
over the period from January 1991 to December 2000, we show that those funds that are
the most consistent in their investment styles over time repeatedly produce better absolute
and relative performance than those funds demonstrating less style consistency. This
result proves to be generally robust to the fund’s investment style category, the time
period over which performance is measured, and the return-generating process used to
measure the fund’s expected returns. Further, the evidence presented is also strongly
supportive of the hypothesis that high style-consistent funds have lower portfolio
turnover than low style-consistent funds and that, controlling for turnover as well as fund
expenses, style consistency is still a dominant explanatory factor. Finally, we document

3
the positive relationship that exists between the consistency of a fund’s investment style
and the persistence of its return performance, even after accounting for momentum and
past abnormal performance effects. This finding provides an interesting counterpoint to
the work of Chan, Chen, and Lakonishok (1999) who show that style drift is more likely
to occur in funds with poor past performance. Taken as a whole, our results support the
conclusion that the ability of a manager to maintain a consistent investment style is a skill
valued in the marketplace.5
The remainder of the paper is organized as follows. In the next section, we present
a brief summary of the academic and practitioner literature on the measurement of mutual
fund performance and the role that investment style analysis has played in how funds are
classified and evaluating the persistence of fund performance. Section II reviews the
analytics for determining a mutual fund’s investment style and develops hypotheses
about the relationship between fund performance and style consistency. In the next two
sections, we discuss the data and empirical methodology used to test these hypotheses
and then present and summarize our results. Section V concludes the study.

I. Investment Style, Fund Classification, and the Performance Persistence in

Fund Returns: An Overview

A. Investment Style and the Classification of Mutual Funds

From the inception of the industry, mutual funds have attempted to inform potential
investors about their intended investment strategy by committing to a specific objective
classification. These investment objectives, which currently number 33 according to the
Investment Company Institute, are listed in the fund’s prospectus and include such
categories as aggressive growth, growth, growth and income, balanced, global, and
income. Prior to the advances that have been made in defining investment style during
the last several years, researchers and investors alike often used these objective classes as
surrogates for the risk-expected return tradeoff a given fund was likely to produce. In
fact, one of the earliest indications that investment style might play a significant role in

5
One other study that also makes intra-objective class comparisons of fund performance is Bogle (1998).
However, he does not consider the issue of style consistency, concentrating instead on the relationship
between fund returns and expenses ratios.

4
portfolio performance comes from McDonald (1974), who examines the returns
generated by a sample of mutual funds segmented by their stated objectives. In
particular, he finds that measures of both risk and return increased as the fund objective
became more aggressive and that the risk-adjusted performance of the more aggressive
funds dominated that of the more conservative funds during the sample period. More
recently, Malkiel (1995) offers evidence that a fund’s ability to outperform a benchmark
such as the S&P 500 was also related to its objective classification.6
Despite their documented connections with risk and performance, traditional fund
objective categories appear to have fallen out of favor as methods of classifying funds.
One reason for this is that the selection process for these objectives can be subjective and
might not always represent a fund’s actual holdings very well. More typical of current
fund classification methods is the effort to define a portfolio’s investment style directly
by a decomposition of its security characteristics. This is the approach taken in the work
of Fama and French cited earlier as well as that of Roll (1995), who examines the risk
premiums produced by portfolios sorted on factors such as market capitalization, price-
earnings, and price-book ratios. Not surprisingly, a consequence of such efforts has been
the finding that funds are often classified improperly using the traditional categories.
Brown and Goetzmann (1997) develop an entirely new classification system based on
style factors that is superior to the conventionally defined categories in predicting future
fund returns. Further, diBartolomeo and Witkowski (1997) use a multi-factor
decomposition of fund security holdings to conclude that 40 percent of the 748 equity
funds in their sample were misclassified, a problem they attribute primarily to the
ambiguity of the current objective classification system.7

6
Malkiel (1995) further indicates that the survivorship bias phenomenon introduced by Brown,
Goetzmann, Ibbotson, and Ross (1992) differs across his equity fund sample by objective class, with capital
appreciation and growth funds affected the most severely.
7
diBartolomeo and Witkowski (1997) also note that competitive pressure and the nature of compensation
contracting in the fund industry also lead to the potential for “gaming” the category listing. This is
consistent with the tournament hypothesis of Brown, Harlow, and Starks (1996), who show that managers
of different funds in the same objective class have different incentives to adjust portfolio risk depending on
relative performance.

5
B. Investment Style and Performance Persistence
Although analyzing overall performance has been the primary focus of the fund
performance literature, a related topic that has received considerable recent attention has
been the persistence of that performance—whether good or bad—over time. Against the
backdrop of Jensen’s (1968) original finding that managers generally are not able to
sustain superior performance, much of the more current research reports data supporting
persistence. Some of these studies, such as Hendricks, Patel, and Zeckhauser (1993) and
Brown and Goetzmann (1995), document a short-run, positive correlation between
abnormal returns in successive years. This phenomenon is attributed to managers with
“hot hands”, but the evidence in both studies appears to be driven by those funds
sustaining poor performance (i.e., “icy hands”).8 Additionally, Grinblatt and Titman
(1992) and Elton, Gruber, and Blake (1996) find that past risk-adjusted performance is
predictive of future performance over periods as long as three years, although Malkiel
(1995) stresses that these results are sample-period dependent. Finally, Carhart (1997)
and Wermers (2001) document that the dominance of past winner funds over past losers
is largely driven by momentum investing and is most pronounced in growth-oriented
portfolios.
Obviously, an important issue in establishing persistence is how abnormal
performance is measured and this is one point where a fund’s investment style comes into
play. In these studies, risk-adjusted performance is typically measured in terms of a
multi-factor return generating process designed to capture the essence of the fund’s style
in either an implicit or explicit fashion. Some use variations of the Fama-French
characteristic-based model while others, such as Grinblatt and Titman (1992), use a
multiple benchmark portfolio model. While nominally a study of the performance of
private asset managers rather than the public fund industry, Christopherson, Ferson, and
Glassman (1998) extend this literature in two interesting ways while corroborating the
finding that bad performance persists. First, they calculate abnormal performance
directly against returns to specific (i.e., Russell) style indexes. Second, the authors
exploit a statistical technique that allows them to assess performance conditioned on the

8
Brown and Goetzmann (1995) also show that those funds with persistently poor performance are the one
most likely to disappear from the industry, thus linking the persistence and survivorship literatures.

6
myriad macroeconomic information that was publicly available at the time the returns
were generated.
Teo and Woo (2001) provide evidence that investment style and performance
persistence may be connected. Based on their sample of style-adjusted returns (i.e., fund
returns in excess of the returns of the average fund in a given style group), they
demonstrate that portfolios of past winners and losers continue to mimic their previous
behavior. They also note that this persistence effect declines slowly as the length of the
initial period for measuring style-adjusted past returns increases. Although Teo and Woo
suggest that investors might profit from attempting to “time” style movements, it remains
unclear how the performance persistence phenomenon relates to the consistency with
which managers execute their respective investment mandates.

II. Investment Style Analysis and Style Consistency

A. Measuring Investment Style

As developed by Sharpe (1992), returns-based style analysis is an attempt to explain the
systematic exposures that the observed returns on a security portfolio have to the returns
on a series of benchmark portfolios designed to capture the essence of a particular
security characteristic. This process involves using the past returns to a manager’s
portfolio along with those to a series of indexes representing different investment styles
in an effort to determine the relationship between the fund and those specific styles.
Generally speaking, the more highly correlated a fund’s returns are with a given style
index, the greater the weighting that style is given in the statistical assessment.
Formally, returns-based style analysis can be viewed as a straightforward
application of an asset class factor model:
 K 
R jt = b j0 + ∑ b jk Fkt  + e jt (1)
 k =1 

where: Rjt is the t-th period return to the portfolio of manager j,

Fkt is the t-th period return to the k-th style factor,
bjk is the sensitivity of portfolio j to style factor k,
bj0 is the “zero-beta” component of fund j’s returns

7
 ejt is the portion of the period t return to fund j not explained by variability in
the set of style factors.

Using (1), the set of style factor sensitivities that define a given fund (i.e., {bjk}) are
established by standard constrained least squares methods, with at least two constraints
usually employed: (i) the estimated factor loadings sum to one, and (ii) all of the loadings
must be non-negative.
The coefficient of determination (i.e., R2) for (1) is defined as R2 = 1 -
[σ2(ej)/σ2(Rj)] and can be interpreted as the percentage of fund j’s return variability due
to the fund’s style decision. Of course, critical to this interpretation is the proper
specification of the benchmark portfolios representing the style factors, which should
ideally reflect the fund’s entire investment universe and be orthogonal to one another. In
practice, three general designations of the factor structure in (1) are typically used: (i) a
single-index market model (e.g., Jensen (1968)), (ii) multi-factor models based on pre-
formed style indexes (e.g., Sharpe (1992), Elton, Gruber, and Blake (1996)), and (iii)
multi-factor models based directly on portfolios created by characteristic-based stock
sorts (e.g., Fama and French (1993), Carhart (1997)).9

B. Defining Style Consistency

There are two ways that a manager’s investment style consistency can be defined and
measured in practice. First, from the specification of (1), it is clear that the statistic [1–
R2] captures the portion of fund j’s return variability that is not systematically related to
comovements in the returns to the style benchmarks. Accordingly, [1-R2] serves as a
proxy for the extent to which the manager is unable to produce returns consistent with a
tractable investment style. There are three plausible reasons why R2 measured from (1)
for any given fund might be less than one. First, assuming that the designated factor
model correctly summarizes the universe of securities from which the manager forms his
or her portfolio, [1-R2] might simply indicate that the fund has not diversified all
company-specific risk elements. Second, it is also possible that the manager is

9
BARRA, Inc., which produces a popular set of style factors, uses portfolios formed around 13 different
security characteristics, including variability in markets, success, size, trading activity, growth, earnings to
price ratio, book to price ratio, earnings variability, financial leverage, foreign income, labor intensity,
yield, and low capitalization. See Dorian and Arnott (1995) for a more complete description of these
factors are defined and used to make tactical investment decisions.

8
employing an investment style that the factor model is not capable of capturing; this is
the benchmark error problem discussed earlier. Finally, if (1) is estimated with the
additional constraint that bj0 = 0, as in Sharpe (1992) and Kahn and Rudd (1995), [1-R2]
can be interpreted as a measure of the manager’s security selection skill.
While each of the preceding explanations differ in its interpretation of [1-R2],
neither the first nor the third ultimately present a challenge for using R2 as a cross-
sectional measure of style consistency. That is, as long as the basic factor structure fairly
represents the style universe confronting the manager, the component of that fund’s
returns not explained by the model must be related to non-style elements.10 Conversely,
if the empirical form of (1) is an incomplete representation of the manager’s investment
style, then [1-R2] might artificially understate his or her ability to maintain a style-
consistent portfolio. With this caveat in mind, we use R2 as our first proxy for the
relative consistency of a fund’s observed investment style, subject to robustness checks
on the specification of the underlying factor model used to generate expected returns.11
A second way in which a fund’s style consistency can be measured involves the
calculation of the portfolio’s tracking error. Tracking error can be estimated as the
volatility of the difference between the fund’s returns and those to a corresponding
benchmark portfolio summarizing the style universe. To define this more precisely, let:
N
∆ jt = ∑ x ji R jit - R bt = R jt - R bt (2)
i =1

where xji is the weight in managed fund j for security i and Rbt is the period t return to the
style benchmark portfolio. Notice two things about the return differential defined in (2).
First, given the returns to the N assets in the managed portfolio and the benchmark, ∆ is a
function of the investment weights that the manager selects (i.e., ∆ = f({xi}│{Ri}, Rb)).

10
Although this interpretation is ultimately valid whether or not bj0 is included in (1), the cleanest
specification of the model constrains the intercept to be zero because this forces all non-style return
components (i.e., noise and security selection skills) into the error term.
11
Chan, Chen, and Lakonishok (1999) present a style classification scheme that can be seen as a variation
on this approach. Specifically, they rank funds by their exposure to a characteristic (e.g., firm size) or
factor loading and then scale them to fall between zero and one. Using this approach, they show that the
correlation of a fund’s past and future style averages between 70 and 80 percent, indicating a broad degree
of style consistency in their sample.

9
Second, ∆ can be interpreted as the return to a hedge portfolio long in the managed fund
and short in the benchmark (i.e., xb = -1).12
From (2), periodic tracking error can be measured by the standard deviation of ∆
(σ∆) so that annualized tracking error (TE) can be calculated:

TE = σ ∆ P (3)

where P is the number of return periods in a year. TE represents a second measure of the
extent to which a manager is able to deliver an investment style consistent with that
implied by a style benchmark. It differs fundamentally from the R2 statistic generated
from (1) in that it does not involve the specification of explicit functional form for the
style-based return-generating model. However, (3) does require the selection of a
benchmark portfolio whose returns adequately capture the relevant style characteristics of
the security universe from which the manager chooses {xji}. Naturally, this selection
may be fraught with the same sort of peril as the designation of the style factor structure
in (1). Thus, the earlier robustness caveat regarding the use of R2 as a cross-sectional
measure of style consistency applies to TE as well.13
Figure 1 illustrates the way that changes in investment style over time can be
measured. At any given point, any fund can have its position plotted in a 3 x 3 style grid
by using available return data to estimate the optimal combinations of the mimicking
style indexes in a factor model such as (1). As more performance data become available,
additional plot points can be calculated and overlaid in the same grid to indicate how the
fund’s style either drifts or remains relatively constant. Figure 1 shows the connected
plot points (or “snail trails”) for two representative large-cap value funds, with circles of
increasing size highlighting the most recent plot points. For comparison, the average
positions of several different style and market indexes are shown as well.
[Insert Figure 1 About Here]

12
For more discussion of this development, see Grinold and Kahn (1995) who also refer to tracking error
as the fund’s active risk relative to the benchmark.
13
Ammann and Zimmerman (2001) note that while (3) is used frequently in practice, tracking error can
also be estimated as the standard deviation of the residuals of a linear regression between the returns to the
managed and benchmark portfolios. However, as this approach essentially relies on a single-factor version
of (1), it will be considered as a special case of the R2-based style consistency measure.

10
 The fund in the left-hand panel of the display (Fund A) has an R2 value of 0.92
while the Fund B in the right-hand panel has an R2 value of 0.78 with respect to the same
factor model.14 Clearly, Manager A has maintained the portfolio’s investment style
position to a greater degree than Manager B, who exhibits substantially more style drift.
Accordingly, we will define Fund A as being more style consistent than Fund B.
Whether such differences in the decision to stay consistent to a given investment style are
associated with measurable differences in fund return performance is the purpose of the
empirical work that follows.

C. Testable Hypotheses
There are three specific hypotheses that we will test in the subsequent sections. First, the
style position patterns illustrated in Figure 1 suggest that Manager B is more likely than
Manager A to attempt to add value through superior stock selection skills or tactical style
adjustments. In either case, it is quite possible that Fund B requires a higher degree of
portfolio turnover (measured in a given period as the dollar level of fund sales divided by
the average market value of the fund’s total assets) than Fund A. Note, however, that
style consistency does not imply a buy-and-hold portfolio; indeed, matching the
movements in oft-volatile benchmark returns in order to maintain constant style factor
loadings may require frequent rebalancing. Nevertheless, these adjustments may be
fewer in number than the trading required to execute a more active portfolio strategy.

Hypothesis One: Style-consistent (i.e., high R2, low TE) funds have lower portfolio
turnover than style-inconsistent (i.e., low R2, high TE) funds.

Related to the last supposition, it is possible that more frequent trading leads in turn
to inferior return performance. There are two reasons why this could be true. First,
several studies establish a significant negative correlation between fund expense ratios
and returns (e.g., Carhart (1997), Bogle (1998)). More active management, with its
attendant higher portfolio turnover, could increase fund expenses to the point of
diminishing relative performance. Second, regardless of whether style-inconsistent funds
have higher portfolio turnover, it may also be the case the managers of these portfolios

14
The model specifications and return analysis that produced these examples will be detailed in the next
section.

11
are chronically underinvested in the “hot” sectors of the market through their more
frequent tactical portfolio adjustments.15 There is, in fact, a long-standing literature
suggesting that professional asset managers generally possess negative market and style
timing skills; see, for example, Kon (1983), Chang and Lewellen (1984), and Coggin,
Fabozzi, and Rahman (1993), and Daniel, Grinblatt, Titman, and Wermers (1997).16
Thus, if the value lost through poor timing decisions is sufficient to offset the marginal
addition of the manager’s selection skills, we would expect managers demonstrating less
style consistency to perform relatively worse than their more disciplined peers.

Hypothesis Two: Style-consistent funds have higher total and relative returns than style-
inconsistent funds.

The final hypothesis we test involves the relationship between style consistency and
the persistence of fund performance. From the literature on performance persistence
reviewed earlier, a finding that appears with some regularity is that it is usually bad
performance that persists from one period to the next (e.g., Brown and Goetzmann
(1995), Christopherson, Ferson, and Glassman (1998)), especially when fund returns are
adjusted for a momentum effect (e.g., Carhart (1997), Wermers (2001)). In the present
context, while style-consistent funds—which, by definition, produce returns that are
closely correlated with a benchmark or specific style exposure—may or may not produce
superior performance, it is unlikely either that they will regularly produce inferior relative
returns. On the other hand, managers of portfolios that rely more on security selection or
market/sector timing than style discipline to justify their active management fees will
generate less reliable performance relative to the benchmark. If these return deviations
tend to be more negative than positive—as might occur if they require a larger number of
portfolio transactions—then style-inconsistent funds may be responsible for the adverse

15
Barberis and Shleifer (2000) have modeled an economy where some investors shift assets between style
portfolios in an attempt to exploit perceived contrarian and momentum opportunities. The authors
demonstrate that prices in such a market can deviate from long-term fundamental values so as to look like
bubbles. However, without knowledge of which style is currently in favor, they argue that arbitrage is not a
riskless proposition and that there are no consistent profits available.
16
More recent evidence in Bollen and Busse (2001) suggests that mutual fund managers may exhibit
significant positive timing skills when measured using daily returns.

12
performance persistence phenomenon.17 Conversely, better managers might decide to
maintain a more style-consistent portfolio as a means of conveying their investment
prowess to the market.

Hypothesis Three: There is a positive correlation between the consistency of a fund’s

investment style and the persistence of its future performance.

III. Data, Methodology, and Preliminary Analysis

A. Sample Construction and Descriptive Statistics

The data for this study consist of monthly returns to a collection of equity mutual funds
over the thirteen-year period spanning January 1988 to December 2000. The source of
these returns is a Morningstar mutual fund database. Investment category classifications
for each fund as well as portfolio turnover and expense ratio statistics were obtained from
the Morningstar database and the Center for Research in Security Prices (CRSP) mutual
fund database. Following industry conventions, Morningstar classifies funds along two
dimensions: average firm size, based on median market capitalization, and “value-
growth” characteristics, based on an asset-weighted composite ranking of the relative
price-earnings and price-book ratios of the stocks in the portfolio. Separating each
dimension into three parts places each fund in the sample universe into one of nine style
categories: large-cap value (LV), large-cap blend (LB), large-cap growth (LG), mid-cap
value (MV), mid-cap blend (MB), mid-cap growth (MG), small-cap value (SV), small-
cap blend (SB), and small-cap growth (SG).18 This database is also constructed so as to
be free of the sort of survivorship bias problems documented by Brown, et al (1992).
Finally, notice that by using these style categories we create a sample that includes index
funds, but excludes specialty funds such as sector, balanced, and asset allocation funds.
17
In fact, Gallo and Lockwood (1999) have shown that about two-thirds of funds that changed poor-
performing managers subsequently changed their investment styles, as determined by a shift in the primary
factor loading in an equation similar to (1) following the installation of the new manager.
18
Morningstar began using this style classification system in 1992. For the purpose of classifying the
investment style of funds in the first year of our forecast period (i.e., 1991), we use Morningstar’s initial
assessments made in 1992. To test whether this decision, which was made to create a ten-year forecast
period that could be split into two five-year subperiods, affected the analysis, we also replicated the study
using data from just the 1992-2000 time frame. Additionally, we reproduced the study using alternative
style classification and objective groups (e.g., Lipper Analytical). All of these modifications generated
highly similar findings and are therefore not reported here.

13
 Table I summarizes the number of funds in each style category for every year of the
sample period, the total funds in the sample listed annually, as well as the average
number of funds that existed in each category over two non-overlapping subperiods. The
numbers reported represent those funds with at least 36 months of return history prior to a
given classification year. Thus, with this inclusion criterion, the earliest style category
year possible is 1991, with all funds reported for this period having returns dating to
January 1988. The final column of the display documents the dramatic increase in the
total number of funds eligible for style classification and hence included in the study.
Starting with a collection of 698 separate portfolios in 1991, the sample grew at a year-
over-year rate of more than 18 percent to its terminal level of 3,177 in 2000.
[Insert Table I About Here]
This display also indicates that the distribution of funds across the various style
classes is not uniform, nor has the growth of each category over time been comparable.
In particular, consistent throughout the entire sample period, the biggest collection of
funds fall into the three large-cap categories, with the large-cap blend classification
(which includes, among others, funds based on the Standard & Poor’s 500 benchmark)
being the most popular in every individual year. At the other extreme, small-cap funds
were the least well represented for the majority of the sample period, although the gap
between small- and mid-cap funds narrowed over time; in fact, the SB category surpassed
the MB class in the later years of the sample. Further, the small-cap categories were the
fastest growing over the classification period, followed by the large-cap and mid-cap
style classes.
[Insert Table II About Here]
Table II provides several initial indications of the myriad practical differences that
exist between the Morningstar style categories. Panel A lists descriptive statistics over
various periods for several category-wide average characteristics, including annual total
return (i.e., capital gain plus income distribution, net of expenses), standard deviation,
firm size, expense ratio, and portfolio turnover (i.e., the ratio of fund sales to total fund
holdings, measured in dollar volumes). Panel B then displays differences in those
characteristics across “extreme” categories (e.g., [LV-LG] for the value-growth

14
dimension, [LV-SV] for the size dimension), along with the associated p-values
summarizing the statistical significance of those differences.
The results in Table II confirm much of the conventional wisdom about investment
style and fund performance. For instance, Panel B shows that, controlling for market
capitalization over the entire sample period, value-oriented funds produced average
annual returns as much as 5.24 percent higher than those for growth-oriented portfolios.
Further, the average large- and small-cap value fund standard deviation are more than
seven percentage points lower than the total risk level of comparably sized growth funds.
These results are consistent with the existence of a risk-adjusted value premium reported
by Capual, Rawley, and Sharpe (1993) and Fama and French (1998). Alternatively,
controlling for value-growth characteristics, small-cap funds outperformed large-cap
funds by an average of between 6.10 and 9.98 percent, but with total risk that was
commensurately higher, which is consistent with the findings of first published by Banz
(1981).
This display also reveals important differences about the manner in which portfolios
in different style categories are managed. Specifically, over the entire sample period,
there were substantial differences between style groups in portfolio turnover and expense
ratios. Generally, the data show that growth funds have higher turnover ratios than value
funds (e.g., MG turnover exceeds MV turnover by 48.23 percent) and large-cap funds
have lower turnover ratios than small cap funds (e.g., LG turnover is 26.96 percent lower
than SG turnover). The only deviation from these conclusions is that the [LV – SV]
turnover ratio is positive, although not always significantly so. Consistent with this
pattern of higher trading, the results in Panel B also support the conclusion that small-cap
and growth funds have higher expense ratios than large-cap and value funds, respectively.
Finally, while these findings are relatively robust over time, it does appear that most all
investment styles had higher turnover and higher expense ratios in the latter half of the
sample period.
An important implication of the preceding results is that it may be quite difficult to
directly compare the return performance of two funds that have contrasting investment
styles. Said differently, fund investment prowess is more appropriately viewed on a
relative basis within—rather than across—style categories; this is the tournament

15
approach adopted by Brown, Harlow, and Starks (1996) and Chevalier and Ellison
(1997), where a manager’s performance and compensation are determined in comparison
with their peers within a style class or a style-specific benchmark. Further, Khorana
(1996) shows that managers exhibiting higher portfolio turnover and higher expense
ratios relative to their style-matched peers are more likely to be replaced. Of course,
these industry practices are likely driven by the tendency for investors to concentrate on a
fund’s past total returns when making their investment decisions within a given style
class (e.g., Sirri and Tufano (1998), Capon, Fitzsimons, and Prince (1996)).
Consequently, in the subsequent analysis, we will consider the issue of investment style
consistency in the context of the nine style “tournaments” defined by the Morningstar
categories.

B. Style Consistency Behavior

As noted earlier, the consistency of a fund’s investment style can be measured either
with the coefficient of determination relative to a return-generating model or by tracking
error compared to a style-specific benchmark portfolio. To calculate the former (i.e., R2),
we adopt as an empirical specification of equation (1) Carhart’s (1997) extension of the
Fama-French three-factor model that includes Jegadeesh and Titman’s (1993) return
momentum factor:
Rjt = aj + bjMRMt + bjSMBRSMBt + bjHMLRHMLt + bjPR1YRRPR1YRt + ejt (4)

Equation (4) employs the following factor definitions: (i) RMt is the month t excess return
on the CRSP value-weighted portfolio of all NYSE, AMEX, and NASDAQ stocks; (ii)
RSMBt is the difference in month t returns between small cap and large cap portfolios; (iii)
RHMLt is the difference in month t returns between portfolios of stocks with high and low
book-to-market ratios; and (v) RPR1YRt is the difference in month t returns between
portfolios of stocks with high and low stock return performance over the preceding year.
Return data for the first three factors were obtained from Eugene Fama and Ken French
while the momentum factor was constructed using Carhart’s procedure with return data
from constituents of the Russell 3000 index. Finally, individual fund returns and returns
to the market risk factor are computed in excess of the corresponding one-month U.S.

16
Treasury bill yield, which allows for usual interpretation of aj (i.e., alpha) as an abnormal
performance measure for fund j.19
In order to estimate the consistency of a fund’s investment style using the tracking
error measure in (2), it is necessary to designate style category-specific indexes to
represent the benchmark portfolio in each of the nine style classes. One challenge in this
effort is to select a set of indexes that is uniform in its construction and meaning. For that
reason, we adopted the following benchmarks for each of the cells in the 3 x 3 style grid:
Russell 1000-Value (LV), Russell 1000-Blend (LB), Russell 1000-Growth (LG), Russell
Mid-Cap-Value (MV), Russell Mid-Cap-Blend (MB), Russell Mid-Cap-Growth (MG),
Russell 2000-Value (SV), Russell 2000-Blend (SB), and Russell 2000-Growth (SG). The
return data for these indexes was obtained directly from Frank Russell Company.
We calculate both R2 and TE values on an annual basis for all nine style classes,
using returns for the prior three years (e.g., consistency measures for 1999 are calculated
using returns from 1996-98). Funds are then rank ordered in separate listings by both
statistics and sorted into “high consistency” (i.e., high R2 or low TE) and “low
consistency” (i.e., low R2 or high TE) subsamples according to where their consistency
measure falls relative to the median for the objective class. Separate consistency
subgroups are maintained for the R2 and TE sorts and we then reclassify these fund
consistency portfolios on a year-to-year basis.
Panel A of Table III summarizes the characteristics of the fund sample split into
high and low consistency groupings by R2, while Panel B separates the funds by the TE
criterion. Each panel lists median values for the following statistics: R2, annual TE, peer
group ranking (i.e., the fund’s relative position in the annual performance tournament,
based on total return), annual total return, return standard deviation, portfolio turnover,
and expense ratio. In both panels, the numbers reported represent aggregated values of
these statistics; the funds were sorted into consistency groups on an annual basis to
produce the base levels of the various statistics and then these annual values were then
averaged to produce the display.

19
We estimated two other versions of (4) as well, including the basic three-factor version of the Fama-
French model and Elton, Gruber, and Blake’s (1996) variation of that model that includes as risk factors
excess returns to a bond index and a global stock index. The R2 rankings produced by these alternative
specifications were quite similar and are not reproduced in full here. They are, however, available upon
request.

17
 [Insert Table III About Here]
Several relevant observations can be made about the results listed in Table III.
First, regardless of whether funds are sorted by R2 or TE, it appears that large-cap funds
demonstrate more investment style consistency than do small- or mid-cap funds. For
instance, the median R2 value for the high consistency portion of the three large-cap style
categories is 0.93 while the median TE for this grouping is 3.70%. By contrast, the high-
consistency portions of the small- and mid-cap objectives yield a median R2 value of 0.87
and a “typical” TE of around 5%. Comparable results obtain for the low-consistency
groupings: median large-cap R2 and TE values are 0.86 and 5.27%, respectively, with the
analogous values for the other two size-based categories were in the range of 0.77 and
8.30%. Although not shown, the findings from each of the five-year subperiods of the
sample confirm these patterns.
Table III also provides indirect evidence supporting the first two hypotheses listed
in the previous section. Specifically, the first hypothesis maintained that high-
consistency funds would have lower portfolio turnover than low-consistency funds.
Based on a simple comparison of median turnover ratios, this is true for all nine style
groups in Panel A and eight of the nine (MV being the exception) in Panel B. Further, it
is also the case that high-consistency funds have lower average expense ratios; all of the
18 style categories across the two panels support this conjecture. As to the second
hypothesis, which held that high-consistency funds should produce higher total and
relative returns than low-consistency funds, the median annual fund return is larger for
the former grouping in seven of nine cases using both the R2 and TE ranking criteria (MV
and MB excepted). Additionally, the managers of more style-consistent portfolios
produced a higher median peer group ranking eight out of nine times in both Panels A
and B, with MG being the non-conforming category in each case. Thus, while more
formal tests of these propositions will be developed in the next section, this initial
evidence corroborates the view that investment style consistency matters.
Given the similarity of the findings for the consistency measures just described, it is
reasonable to ask whether the R2 and TE statistics generate unique rank orderings of
funds in a given style class. For instance, for every style category it is true that when
consistency is defined by R2, the median TE values for the resulting low- and high-

18
consistency groupings are supportive (and vice versa). Nevertheless, while the rankings
produced by the model-based and benchmark-based consistency measures are indeed
comparable, they are not identical. The Pearson correlation coefficient between the fund-
specific level of R2 and TE is –0.582, which is significant at the 0.01% level. (Recall that
high consistency is defined by high R2 values, but low TE values; thus, a negative
correlation level between these variables would be expected.) The Spearman correlation
coefficient of the rankings produced by these measures is –0.629, which is also highly
statistically significant. Thus, we conclude that R2 and TE provide alternative methods
for calculating the temporal consistency of a mutual fund’s investment style.

IV. Extended Empirical Results

A. Basic Correlation Tests

A more direct test of the first two consistency hypotheses is possible by considering how
the pattern of correlation between the style consistency measures and certain fund
management and performance variables evolved over the sample period. Specifically,
the proposition that consistency and turnover are negatively related can be judged by the
cross-sectional correlation between a fund’s R2 or TE measure and its portfolio turnover
ratio. Similarly, the correlation between R2 (or TE) and future fund returns provides
direct evidence on the proposition that consistency and subsequent performance are
positively related.
[Insert Table IV About Here]
Table IV reports these Pearson correlation statistics for the 1991-2000 sample
period as a whole as well as for each year individually. Panel A of the display defines
consistency with respect to the coefficient of determination while Panel B focuses on
tracking error. In both cases, the consistency measures are correlated with the following
five variables: annual portfolio turnover, annual fund expense ratio, actual annual fund
return, “tournament” fund return (i.e., actual returns standardized by year within a fund’s
style classification), and peer ranking of the tournament return. As before, the
2
consistency statistics are measured out-of-sample; that is, R and TE are based on fund
returns for the 36-month period preceding the year for which the management and
performance variables are produced.

19
 Hypothesis One is tested with the correlation between a particular consistency
measure and fund turnover. By the way that consistency is defined, this correlation is
predicted to be negative for R2 (i.e., high R2, low turnover) and positive for TE (i.e., low
TE, low turnover). The results from both panels of the display unambiguously support
the notion that more style consistent funds have lower portfolio turnover. In fact, there is
not a single year in which either consistency measure provides contrary evidence and the
sample period-wide absolute correlation value is just under 0.25. Further, although not
formally part of the first hypothesis, Table IV also indicates that funds with stricter
adherence to their investment style also tend to have lower expense ratios. This suggests
the possibility that managers who charge higher fees (i.e., have higher expense ratios) are
more likely to be active investors who seek to obscure their performance by letting their
investment style drift. Taken together, these findings also imply an interesting extension
of Khorana’s (1996) conclusion reported earlier: Managers who remain more consistent
to their designated style mandate may be able to reduce the probability that they will be
replaced.
To test the second hypothesis fully, it is necessary to define both absolute and
relative fund returns. As noted, although investors often focus on actual returns when
selecting funds (e.g., Capon, Fitzsimons, and Prince (1996)), it is also true that fund
complexes and managers act as if they compete in more narrowly defined style-specific
tournaments (e.g., Brown, Harlow, and Starks (1996)). Accordingly, in addition to
calculating a fund’s total return during a particular sample year, we also convert this
value to a z-score by standardizing within the fund’s Morningstar investment
classification. We refer to this standardized value as the fund’s “tournament” return and
it is one of two relative return measures we employ, the other being peer ranking (i.e.,
tournament ranking) based on these standardized returns. This adjustment also allows for
the aggregation of performance statistics across time and investment styles, which
facilitates the analysis in the next section.
The evidence presented in Panel A of Table IV strongly supports the proposition
that more style consistent funds produce higher absolute and relative returns. Under this
hypothesis, the correlation coefficient between R2 and each of the return metrics is
expected to be positive. This is indeed the case for the entire sample period as well as

20
during eight of the ten individual sample years. Overall, the correlation between R2 and
the relative return measures is stronger—11.0% for tournament returns and 9.2% for
tournament ranking—than with unadjusted total returns (i.e., 2.9%), although the latter is
still statistically significant. Further, the correlations are particularly strong during the
middle years of the sample (i.e., 1994-1998) for all of the return statistics.
The findings in Panel B for the TE consistency measure tell a similar, if more
modest, story. The expected correlation coefficient for this statistic should be negative
and, for the entire sample period, the findings support this conclusion. However, the
absolute coefficient values for the relative returns—9.1% for tournament returns and
7.8% for tournament rankings—are lower than those for the R2 consistency measure and
the correlation between TE and actual returns is not statistically significant. Further, for
the relative return measures, six of the ten annual tournaments produce coefficients that
conform to second consistency hypothesis, with the strongest values once again being
generated in the middle years of the sample. Interestingly, despite not producing a
significant decade-wide relationship, the correlation coefficient between TE and actual
returns is in the predicted direction in seven of the ten individual years.
In addition, to confirming the first two hypotheses concerning the value of
maintaining a consistent investment style, the findings in Table IV suggest two notable
implications. First, regardless of how consistency is measured or when it is assessed, the
relationship between style consistency and portfolio turnover is remarkably strong. So
strong, in fact, that it may be the case that style consistency is merely a surrogate for low
turnover and, hence, low transaction costs. We investigate this possibility in the
following sections. Second, while suggested previously, it is now more apparent that R2
and TE produce measurably different indications of style consistency and that the model-
based metric is a more reliable indicator of subsequent return performance. One possible
explanation for this is that while TE measures consistency relative to a single benchmark,
depending on the model R2 can tie the consistency measure to a more expansive
definition of the investment mandate. This appears to be a useful expansion when
judging performance on a total, rather than a relative, basis.

21
B. Style Consistency and Return Persistence: Initial Tests
The final hypothesis specified earlier holds that the consistency of a fund’s investment
style should be positively related to the manager’s ability to produce consistently superior
relative returns. To test this notion, we first need to define a fund-specific measure of
past successful (or unsuccessful) investment performance. Given our out-of-sample
methodological design, the intercept term from the excess return-generating model in
(4)—i.e., alpha—serves this purpose. In fact, to facilitate comparisons, for each fund in a
given year we compute an alpha statistic using two forms of (4): the conventional Fama-
French three-factor model and an extension of this model that includes a return
momentum factor. As we discuss below, the reason for measuring past performance with
two different models lies in Carhart’s (1997) finding that any performance persistence
present with the three-factor model largely disappears when return momentum is added
as a fourth explanatory factor.
We test for performance persistence in the following manner. Using a 36-month
return window at a given point in time, we estimate (4) for each fund in the sample. This
estimation yields estimates for both alpha and R2, which becomes our main measure of
style consistency.20 We then calculate the fund’s tournament (i.e., standardized) return
during the t-month period immediately following the end of the model estimation
window. Two values of t are employed: three (i.e., the fund’s next quarter return) and 12
(i.e., the fund’s next year return). Repeating this process for each fund throughout
sample period by rolling the 36-month estimation window forward as necessary produces
a full set of data for three-year past performance (and consistency) as well as t-month
subsequent performance.
To examine the dynamics of the various relationships between future performance,
past performance, and investment style consistency, we regress the three- or 12-month
standardized return on the prior levels of fund alpha and R2. In various forms of this
regression, we also include the following control variables: portfolio turnover (TURN),
fund expense ratio (EXPR), and fund size (TNA), as measured by the market value of its

20
Given the analysis in Table IV, the regression results produced below will be reported for just the
model-based consistency measure. We have replicated these findings using TE as well, which generates a
largely comparable set of conclusions to those we offer for R2. These supplementary results are available
upon request.

22
assets under management at the end of the 36-month estimation period. In order to
aggregate these data across different annual style tournaments into a single calculation,
all of the variables just described were standardized by year and style group. This
normalization process also allows for the direct comparability of the magnitude and
significance of the various parameter estimates.
[Insert Table V About Here]
Table V reports the results for several different versions of the performance
regression over the entire 1991-2000 sample period. The findings in Panel A use three-
month future returns as a dependent variable while Panel B lists results for one-year out-
of-sample returns. In each panel of the display, a duplicate set of coefficient estimates
are reported for past performance (ALPHA) and style consistency (RSQ) measured by
the Fama-French three-factor version of (4) (i.e., FF) and Carhart’s momentum-
supplemented extension (i.e., FFC). For both return-generating models, we estimated
parameters for six different combinations of the independent variables, starting with
simple models involving ALPHA or RSQ alone and ending with one that includes all five
regressors.
The findings in Table V support several general conclusions. Most broadly, the
overall level of future return predictability is low, as indicated by the adjusted coefficient
of determination values reported in the last row of each panel. Within this context,
longer-term (i.e., twelve month) out-of-sample performance appears to be marginally
more predictable than short-term future returns. Despite these small regression-wide
statistics, however, the individual parameters on the independent variables are all highly
significant at conventional levels. This is clearly a by-product of the large sample sizes
created by the pooling of data across time and investment style groups.21 Nevertheless,
the reported parameters are useful for the information they contain about the direction
and magnitude of the various relationships, as well as the comparative connections they
suggest.
Model 1, which regresses future returns on past fund performance alone, provides a
baseline analysis of the persistence phenomenon. The positive coefficient values in all

21
In the next section, we examine these relationships within the context of each of the nine investment
style groups.

23
four versions of this model indicate that relative performance did indeed persist
throughout the sample period. More interestingly, however, is that it is also the case that
the inclusion of a momentum factor to the return-generating process (i.e., moving from
FF to FFC) substantially reduces the magnitude of the relationship between ALPHA and
future returns. For instance, looking at the three-month subsequent returns in Panel A,
the coefficient value declines by more than two-thirds, from 0.075 to 0.021; the
comparable decline for twelve-month future return sample is from 0.091 to 0.056. These
reductions confirm, at least partly, the contention that persistence is actually explained by
return momentum.
The remaining five models represented in Table V examine the role that investment
style consistency plays in predicting future fund performance. In Model 2, the simplest
form of the relationship between RSQ and subsequent returns is tested. All four versions
produce positive coefficient values: 0.050 (FF) and 0.049 (FFC) for three-month returns
versus 0.109 (FF) and 0.110 (FFC) for twelve-month returns. The direction of this
relationship is in line with that implied by Hypothesis Three. Additionally, notice that
unlike ALPHA, the inclusion of the momentum factor in the expectations model has no
effect on the strength of the relationship between RSQ and future fund returns.
Models 3-6 explore this relationship further by controlling for other mitigating
influences. Most importantly, the four forms of Model 3 show that the consistency
variable is not a simple surrogate for ALPHA. In fact, the coefficient level for RSQ does
not change appreciably with the addition of the past performance metric. The results for
Model 4, which includes TURN in addition to ALPHA and RSQ, allows this conclusion
to be extended with respect to portfolio turnover; that is, adding TURN also does nothing
to diminish the magnitude of the style consistency variable.22 Thus, it also appears that
RSQ is not a proxy for TURN either. Finally, the connection between RSQ and future
performance remains strong, although at a somewhat reduced level, after adding fund
expense ratios and total net asset values (i.e., Models 5 and 6) as regressors. Viewed
collectively, the findings in Table V provide strong, broad support for the proposition that

22
An interesting related finding documented in Table V is the positive coefficient defining the
relationship between future fund returns and portfolio turnover. Wermers (2000) documents this same
connection and interprets it as supporting the value of active fund management.

24
the consistency of a fund’s investment style and its future performance are positively
related.
All of the results presented thus far have been based on our full sample of mutual
funds that includes index funds. This permits the possibility that the effects we have
documented are actually being driven by a large passive investment element where the
“consistency” of the style is mandated rather than voluntary. One fact that makes this
unlikely, however, is that indexed portfolios represent a relatively small percentage of the
collection of funds included in the study; for instance, in 2000 there were only 140 index
funds a total sample of 3,177 (i.e., 4.4 percent). Nevertheless, to test more formally the
possibility that style consistency is driven by a passive investment mandate, we replicated
the findings in Table V for the sample excluding index funds. Although not reproduced
in full here, the estimated regression parameters are virtually identical whether or not
index funds are included in the sample. Typical of this outcome are the results for FFC-
Model 6 using 12-month future returns as the dependent variable. The coefficients
calculated with (without) index funds are: Intercept: 0.000 (0.000); ALPHA: 0.038
(0.038); RSQ: 0.077 (0.076); TURN: 0.062 (0.062); EXPR: -0.145 (-0.141); and TNA:
-0.019 (-0.018).23 Thus, we conclude that the style consistency phenomenon is not
related to active versus passive management issues.

C. Style Consistency and Return Persistence: Style Tournaments

In this section, we extend the preceding analysis by estimating the parameters of the
regression of future fund returns on ALPHA, RSQ, and the various control variables
within each of the nine Morningstar investment style groups. Specifically, we used the
four-factor (i.e., FFC) version of (4) to generate the two main regressors and computed
future returns over the 12 months following the estimation interval. After standardizing
the variables by year only, we then calculated the coefficients of Model 6 for each style
group over the entire 1991-2000 sample period. Model 6, which includes all five
independent variables, was chosen as it represents the most severe test for the style
consistency hypotheses. These findings are reported across the nine rows of Panel A in

23
The comparable set of estimated parameters for FFC-Model 6 using three-month future returns is:
Intercept: 0.000 (0.000); ALPHA: 0.011 (0.012); RSQ: 0.030 (0.030); TURN: 0.033 (0.034); EXPR: -0.082
(-0.080); and TNA: -0.008 (-0.007).

25
Table VI.24 Panel B then reports regression results for funds aggregated across the three
divisions of each style dimension (i.e., large-, mid-, and small-cap for the firm size
dimension; value, blend, and growth for the firm characteristic dimension).
[Insert Table VI About Here]
The first thing to notice is that, consistent with the demographic data reported in
Table I, there are far more observations for large-cap funds than for mid- or small-cap
funds. In the Growth category, for instance, the numbers in the first column of the first
panel in Table VI indicate that large-, mid-, and small-cap portfolios accounted for 2003,
1,230, and 982 returns, respectively. Thus, it is reasonable to conclude that the pooled
results of the last section were weighted more heavily toward large-cap funds than the
other two size-based categories. This becomes an important consideration because the
data in Table VI suggest that the persistence and consistency effects described above are
not completely uniform across the various style groups.
In particular, the parameter on RSQ is statistically significant at conventional levels
in the direction predicted by Hypothesis Three for six of the nine style classes, with LV,
MB, and MG being the exceptions. For two of these exceptions (LV and MB), the
coefficient on the style consistency variable is positive. The RSQ parameter estimates for
all three small-cap styles are particularly strong. By contrast, the estimated parameter
values for ALPHA are statistically insignificant in five of nine cases and, in one of the
four groups where the parameter is significant (i.e., SV), it has the wrong sign to support
the notion that past return performance persists in subsequent periods. Additionally, of
the control variables, only fund expense ratios proves to have a consistently reliable
effect on the generation of future returns; in fact, the coefficient on EXPR is significantly
negative in eight of the nine style group regressions.
The regression results for funds aggregated within style dimensions in Panel B
underscore these tournament-specific findings. In particular, TURN and EXPR are
statistically significant for all six broad style groups. Importantly, though, the
significance of these control factors does not diminish the influence that style consistency
has on future fund returns; the estimated parameter on RSQ is positive and statistically

24
Although not presented in the display, for every style tournament we also calculated the parameter
estimates for each of other five forms of the regression equation employed earlier. These data support the
conclusions discussed below and are available from the authors upon request.

26
significant in five of the six cases, with the mid-cap group producing an insignificantly
positive coefficient. Conversely, the relationship between ALPHA and future returns is
only significant for three of the six dimension divisions.
The primary conclusion that can be drawn from the findings in Table VI is that a
manager’s commitment to running a style-consistent portfolio will tend to signal his or
her chances to produce superior future returns. As noted, this style consistency effect
remains in place even after accounting for other mitigating influences documented
elsewhere in the literature, such as past performance, portfolio turnover, and fund
expenses. It is now also apparent, however, that this relationship is more likely to hold
for certain investment styles than others. The connection is particularly strong for small-
cap funds and far less meaningful for mid-cap funds. Further, the aggregated evidence
from value, blend, and growth groups strongly supports Hypothesis Three, even though
one of the three size-based cells in each of these classes is not statistically significant.
Thus, while not pervasive, style consistency does appear to play a tractable role in future
fund performance.

D. Style Consistency and Return Persistence: Logit Analysis

The results of the preceding two subsections document the effect that variables such as
past performance, style consistency, portfolio turnover, and expense ratios have on the
level of future fund returns. While providing a clear picture of how these factors are
connected, there is some evidence (e.g., Brown, Harlow, and Starks (1996)) to suggest
that compensation contracting among fund managers may depend on an even more basic
level of fund performance: Are managers above or below average compared to their peer
groups? Consequently, a related question worth exploring is whether these same factors
influence where a manager ranks relative to the median competitor within a particular
style tournament. We examine this issue in two ways. First, to provide a comparison
with the continuous dependent variable results just presented, we re-estimate the
regression equations in Table V using a logit model with a dependent variable that takes a
value of one if a fund’s annual return exceeds the median for a particular style group in a
given year and zero otherwise. Second, we use these logit regressions to assess the
probability of finishing as an above-median manager in a two-way classification

27
involving the relative levels of a fund’s alpha and style consistency statistics. In this
way, we can attempt to quantify the economic significance of the connection between
style consistency and return persistence.
[Insert Table VII About Here]
Table VII reports estimated coefficient values for a logit regression using the
same six combinations of explanatory variables explained previously, as well as a
seventh model that adds a term capturing the interaction between ALPHA and RSQ to
Model 6. To document the effect that the inclusion of a momentum factor has on return
persistence, we once again estimate ALPHA and RSQ with both the FF (i.e., three-factor)
and FFC (i.e., four-factor) forms of the return-generating model in (4). The conclusions
that can be drawn from these data are, if anything, more dramatic than those suggested by
Panel B of Table V. In particular, while the alpha persistence effect exists in isolation, its
magnitude is: (i) reduced substantially when measured net of a return momentum factor,
and (ii) virtually eliminated when all additional control variables are added as regressors
using the FFC definitions of ALPHA and RSQ (i.e., ALPHA becomes statistically
insignificant in Models 5 and 6). On the other hand, style consistency continues to have a
positive impact on future performance, regardless of how RSQ was estimated or what
other variables are included as explanatory variables. Further, the effect that style
consistency has on a manager’s ability to generate returns in the upper half of his or her
peer group, while reduced somewhat, remains strong even after controlling for portfolio
turnover and fund expenses. Finally, the interaction between ALPHA and RSQ is, at
best, only marginally significant and appears to be highly dependent on how the two
variables are measured.
[Insert Table VIII About Here]
To get a better sense of how alpha persistence and managing a style consistent
portfolio can indicate an improvement in an investor’s chance of delivering superior
future returns, Table VIII lists the probability of beating the median peer manager when
ALPHA and RSQ fall within a particular cohort cell while holding the other explanatory
variables constant. Specifically, Panel A sets the levels of TURN, EXPR and TNA equal
to their standardized mean values of zero while Panel B modifies these controls by setting
EXPR two standard deviations below its mean (i.e., the lowest expense ratio cohort with

28
average turnover and average assets under management). In both panels, funds within a
style group and year are sorted into cohorts delineated by the number of standard
deviations each variables falls from its mean (e.g., a fund in (-2, +1) cohort produced an
ALPHA at least two standard deviations below the average and a RSQ at least one
standard deviation above the norm). The columns of the display represent the differential
effect of ALPHA for a given level of RSQ, while reading across a row shows how style
consistency increases the probability of being an above-average manager given a certain
level of past abnormal performance. The final row and column report the difference in
proportions for the highest and lowest ALPHA and RSQ effects, controlling for the other
effect, respectively. The FFC form of Model 7 was used to calculate these probabilities.
For the base case in Panel A, the first effect to notice is that funds in the (0,0)
cohort—those producing average past alpha and style consistency levels—essentially
have an equal chance (i.e., a reported proportion of 0.5010) of finishing above the
average in a subsequent annual style tournament. With that as a benchmark, notice also
that funds falling in the two extreme cohorts have markedly different possibilities of
future success: a 44.67 percent probability for the (-2, -2) cohort and a 60.24 percent
probability for the (+2, +2). More importantly, however, it appears that remaining
consistent to a particular style mandate appears to improve the chance of future
outperformance more than producing superior past performance does. For instance,
looking down the first column of data, for the most style-inconsistent cohort, the
probability of being an above-median manager is less than 50 percent regardless of what
past alpha level the manager generated. Conversely, across the lowest ALPHA cohort,
the probability of producing above-average future returns increases from less than 45
percent to more than 51 percent as style consistency improves. At the other extreme,
although higher alpha levels increase the chance of outperformance in the highest RSQ
cohort—from 51.27 to 60.24 percent—the effect in the other direction is even larger; the
difference between the least and most style-consistent funds in the highest alpha cohort is
over 16 percent. Panel B of Table VIII indicates that these differential effects remain
relatively stable when the base case is changed to include those managers in the lowest
expense cohort, although the specific cell levels are uniformly higher (e.g., the
probability in the (+2, +2) cell increases from 60.24 to 69.33 percent). Thus, these

29
findings again support the conclusion that a fund’s investment style is positively related
to its future outperformance, even after accounting for past abnormal returns.

E. Style Consistency-Based Trading Strategies

The findings presented thus far mainly emphasize the strong degree of statistical
significance that defines the relationship between the ability of a fund manager to
maintain a consistent investment style and that fund’s subsequent return performance. In
this section, we extend this analysis by examining another perspective of the economic
impact of style-consistent investing. Specifically, we ask the following question:
Controlling for portfolio expenses and past performance, would investors be able to
exploit the return differential (if any) generated by style-consistent and style-inconsistent
portfolios? To address this issue, we calculate the returns to several portfolios sorted by
various combinations of fund expense ratios (EXPR), past fund performance (ALPHA),
and style consistency (RSQ). In particular, at the beginning of each quarter during the
1991-2000 period, the total sample of funds was sorted into quintile portfolios based on
their EXPR levels. The following different sample universes were then defined using
funds from all style groups to quantify the return impact of style consistency: (i) the
whole sample divided into the “extreme” combinations of expense control quintiles: [low
expense quintile (Lo EXPR); high expense quintile (Hi EXPR)], (ii) the sample divided
into expense control quintiles and above or below median RSQ [high consistency and
low expenses (Hi RSQ, Lo EXPR); low consistency and high expenses (Lo RSQ, Hi
EXPR)], (iii) the sample divided into extreme combinations of expense-alpha control
quintiles [low expense and high alpha (Lo EXPR, Hi ALPHA); high expense and low
alpha (Hi EXPR, Lo ALPHA)], and (iv) the sample divided into expense-alpha control
quintiles and above or below median RSQ [high consistency, low expense and high alpha
(Hi RSQ, Lo EXPR, Hi ALPHA); low consistency, high expense and low alpha (Lo
RSQ, Hi EXPR, Lo ALPHA)]. With each sorting procedure, returns for every quintile
portfolio were calculated for the subsequent three-month period, at which time all
portfolios were rebalanced by rolling the 36-month estimation interval forward one
quarter.
[Insert Figure 2 About Here]

30
 Figure 2 illustrates the investment performance for the various portfolios just
described. The data displayed indicate the cumulative impact of remaining invested
(with rebalancing) in a given portfolio over the entire ten-year period. Performance is
scaled relative to an initial January 1991 value of 1.00 and then continues through
December 2000. Panel A compares the cumulative performance differential of investing
in high and low style-consistent funds that have first been segmented into extreme control
quintile cross-sections based on expenses. Panel B then illustrates this return differential
using similar control quintile portfolios sorted using both expenses and alphas.
Panel A shows the cumulative performance of four separate portfolios. To set a
benchmark for the value added by a style-consistent investment mandate, the first set of
portfolios compare investments in a low expense portfolio (i.e., Lo EXPR) and high
expense portfolio (i.e., Hi EXPR) without regard to a fund’s RSQ level. The respective
terminal values of a one dollar investment are $4.1024 and $3.4202, which correspond to
average annual returns of 15.58 and 13.44 percent. The second set of portfolios modifies
the first by adding the style consistency dimension to the mix. Specifically, the solid
lines in the chart highlight the effect of isolating: (i) the high RSQ portion of the (Lo
EXPR) quintile and (ii) the low RSQ portion of the (Hi EXPR) quintile. The terminal
wealth levels for these two modifications are $4.1747 and $3.3203, respectively, with
corresponding average annual returns of 15.79 and 13.10 percent. Notice that the 2.69
percent differential in annual returns between the latter set of portfolios—which can be
interpreted as the return to a hypothetical hedge fund that is long in high style consistency
and short in low consistency—is greater than the 2.14 percent difference between the two
control portfolios. This 55 basis point differential can be thought of as a “consistency
premium” across all individual tournaments with comparable expense ratios.25

25
It is worth noting that this 55 basis point return differential very likely understates the true consistency
premium. To see why, notice that the 214 basis point difference in average annual returns between the low
and high expense subsamples is actually a larger effect than can be explained by expense ratio differentials
alone, which according to the descriptive statistics in Table II only ranged on average from about 1.25 to
1.65 percent depending on the style group. As importantly, however, Table IV reported a strong negative
correlation between a fund’s style consistency measures and its expense ratio. Consequently, even before
the Lo EXPR and Hi EXPR subsamples are further segmented by RSQ, we would expect that part of the
return differential between them owes to style consistency effects rather than fund expense differentials per
se. This is in turn suggests that what we label a “consistency premium” is a downward-biased measure of
the true, total impact that consistency exerts on returns.

31
 Panel B shows the performance of style-based strategies controlling for expenses as
well as prior performance. The two control portfolios are based on low expense and high
alpha quintiles (Lo EXPR, Hi ALPHA) and high expense and low alpha quintiles (Hi
EXPR, Lo ALPHA). The terminal values of these portfolios are $4.1019 and $2.9262,
respectively. This corresponds to average annual returns of 15.58 and 11.64 percent.
The second set of portfolios includes high and low style consistency as part of the
rebalancing strategy. The high consistency, low expense and high alpha portfolio (Hi
RSQ, Lo EXPR, Hi ALPHA) has a terminal value of $4.2782 while the low consistency,
high expense and low alpha strategy (Lo RSQ, Hi EXPR, Lo ALPHA) has a terminal
value of $2.5639. The respective annual returns are 16.08 and 10.14 percent. This latter
trading strategy, which controls both for expenses and past performance, produces a
much higher “consistency premium” of 200 basis points (3.94 percent return differential
for the control groups versus 5.94 percent for the style-based strategies). Thus, it appears
that style consistency does add economic value with respect to expense and alpha-based
trading strategies.26

V. Conclusion
One of the more interesting intellectual developments in the investment management area
during the past few decades has been the evolution in the way in which a portfolio’s
investment style is defined and the role that this style subsequently plays in determining
fund returns. Both theory and practice appear to have settled on two salient dimensions
that define a portfolio’s style: the market capitalization of the typical fund holding (i.e.,
the “size” dimension) and the fundamental attributes of that composite holding (i.e., the
“value-growth” dimension). While considerable effort has been put toward establishing
whether a manager’s selection of a particular set of style characteristics over another
matters, relatively little is known about whether the manager’s ability to consistently

26
Recognize that the consistency premium just described is an average across all the individual style
tournaments. Although not shown in Figure 2, we also calculated a similar methodological design within
each of the nine separate style groups. This experiment produced an advantage for the Hi RSQ portfolio in
eight cases. Specifically, after the controlling for expenses and past performance, the consistency premium
in each of the style tournaments (i.e., average annual return for Hi RSQ partition minus average annual
return for Lo RSQ partition) is as follows: LV: 3.07%, LB: 0.85%, LG: 1.89%, MV: 2.40%, MB: 0.54%,
MG: 0.19%, SV: -1.80%, SB: 7.16%, and SG: 4.60%.

32
execute his or her style mandate—whatever that may be—also has a significant impact
on investment performance.
Does investment style consistency matter? The results of this study strongly
suggest that the answer is “yes”. Using two different statistical measures of consistency,
we test three specific hypotheses related to this issue, namely that: (i) a negative
relationship exists between portfolio style consistency and portfolio turnover, (ii) a
positive relationship exists between a fund’s style consistency and the future actual and
relative returns it produces, and (iii) a positive relationship exists between the consistency
of a portfolio’s investment style and the persistence of its performance over time. Based
on a survivorship bias-free sample of several thousand mutual funds drawn from nine
distinct style groups over the period 1991-2000, the data provide support for all three
propositions under a wide variety of different conditions and alternative possibilities.
Regardless of whether the definition of style consistency is model-based (i.e., R2) or
benchmark-based (i.e., tracking error), high-consistency funds do indeed tend to have
lower portfolio turnover and expense ratios than low-consistency funds. This
undoubtedly contributes to the additional result that greater style consistency is positively
associated with both higher overall returns as well as higher relative returns within a
given investment class. Importantly, however, style consistency is not simply a surrogate
for portfolio turnover; even after controlling for the latter, the relationship between a
portfolio’s style consistency and its future returns remains significant. Thirdly, we also
confirm the positive correlation between consistency and the persistence of fund returns
and show that this connection is distinct from—and of comparable magnitude to—past
performance (i.e., alpha) and fund expense ratio. The inclusion of a momentum factor in
the measurement of expected returns does nothing to diminish this consistency effect, nor
does the exclusion of index funds from the sample. Finally, the performance of simulated
consistency-based trading strategies suggests that these effects are economically as well
as statistically significant.
These findings evoke several implications and extensions. Most notably, it appears
that the ability for a portfolio manager to remain consistent to his or her designated
investment style is a valuable skill. It may, in fact, be the case that maintaining an
observable level of consistency in their investment style is one of the ways in which

33
superior managers attempt to signal their skills to investors. Further, while this
conclusion does not appear to be time dependent, there is some evidence to suggest
consistency is a more valuable talent within some style classes (e.g., large- and small-
cap) than others (e.g., mid-cap). Also, although our results do not negate the possibility
that managers who follow an explicit tactical style timing strategy can be successful, they
do suggest that unintentional style drift can lead to inferior relative performance; indeed,
the decision to remain style consistent may be more useful in helping managers avoid
consistently poor performance than creating an environment that fosters persistent
superior relative returns. Lastly, given related research in this area, it also may be the
case that the ability to maintain a style-consistent portfolio increases the likelihood that
the manager will remain employed at the end of an evaluation period. At a minimum, it
seems clear that style consistency is another element that must be factored into the on-
going debate of whether mutual fund performance persists over time.

34
 REFERENCES

Ammann, Manuel, and Heinz Zimmerman, 2001, Tracking error and tactical asset
allocation, Financial Analysts Journal 57, 32-43.

Banz, Rolf W., 1981, The relationship between return and market value of common
stocks, Journal of Financial Economics 9, 3-18.

Barberis, Nicholas, and Andrei Shleifer, 2000, Style investing, Working Paper,
University of Chicago and Harvard University.

Basu, Sanjoy, 1977, Investment performance of common stocks in relation to their price-
earnings ratios: A test of the efficient market hypothesis, Journal of Finance 32, 663-682.

Bogle, John C., 1998, The implications of style analysis for mutual fund performance
evaluation, Journal of Portfolio Management 24, 34-42.

Bollen, Nicolas P. B., and Jeffrey A. Busse, 2001, On the timing ability of mutual fund
managers, Journal of Finance 56, 1075-1094.

Brown, Keith C., and Gregory D. Brown, 1987, Does the composition of the market
portfolio really matter?, Journal of Portfolio Management 13, 26-32.

Brown, Keith C., W. V. Harlow, and Laura T. Starks, 1996, Of tournaments and
temptations: An analysis of managerial incentives in the mutual fund industry, Journal of
Finance 51, 85-110.

Brown, Stephen J., William N. Goetzmann, Roger G. Ibbotson, and Stephen A. Ross,
1992, Survivorship bias in performance studies, Review of Financial Studies 5, 553-580.

Brown, Stephen J., and William N. Goetzmann, 1995, Performance persistence, Journal
of Finance 50, 679-698.

Brown, Stephen J., and William N. Goetzmann, 1997, Mutual fund styles, Journal of
Financial Economics 43, 373-399.

Carhart, Mark M., 1997, On persistence in mutual fund performance, Journal of Finance
52, 57-82.

Carlson, Robert S., Aggregate performance of mutual funds, 1948-1967, Journal of

Financial and Quantitative Analysis 5, 1-32.

Capaul, Carlo, Ian Rawley, and William F. Sharpe, 1993, International value and growth
stock returns, Financial Analysts Journal 49, 27-36.

35
Capon, Noel, Gavin Fitzsimons, and Roger Prince, 1996, An individual level analysis of
the mutual fund investment decision, Journal of Financial Services Research 10, 59-82.

Chan, Louis K.C., Hsiu-Lang Chen, and Josef Lakonishok, 1999, On mutual fund
investment styles, National Bureau of Economic Research Working Paper.

Chang, Eric C., and Wilbur G. Lewellen, 1993, Market timing and mutual fund
investment performance, Journal of Business 57, 57-72.

Chevalier, Judith, and Glenn Ellison, 1997, Risk taking by mutual funds as a response to
incentives, Journal of Political Economy 105, 1167-1200.

Christopherson, Jon A., Wayne E. Ferson, and Debra A. Glassman, 1998, Conditioning
manager alphas on economic information: Another look at the persistence of
performance, Review of Financial Studies 11, 111-142.

Coggin, T. Daniel, Frank J. Fabozzi, and Shafiqur Rahman, 1993, The investment
performance of U.S. equity pension fund managers: An empirical investigation, Journal
of Finance 48, 1039-1055.

Daniel, Kent, Mark Grinblatt, Sheridan Titman, and Russ Wermers, 1997, Measuring
mutual fund performance with characteristic-based benchmarks, Journal of Finance 52,
1035-1058.

Daniel, Kent, and Sheridan Titman, 1997, Evidence on the characteristics of cross
sectional variation in stock returns, Journal of Finance 52, 1-33.

diBartolomeo, Dan, and Erik Witkowski, 1997, Mutual fund misclassification: Evidence
based on style analysis, Financial Analysts Journal 53, 32-43.

Dorian, John L., and Robert D. Arnott, 1995, Tactical style management, in T. Coggin
and F. Fabozzi, eds.: The Handbook of Equity Style (Fabozzi Associates, New Hope, PA).

Fama, Eugene F., and Kenneth R. French, 1992, The cross-section of expected returns,
Journal of Finance 47, 427-466.

Fama, Eugene F., and Kenneth R. French, 1993, Common risk factors in the returns on
stocks and bonds, Journal of Financial Economics 33, 3-56.

Fama, Eugene F., and Kenneth R. French, 1998, Value versus growth: The international
evidence, Journal of Finance 53, 1975-1999.

Elton, Edwin J., Martin J. Gruber, and Christopher R. Blake, 1996, The persistence of
risk-adjusted mutual fund performance, Journal of Business 69, 133-157.

36
Gallo, John G., and Larry J. Lockwood, 1999, Fund management changes and equity
style shifts, Financial Analysts Journal 55, 44-52.

Grinblatt, Mark, and Sheridan Titman, 1992, Mutual fund performance: An analysis of
quarterly portfolio holdings, Journal of Business 62, 393-416.

Grinblatt, Mark, and Sheridan Titman, 1992, The persistence of mutual fund
performance, Journal of Finance 47, 1977-1984.

Grinold, Richard C., and Ronald N. Kahn, 1995, Active Portfolio Management (Probus
Publishing, Chicago, Ill.).

Hendricks, Daryll, Jayendu Patel, and Richard Zeckhauser, 1993, Hot hands in mutual
funds: Short-run persistence of relative performance, 1974-1988, Journal of Finance 48,
93-130.

Investment Company Institute, 2001, Mutual Fund Fact Book (Investment Company
Institute, Washington, DC).

Jegadeesh, Narasimham, and Sheridan Titman, 1993, Returns to buying winners and
selling losers: Implications for stock market efficiency, Journal of Finance 48, 65-91.

Jensen, Michael, 1968, The performance of mutual funds in the period 1945-1964,
Journal of Finance 23, 389-416.

Kahn, Ronald N., and Andrew Rudd, 1995, Does historical performance predict future
performance, Financial Analysts Journal 51, 43-52.

Khorana, Ajay, 1996, Top management turnover: An empirical investigation of mutual

fund managers, Journal of Financial Economics 40, 403-427.

Kon, Stanley J., 1983, The market-timing performance of mutual fund managers, Journal
of Business 56, 323-347.

Lakonishok, Josef, Andrei Shleifer, and Robert W. Vishny, 1994, Contrarian investment,
extrapolation, and risk, Journal of Finance 49, 1541-1578.

Lehman, Bruce N., and David M. Modest, 1987, Mutual fund performance evaluations: A
comparison of benchmarks and benchmark comparisons, Journal of Finance 42, 233-
265.

Loughran, Tim, 1997, Book-to-market across firm size, exchange, and seasonality: Is
there an effect?, Journal of Financial and Quantitative Analysis 32, 249-268.

Malkiel, Burton G., 1995, Returns from investing in equity mutual funds: 1971 to 1991,
Journal of Finance 50, 549-572.

37
McDonald, John G., 1974, Objectives and performance of mutual funds, 1960-1969,
Journal of Financial and Quantitative Analysis 9, 311-333.

Roll, Richard, 1995, Style return differentials: Illusions, risk premiums, or investment
opportunities, in T. Coggin and F. Fabozzi, eds.: The Handbook of Equity Style (Fabozzi
Associates, New Hope, PA).

Sharpe, William F., 1966, Mutual fund performance, Journal of Business 39, 119-138.

Sharpe, William F., 1992, Asset allocation: Management style and performance
measurement, Journal of Portfolio Management 18, 7-19.

Sirri, Eric R., and Peter Tufano, 1998, Costly search and mutual fund flows, Journal of
Finance 53, 1589-1622.

Sorenson, Eric H., and Craig J. Lazarra, 1995, Equity style management: The case of
growth and value, in R. Klein and J. Lederman, eds.: Equity Style Management:
Evaluating and Selecting Investment Styles (Irwin, Chicago, IL).

Teo, Melvyn, and Sung-Jun Woo, 2001, Persistence in style-adjusted mutual fund
returns, Harvard University working paper.

Wermers, Russ, 2000, Mutual fund performance: An empirical decomposition into stock-
picking talent, style, transaction costs, and expenses, Journal of Finance 55, 1655-1695.

Wermers, Russ, 2001, Predicting mutual fund returns, Working Paper, University of
Maryland.

38
 Table I
Mutual Fund Style Sample By Year

This table reports the number of mutual funds included in each Morningstar style objective category by year for the sample period
spanning January 1991 to December 2000. The numbers listed represent those funds with at least 36 months of return history prior
to the given date. Morningstar uses the following objective classifications: large-cap value (LV), large-cap blend (LB), large-cap
growth (LG), mid-cap value (MV), mid-cap blend (MB), mid-cap growth (MG), small-cap value (SV), small-cap blend (SB), small-
cap growth (SG). Averages (rounded to the nearest fund) are also listed for two non-overlapping subsets of the 10-year sample
period. The compound annual growth rate for the number of funds in each style category are reported in the last row.

Morningstar Mutual Fund Style Category:

Year LV LB LG MV MB MG SV SB SG Total

1991 135 163 118 60 47 79 25 29 42 698

1992 140 172 120 60 49 78 28 30 44 721
1993 156 184 126 65 54 78 31 30 49 773
1994 169 203 139 67 54 82 38 37 59 848
1995 215 245 178 69 62 106 47 52 78 1052
1996 273 314 233 87 71 150 62 71 113 1374
1997 350 382 297 102 99 183 79 97 152 1741
1998 410 446 355 127 104 221 97 123 206 2089
1999 504 584 425 167 125 289 121 147 262 2624
2000 564 729 549 199 138 333 162 194 309 3177

Averages:
91-95 136 161 114 54 44 71 28 30 45 ---
96-00 350 409 310 114 90 196 87 105 194 ---

% Annual
Growth: 17.22% 18.11% 18.63% 14.25% 12.71% 17.33% 23.08% 23.51% 24.82% 18.34%

39
 Table II
Mutual Fund Style Sample: Descriptive Statistics

This table reports descriptive statistics for the mutual fund sample, broken down by style classification and time period. Reported in
Panel A for each style category are: the average fund annual total (i.e., capital gain and income distribution) return, average fund
return standard deviation, average market capitalization of fund holdings, average fund expense ratio, and average annual fund
turnover (defined as the ratio of fund sales to total fund holdings, measured in dollar volumes). Panel B provides the numerical
differences in each characteristic between extreme category pairs, with p-values summarizing the statistical significance of those
differences listed parenthetically.

Panel A. Characteristics of Mutual Funds by Style

Avg. Fund
Avg. Annual Avg. Fund Avg. Fund Firm Expense Ratio Avg. Fund
Style Group Period Fund Return (%) Std. Dev. (%) Size ($MM) (%) Turnover (%)

Large Value 1991 -2000 12.09 15.21 25,298 1.38 67.57

(LV) 1991-1995 16.90 9.37 9,909 1.19 60.28
1996-2000 12.01 15.85 26,913 1.38 67.65

Large Blend 1991 -2000 7.76 16.47 44,611 1.22 69.14

(LB) 1991-1995 16.64 9.54 13,058 1.10 69.30
1996-2000 7.63 17.40 46,969 1.22 68.83

Large Growth 1991 -2000 6.85 22.39 45,381 1.45 92.93

(LG) 1991-1995 15.80 11.98 10,622 1.33 83.77
1996-2000 6.89 23.43 48,034 1.44 93.20

Mid Value 1991 -2000 18.06 16.00 5,731 1.43 84.73

(MV) 1991-1995 16.91 9.88 3,641 1.36 73.66
1996-2000 17.93 17.07 5,887 1.42 85.45

Mid Blend 1991 -2000 11.74 17.63 6,782 1.45 79.39

(MB) 1991-1995 13.82 11.25 2,392 1.37 56.35
1996-2000 11.97 19.13 7,044 1.45 80.23

Mid Growth 1991 -2000 14.94 31.04 4,917 1.55 132.96

(MG) 1991-1995 18.25 14.15 1,882 1.37 106.92
1996-2000 14.69 33.36 5,109 1.55 133.79

Small Value 1991 -2000 20.60 17.64 643 1.48 61.43

(SV) 1991-1995 20.26 11.29 456 1.28 54.52
1996-2000 19.33 18.18 657 1.47 60.80

Small Blend 1991 -2000 17.74 21.71 1,283 1.50 82.17

(SB) 1991-1995 16.86 12.09 2,644 1.55 69.43
1996-2000 17.39 22.49 1,297 1.49 81.72

Small Growth 1991 -2000 12.95 34.69 1,057 1.64 119.89

(SG) 1991-1995 19.71 15.17 800 1.49 89.93
1996-2000 12.33 35.80 1,069 1.64 120.54

40
 Table II (cont.)
Mutual Fund Style Sample: Descriptive Statistics

Panel B. Differences in Characteristics

Avg. Fund
Style Group Avg. Annual Avg. Fund Avg. Fund Firm Expense Ratio Avg. Fund
Comparison Period Fund Return (%) Std. Dev. (%) Size ($MM) (%) Turnover (%)

Ratio-Based:

LV - LG 1991-2000 5.24 -7.18 -20,083 -0.08 -25.36

(0.00) (0.00) (0.00) (0.10) (0.00)
1991-1995 1.10 -2.60 -713 -0.14 -23.49
(0.18) (0.00) (0.45) (0.21) (0.00)
1996-2000 5.13 -7.59 -21,121 -0.07 -25.55
(0.00) (0.00) (0.00) (0.15) (0.00)
MV - MG 1991 -2000 3.12 -15.05 814 -0.12 -48.23
(0.19) (0.00) (0.01) (0.01) (0.00)
1991-1995 -1.33 -4.27 1,759 -0.01 -33.26
(0.45) (0.00) (0.00) (0.92) (0.00)
1996-2000 3.24 -16.29 778 -0.13 -48.34
(0.20) (0.00) (0.02) (0.01) (0.00)
SV - SG 1991 -2000 7.65 -17.05 -413 -0.16 -58.46
(0.00) (0.00) (0.00) (0.01) (0.00)
1991-1995 0.55 -3.88 -343 -0.21 -35.41
(0.86) (0.00) (0.05) (0.13) (0.00)
1996-2000 7.00 -17.63 -412 -0.17 -59.74
(0.00) (0.00) (0.00) (0.01) (0.00)

Size-Based:

LV - SV 1991 -2000 -8.51 -2.44 24,655 -0.10 6.14

(0.00) (0.00) (0.00) (0.15) (0.10)
1991-1995 -3.36 -1.92 9,453 -0.09 5.76
(0.04) (0.00) (0.00) (0.65) (0.43)
1996-2000 -7.31 -2.33 26,256 -0.09 6.85
(0.00) (0.00) (0.00) (0.21) (0.07)
LB - SB 1991 -2000 -9.98 -5.24 43,328 -0.28 -13.03
(0.00) (0.00) (0.00) (0.00) (0.01)
1991-1995 -0.22 -2.55 10,414 -0.46 -0.14
(0.85) (0.00) (0.00) (0.00) (0.99)
1996-2000 -9.76 -5.09 45,672 -0.28 -12.89
(0.00) (0.00) (0.00) (0.00) (0.01)
LG - SG 1991 -2000 -6.10 -12.31 44,324 -0.19 -26.96
(0.00) (0.00) (0.00) (0.00) (0.00)
1991-1995 -3.91 -3.20 9,822 -0.16 -6.16
(0.01) (0.00) (0.00) (0.17) (0.51)
1996-2000 -5.44 -12.37 46,966 -0.19 -27.34
(0.01) (0.00) (0.00) (0.00) (0.00)

41
 Table III
Mutual Fund Style Consistency by Category

This table reports style consistency statistics for the mutual fund sample over the period January 1991 - December 2000. Funds within a style objective are grouped by two
measures related to investment style consistency: (i) average R2, measured relative to the multi-factor return-generating model in equation (4); and (ii) average annual
tracking error relative to the style-specific benchmark, as calculated by equation (3). For each measure and style group, funds are separated into “high” consistency and
“low” consistency groups relative to the category-wide median values of R2 (Panel A) or TE (Panel B). Consistency rankings are based on fund returns for the 36-month
period preceding the year for which the reported characteristics are produced. Average values of R2, annual TE, annual peer rankings, annual portfolio returns, return
standard deviations, portfolio turnover, and expense ratios are aggregated across the ten-year sample period (1991-2000) used to rank fund consistency.

Panel A. Style Consistency Defined by R2

Style Median Tracking Median Peer Group Median Annual Median Fund Std. Median Fund Median Fund
Style Group Consistency Median R2 Error (%) Ranking Fund Return (%) Dev. (%) Turnover (%) Expense Ratio
Large Value Low 0.86 5.23 47.10 11.10 10.40 47.50 1.22
(LV) High 0.93 3.75 52.17 13.05 10.15 45.50 1.02

Large Blend Low 0.88 4.93 38.01 16.69 11.57 77.00 1.25
(LB) High 0.96 2.85 59.17 20.04 11.15 38.00 0.93

Large Growth Low 0.83 7.44 47.79 18.55 15.00 68.00 1.36
(LG) High 0.92 4.94 53.31 19.86 13.18 60.50 1.07

Mid Value Low 0.77 7.46 41.41 17.30 9.90 63.00 1.40
(MV) High 0.87 5.07 54.84 13.58 10.33 60.00 1.16

Mid Blend Low 0.75 8.24 46.02 12.95 12.58 63.00 1.41
(MB) High 0.87 4.89 52.84 12.86 12.05 39.59 1.23

Mid Growth Low 0.80 8.72 54.86 13.90 17.68 115.00 1.40
(MG) High 0.88 5.92 49.44 15.44 15.88 76.00 1.29

Small Value Low 0.75 7.85 46.67 15.83 11.66 50.00 1.39
(SV) High 0.87 5.02 55.26 16.65 12.43 44.82 1.15

Small Blend Low 0.77 8.40 45.95 14.28 13.20 84.50 1.50
(SB) High 0.89 5.85 54.23 15.62 13.93 47.00 1.12

Small Growth Low 0.81 8.74 43.44 12.78 16.75 89.00 1.46
(SG) High 0.90 6.60 55.94 14.21 18.66 78.00 1.33

42
 Table III (cont.)
Mutual Fund Style Consistency by Category

Panel B. Style Consistency Defined by Tracking Error

Style Median Tracking Median Peer Group Median Annual Median Fund Std. Median Fund Median Fund
Style Group Consistency Error (%) Median R2 Ranking Fund Return (%) Dev. (%) Turnover (%) Expense Ratio
Large Value Low 5.27 0.86 46.69 11.86 11.07 50.00 1.19
(LV) High 3.70 0.93 52.63 13.28 9.91 43.00 0.96

Large Blend Low 5.06 0.88 38.42 16.93 12.12 76.00 1.24
(LB) High 2.85 0.96 55.92 20.12 11.09 38.00 0.94
Large Growth Low 7.52 0.85 46.36 16.85 15.57 77.00 1.29
(LG) High 4.55 0.91 52.57 20.94 12.96 48.50 1.09

Mid Value Low 7.63 0.77 48.68 14.65 10.45 63.00 1.33
(MV) High 4.99 0.86 53.64 12.57 10.00 66.00 1.22
Mid Blend Low 8.31 0.75 42.71 15.82 14.52 58.00 1.39
(MB) High 4.89 0.87 55.11 10.74 11.35 51.00 1.20

Mid Growth Low 9.17 0.82 53.63 11.27 19.07 118.50 1.46
(MG) High 5.78 0.87 49.72 15.69 14.93 76.50 1.22

Small Value Low 7.91 0.75 41.07 14.05 11.66 50.00 1.39
(SV) High 5.02 0.87 55.36 18.05 12.03 47.00 1.14
Small Blend Low 8.52 0.77 41.38 14.30 13.22 85.00 1.44
(SB) High 5.58 0.89 58.62 14.66 13.89 45.00 1.19

Small Growth Low 9.18 0.81 43.90 9.32 17.88 93.00 1.41
(SG) High 6.51 0.89 57.32 15.78 18.50 82.00 1.38

43
 Table IV

Style Consistency Correlation Coefficients

This table lists Pearson correlation coefficients between the two measures of investment style consistency (i.e., R2 and TE) and
variables related to fund management and performance. Fund management variables include annual portfolio turnover and annual
fund expense ratio. Fund performance variables include actual annual return, “tournament” annual return (i.e., standardized by year
within a fund’s particular style classification), and the peer ranking of that tournament return. Consistency measures are based on
fund returns for the 36-month period preceding the year for which the management and performance variables are produced. Separate
correlation coefficients are reported for: (i) the entire 1991-2000 sample period, and (ii) each individual year in the sample period. P-
values are listed parenthetically beside each correlation statistic.

Panel A. Correlation with R2

Variable:

Fund Actual Tournament Tournament

Period Fund Turnover Expense Ratio Fund Return Fund Return Return Ranking
1991-2000 -0.216 (0.000) -0.318 (0.000) 0.029 (0.000) 0.110 (0.000) 0.092 (0.000)
1991 -0.185 (0.000) -0.254 (0.000) 0.034 (0.411) 0.031 (0.449) 0.057 (0.170)
1992 -0.246 (0.000) -0.305 (0.000) 0.108 (0.006) 0.110 (0.006) 0.094 (0.018)
1993 -0.195 (0.000) -0.330 (0.000) -0.058 (0.128) -0.054 (0.160) -0.031 (0.417)
1994 -0.260 (0.000) -0.410 (0.000) 0.159 (0.000) 0.170 (0.000) 0.077 (0.037)
1995 -0.277 (0.000) -0.369 (0.000) 0.240 (0.000) 0.278 (0.000) 0.236 (0.000)
1996 -0.240 (0.000) -0.394 (0.000) 0.291 (0.000) 0.301 (0.000) 0.241 (0.000)
1997 -0.180 (0.000) -0.345 (0.000) 0.265 (0.000) 0.329 (0.000) 0.240 (0.000)
1998 -0.166 (0.000) -0.329 (0.000) 0.089 (0.000) 0.147 (0.000) 0.141 (0.000)
1999 -0.246 (0.000) -0.313 (0.000) -0.088 (0.000) -0.082 (0.000) -0.043 (0.058)
2000 -0.233 (0.000) -0.250 (0.000) 0.044 (0.030) 0.035 (0.083) 0.025 (0.217)

Panel B. Correlation with TE

Variable:

Fund Actual Tournament Tournament

Period Fund Turnover Expense Ratio Fund Return Fund Return Return Ranking
1991-2000 0.238 (0.000) 0.364 (0.000) -0.012 (0.177) -0.091 (0.000) -0.078 (0.000)
1991 0.213 (0.000) 0.303 (0.000) 0.110 (0.008) 0.155 (0.000) 0.097 (0.020)
1992 0.242 (0.000) 0.328 (0.000) -0.002 (0.960) 0.024 (0.543) 0.014 (0.731)
1993 0.197 (0.000) 0.358 (0.000) 0.184 (0.000) 0.171 (0.000) 0.120 (0.002)
1994 0.279 (0.000) 0.392 (0.000) -0.174 (0.000) -0.181 (0.000) -0.120 (0.001)
1995 0.289 (0.000) 0.431 (0.000) -0.146 (0.000) -0.177 (0.000) -0.125 (0.000)
1996 0.304 (0.000) 0.465 (0.000) -0.341 (0.000) -0.340 (0.000) -0.273 (0.000)
1997 0.256 (0.000) 0.439 (0.000) -0.358 (0.000) -0.426 (0.000) -0.341 (0.000)
1998 0.199 (0.000) 0.392 (0.000) -0.098 (0.000) -0.166 (0.000) -0.148 (0.000)
1999 0.235 (0.000) 0.350 (0.000) 0.181 (0.000) 0.226 (0.000) 0.177 (0.000)
2000 0.247 (0.000) 0.280 (0.000) -0.116 (0.000) -0.116 (0.000) -0.106 (0.000)

44
 Table V

Style Consistency and Fund Performance Regression Results

This table reports results for the 1991-2000 sample period of the regression of future fund returns on past abnormal returns (ALPHA) and past style consistency (RSQ).
ALPHA and RSQ are estimated over a 36-month period by two different versions of equation (4): the Fama-French three-factor model (i.e., FF) and Carhart’s extension that
includes a return momentum factor (i.e., FFC). Future returns are measured for the t-month period following a given 36-month estimation window; Panels A and B report
values for t=3 and t=12, respectively. Additional control regressors include portfolio turnover (TURN), fund expense ratio (EXPR), and total net fund assets (TNA). All
variables are standardized by year and fund style class. P-values are listed parenthetically beneath each coefficient.

Panel A. Three-Month Future Returns as Dependent Variable

FF Three-Factor Model: FFC Four-Factor Model:

Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Intercept 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000)

ALPHA 0.075 0.073 0.073 0.057 0.058 0.021 0.019 0.020 0.011 0.011
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.012) (0.011)

RSQ 0.050 0.047 0.053 0.034 0.034 0.049 0.048 0.054 0.030 0.030
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

TURN 0.026 0.032 0.032 0.026 0.034 0.033

(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

EXPR -0.067 -0.068 -0.081 -0.082

(0.000) (0.000) (0.000) (0.000)

TNA -0.011 -0.008

(0.012) (0.093)

Adj. R2 0.006 0.002 0.008 0.008 0.012 0.012 0.000 0.002 0.003 0.003 0.009 0.009

# of Obs. 50,709 50,709 50,709 50,709 50,709 50,709 50,709 50,709 50,709 50,709 50,709 50,709

45
 Table V (cont.)

Style Consistency and Fund Performance Regression Results

Panel B. 12-Month Future Returns as Dependent Variable

FF Three-Factor Model: FFC Four-Factor Model:

Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Intercept 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000) (1.000)

ALPHA 0.091 0.090 0.090 0.059 0.060 0.056 0.050 0.052 0.038 0.038
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

RSQ 0.109 0.108 0.119 0.080 0.081 0.110 0.108 0.119 0.076 0.077
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

TURN 0.047 0.060 0.060 0.048 0.062 0.062

(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

EXPR -0.130 -0.134 -0.142 -0.145

(0.000) (0.000) (0.000) (0.000)

TNA -0.021 -0.019

(0.022) (0.038)

Adj. R2 0.008 0.012 0.020 0.022 0.036 0.036 0.003 0.012 0.015 0.017 0.034 0.034

# of Obs. 11,804 11,804 11,804 11,804 11,804 11,804 11,804 11,804 11,804 11,804 11,804 11,804

46
 Table VI

Style Consistency and Return Persistence: Evidence From Style Tournaments

This table reports results for the 1991-2000 sample period of the regression of future fund returns on past abnormal returns
(ALPHA) and past style consistency (RSQ), with three other regressors included as control variables: portfolio turnover
(TURN), fund expense ratio (EXPR), and total net fund assets (TNA). Panel A lists parameters estimates for each of the
nine Morningstar investment style groups separately. Panel B lists parameter estimates for six aggregated style groups:
three size-based (Large-, Mid-, Small-Cap) and three characteristic-based (Value, Blend, Growth). ALPHA and RSQ are
estimated over a 36-month period by Carhart’s extension of the Fama-French return-generating model in (4) that includes a
return momentum factor. Future returns are measured within each style group for the 12 month period following a given
36-month estimation window. All variables are standardized by year. P-values are listed parenthetically beneath each
reported parameter estimate.

Panel A. Individual Style Groups

Independent Variable Estimated Parameter: Coefficient of

Determination
Style Group Intercept ALPHA RSQ TURN EXPR TNA

Large Value (LV) 0.000 0.006 0.026 0.011 -0.251 0.006 0.068
n = 2353 (1.000) (0.753) (0.226) (0.601) (0.000) (0.763)

Large Blend (LB) 0.000 0.019 0.143 0.086 -0.130 0.006 0.050
n = 2616 (1.000) (0.328) (0.000) (0.000) (0.000) (0.759)

Large Growth (LG) 0.000 0.106 0.043 0.085 -0.132 -0.041 0.038
n = 2003 (1.000) (0.000) (0.072) (0.000) (0.000) (0.071)

Mid Value (MV) 0.000 0.027 0.076 0.232 -0.192 -0.050 0.073
n = 798 (1.000) (0.439) (0.033) (0.000) (0.000) (0.168)

Mid Blend (MB) 0.000 -0.001 0.031 -0.043 -0.144 -0.000 0.028
n = 650 (1.000) (0.971) (0.446) (0.276) (0.000) (0.999)

Mid Growth (MG) 0.000 0.017 -0.026 0.022 -0.020 -0.057 0.005
n = 1260 (1.000) (0.551) (0.393) (0.462) (0.484) (0.048)

Small Value (SV) 0.000 -0.119 0.110 0.061 -0.092 0.017 0.037
n = 528 (1.000) (0.006) (0.015) (0.165) (0.043) (0.711)

Small Blend (SB) 0.000 0.112 0.125 0.034 -0.200 -0.084 0.093
n = 606 (1.000) (0.007) (0.005) (0.400) (0.000) (0.047)

Small Growth (SG) 0.000 0.061 0.199 0.062 -0.066 -0.024 0.058
n = 982 (1.000) (0.057) (0.000) (0.053) (0.056) (0.456)

47
 Table VI (cont.)

Style Consistency and Return Persistence: Evidence From Style Tournaments

Panel B. Aggregated Style Groups

Independent Variable Estimated Parameter: Coefficient of

Determination
Style Group Intercept ALPHA RSQ TURN EXPR TNA

Large-Cap 0.000 0.043 0.070 0.061 -0.173 -0.008 0.044

n = 6974 (1.000) (0.000) (0.000) (0.000) (0.000) (0.504)

Mid-Cap 0.000 0.022 0.026 0.071 -0.094 -0.038 0.011

n = 2710 (1.000) (0.2577) (0.200) (0.000) (0.000) (0.055)

Small-Cap 0.000 0.031 0.159 0.049 -0.113 -0.029 0.047

n = 2118 (1.000) (0.154) (0.000) (0.025) (0.000) (0.186)

Value 0.000 0.001 0.057 0.069 -0.205 -0.004 0.048

n = 3681 (1.000) (0.964) (0.001) (0.000) (0.000) (0.807)

Blend 0.000 0.030 0.117 0.054 -0.145 -0.008 0.045

n = 3874 (1.000) (0.056) (0.000) (0.001) (0.000) (0.613)

Growth 0.000 0.073 0.061 0.066 -0.090 -0.039 0.021

n = 4247 (1.000) (0.000) (0.000) (0.000) (0.000) (0.011)

48
 Table VII

Style Consistency and Return Persistence: Logit Analysis

This table reports the findings for a logit analysis of the relationship between a fund manager’s tournament performance and several potential explanatory factors over the
period 1991-2000. Listed are coefficient estimates for logit regressions involving a future performance indicator variable and various combinations of the following
explanatory variables: past abnormal returns (ALPHA), past style consistency (RSQ), portfolio turnover (TURN), fund expense ratio (EXPR), total net fund assets (TNA), and
an interaction term with ALPHA and RSQ. ALPHA and RSQ are estimated over a 36-month period by two different versions of equation (4) using 12-month future returns:
the Fama-French three-factor model (i.e., FF) and Carhart’s extension that includes a return momentum factor (i.e., FFC). The dependent variable assumes the value of one if
a manager’s out-of-sample annual return is above the median for the relevant style group and period, 0 otherwise. P-values are listed parenthetically beneath each coefficient.

FF Three-Factor Model: FFC Four-Factor Model:

Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

Intercept 0.008 0.008 0.007 0.007 0.005 0.005 0.005 0.008 0.007 0.007 0.007 0.005 0.005 0.004
(0.679) (0.703) (0.711) (0.713) (0.789) (0.794) (0.788) (0.679) (0.703) (0.705) (0.709) (0.782) (0.785) (0.821)

ALPHA 0.075 0.074 0.076 0.041 0.042 0.048 0.045 0.039 0.042 0.027 0.027 0.043
(0.000) (0.000) (0.000) (0.038) (0.033) (0.029) (0.014) (0.036) (0.023) (0.161) (0.154) (0.039)

RSQ 0.147 0.147 0.164 0.115 0.116 0.115 0.147 0.145 0.163 0.112 0.112 0.115
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

TURN 0.073 0.092 0.092 0.093 0.074 0.094 0.094 0.096

(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

EXPR -0.189 -0.193 -0.194 -0.195 -0.200 -0.200

(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

TNA -0.021 -0.022 -0.019 -0.020

(0.270) (0.257) (0.314) (0.304)

ALPHA 0.008 0.024

x RSQ (0.548) (0.064)

49
 Table VIII

Style Consistency and Return Persistence: Logit Analysis

This table lists the average probability of producing above-median future performance given the manager’s cell location in a two-way classification involving past alpha and
style consistency. Cell cohorts are determined by the standard deviation rankings of ALPHA and RSQ within a manager’s peer group and tournament year (i.e., -2, -1, 0, +1,
and +2 standard deviations from median value). Panel A sets the value for the other explanatory variables equal to their standardized mean values of zero (i.e., TURN = 0,
EXPR = 0, TNA = 0). Panel B changes this base case by setting the value of EXPR to be two standard deviations below its mean. ALPHA and RSQ are estimated over a 36-
month period by the FFC version of equation (4) using 12-month future returns (11,804 observations).

Panel A. Probability of Being an Above-Median Manager, by ALPHA and RSQ Cohort (TURN = 0, EXPR = 0, TNA = 0)

RSQ:

Std. Dev. Group -2 (Low) -1 0 +1 +2 (High) (High – Low)

-2 (Low) 0.4467 0.4631 0.4796 0.4962 0.5127 0.0660

-1 0.4453 0.4678 0.4903 0.5129 0.5355 0.0902

ALPHA: 0 0.4440 0.4725 0.5010 0.5296 0.5580 0.1140

+1 0.4427 0.4771 0.5118 0.5463 0.5804 0.1377

+2 (High) 0.4414 0.4818 0.5225 0.5628 0.6024 0.1610

(High – Low) -0.0053 0.0187 0.0429 0.0666 0.0897

Panel B. Probability of Being an Above-Median Manager, by ALPHA and RSQ Cohort (TURN = 0, EXPR = -2, TNA = 0)

RSQ:

Std. Dev. Group -2 (Low) -1 0 +1 +2 (High) (High – Low)

-2 (Low) 0.5464 0.5628 0.5790 0.5951 0.6110 0.0646

-1 0.5451 0.5674 0.5895 0.6111 0.6324 0.0873

ALPHA: 0 0.5438 0.5720 0.5998 0.6269 0.6533 0.1095

+1 0.5425 0.5766 0.6100 0.6425 0.6736 0.1312

+2 (High) 0.5412 0.5812 0.6202 0.6577 0.6933 0.1522

(High – Low) -0.0053 0.0184 0.0412 0.0626 0.0824

50
 Fund A (R2 = 0.92): High Style Consistency Fund B (R2 = 0.78): Low Style Consistency

SP500 SP500
100 R1V R1 R1G 100 R1V R1 R1G

90 90
Cap: Small to Large (%)

Cap: Small to Large (%)

80 80

70 70

60 60

50 50

40 40
W4500 W4500
30 30

20 20

10 10

0 R2V R2 R2G 0 R2V R2 R2G

0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Value to Growth (%) Value to Growth (%)

Figure 1. Style Grids, R2, and Changes in Mutual Fund Style Over Time. This figures plots the relative investment style
positions for two portfolios and indicates how those positions have changed over time. Style positions and style consistency
(i.e., R2) were calculated relative to a variation of the multifactor style factor model in equation (4). Also plotted are the
investment style positions of several popular style and market benchmarks: Standard & Poor’s 500 (SP500), Russell 1000
(R1), Russell 2000 (R2), Russell 1000 Value and Growth (R1V and R1G), Russell 2000 Value and Growth (R2V and R2G),
and Wilshire 4500 (WIL4500).

51
Panel A. High Consistency vs. Low Consistency with Expense Control Samples
5.00
Hi RSQ: Lo EXPR

4.50

Lo EXPR
4.00
Hi EXPR

3.50
Growth of a $1

3.00
Lo RSQ: Hi EXPR

2.50

2.00

1.50

1.00
12

06

12

06

12

06

12

06

12

06

12

06

12

06

12

06

12

06

12

06
90

91

91

92

92

93

93

94

94

95

95

96

96

97

97

98

98

99

99

00
19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

20
Date

Panel B. High Consistency vs. Low Consistency with Expense and Alpha Control Samples
5.00
Hi RSQ: Lo EXPR: Hi ALPHA

4.50

4.00

Lo EXPR: Hi ALPHA
3.50 Hi EXPR: Lo ALPHA
Growth of a $1

3.00

2.50

Lo RSQ: Hi EXPR: Lo ALPHA

2.00

1.50

1.00
2

6
1

0
90

91

91

92

92

93

93

94

94

95

95

96

96

97

97

98

98

99

99

00
19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

20

Date

Figure 2. Cumulative Returns for Style Consistency-Sorted Portfolios, 1991-2000. This figures shows the
cumulative performance of one dollar investments in portfolios formed by dividing the fund sample by three
control factors: fund expense ratio (EXPR), past performance (ALPHA), and style consistency (RSQ). Panel A
illustrates the performance of two control portfolios based on the lowest (Lo EXPR) and highest (Hi EXPR)
expense quintiles and two portfolios additionally controlling for fund style consistency ((Hi RSQ, Lo EXPR)
and (Lo RSQ, Hi EXPR)). Panel B displays the performance of two control portfolios ((Lo EXPR, Hi ALPHA)
and (Hi EXPR, Lo ALPHA)) without regard to fund style consistency and two with ((Hi RSQ, Lo EXPR, Hi
ALPHA) and (Lo RSQ, Hi EXPR, Lo ALPHA)).

52