FOCUS TOPIC B R I E F

APRIL 2014

SWAPPING DOGS FOR LIONS:

A PRIMER ON ACTIVE SHARE
CHERYL A. BARKER /

Senior Research Analyst, Global Equities

This focus topic brief is an excerpt from the Fund

Evaluation Group, LLC February 2014 Research Review.
To view the full newsletter, or to be added to our mailing
list, please visit www.feg.com/research.php.

Zoo in China Swaps Dog for Lion, Hopes No One Notices, a headline reads. Ocials at the Louhe city
zoo in the central Henan province hoped no one would notice when they decided to make the switch in
order to send the exhibits regular resident, an African lion, away to a breeding center. Visitors got a rude
surprise, however, when the lion started barking.1
In fairness to the zookeeper, the Tibetan masti is a large, exceptionally furry breed of dog. Many would
conclude that the breed resembles the king of the jungle more closely than perhaps any other domestic
canine. While the switch undoubtedly fooled some patrons, a six-year-old who had recently learned the
various sounds dierent animals make quickly identified that the lion was indeed barking like a dog.
A spokesperson for the zoo cited that Chinese zoos have struggled to make money in recent years due to
a government ruling that animal shows should operate on a non-profit basis.
Regardless of motive or resemblance, a dog was in the lion exhibit and patrons of the zoo were paying to
see a lion. Its no surprise that zoo patrons felt cheated.
APPR OV ED FO R CL I E N T USE .

O ri g i n a l l y p u b l ish e d Fe b r u a r y 2014
2 014 Fu n d E v a l u a t i o n G r o u p , L LC

F O C U S T O P I C B R I E F / A P R I L 2 014

Close enough? A Tibetan mastiff was placed in the African

Lion exhibit at a zoo in Chinas central Henan province.
Source: Ed Jones/AFP/Getty Images/NPR

As the active management versus passive management battle continues to rage, a similar, but less humorous, phenomenon has been observed in the investment world: closet indexing. For decades, active
managers have charged higher management fees for
their strategies than their passively managed competitors. In some cases, the higher fees are understandable, as the active managers apply unique skill,
deep fundamental research, and seek to outperform
a passively managed benchmark. But what happens
when an investment manager claiming to be an active
manager is really a closet indexer? Investors in those
funds are paying active management fees, but are receiving benchmark-like performance. Thus, the investors portfolio has little chance of outperforming the
benchmark, as the higher fees often counteract any
positive performance achieved. Those investors are
paying to view a lion, but seeing a dog instead.

How can investors identify the dogs of the industry? In 2004, Antti Petajisto and Martijn Cremers, then
colleagues in the finance department at the Yale School of Management, pondered that same question.
They were intrigued by a Wall Street Journal article about the Fidelity Magellan Fund, which alleged that
the funds manager at the time, Robert Stansky, was a closet indexer because the funds returns closely
tracked those of the Standard & Poors 500 Index. Knowing there were multiple factors that could lead to
benchmark-like returns and that there was a plethora of industry discussion surrounding closet indexing,
Petajisto and Cremers decided they needed to look under the hood to substantiate the closet indexing
claim with real data.
The duo teamed up to discern how to dierentiate true stock pickers from closet indexers. They completed a first draft of their research in 2006, and in 2009 published a paper, How Active Is Your Fund Manager?
A New Measure That Predicts Performance. That measure, which they named active share, is the percentage of a funds weight-adjusted portfolio that diers from its benchmark.2
We have used the active share metric in manager analysis for a number of years at FEG and believe many
will pay more attention to this measure and that additional follow-up studies will likely be published.
Therefore, this focus topic provides an overview of active share. On one end of the spectrum (assuming no
leverage or short positions), an active share percentage of 0% suggests a portfolios holdings are identical
to the benchmark; while an active share percentage of 100% indicates a portfolio is completely dierent
from the benchmark without any crossover of holdings.

Where wfund,i and windex,i are the portfolio weights of asset i in the fund
and in the index, and the sum is taken over the universe of all assets.
Source: Petajisto and Cremers, 2009 study

Petajisto and Cremers are the first to admit their measure is straightforward, and not exactly rocket science. Like the six-year-old who identified the lion at the zoo was barking like a dog, the simplicity of active share is what gives it strength. At a base level, active share allows investors to easily identify managers
who piggyback the index but charge high fees. Digging deeper, however, active share becomes a useful
tool that can tell much more.
PAG E 2

2 014 Fu n d E v a l u a t i o n G r o u p , L LC

F O C U S T O P I C B R I E F / A P R I L 2 014

P E R F O R M A N C E F O R A L L- E Q U I T Y
M U T UA L F U N D S
19 9 0 -2 0 0 3
Gross
High
ActiveShareQuintile

Because a manager can only add value relative to

a benchmark by deviating from it, active share first
provides information about a funds potential for
outperforming its benchmark. Though active share
is not a sucient factor to explain outperformance
on its own, some level of active management is necessary for a manager to outperform its benchmark.
The degree and type of active management, however, matters considerably for performance. The original study incorporated mutual fund data from 1980
through 2003; however, the performance analysis
within the study used a sample period of 19902003. Almost all funds were active in the 1980s, but
starting around 1990, a meaningful mass of active
funds began having a modest active share of 60% or
less. The study illustrated that funds with the highest active share significantly outperformed their
benchmarks net of fees and their returns persisted
from year to year. Funds with lower active share underperformed after fees.

Net
2.40%

1.13%

1.33%

0.25%

0.81%

0.75%
0.24%

2 1.37%

0.11%

Low 1.42%
2%

1%

0%

1%

2%

Data source: Petajisto and Cremers, 2009 study

Returns are presented gross and net of fees in U.S. dollars.

Tracking error has been the traditional metric used to measure active management, but tracking error
fails to capture the two dierent dimensions of active management: stock selection and factor bets. Additionally, there are dierent approaches to active management that can contribute dierently to tracking
error. When active share is used together with tracking error, it provides a more comprehensive picture.

Active Share: the fraction of portfolio holdings that dier from the benchmark;
thus a measure that describes a managers stock selection bets.
Tracking Error: the volatility of active return relative to the benchmark: thus, a
measure that describes how closely a manager tracks its benchmark.
Data Source: Cremers + Petajisto and Zephyr Associates, Inc.

Active share is a holdings-based measure that describes a managers current stock selection bets. Tracking
error is a returns-based measure of the volatility of past active returns relative to the benchmark. When
used together, they provide a comprehensive picture of how active a fund is on the dimensions of both
holdings and returns. When plotting investment managers by their active share and tracking error metrics,
relative categories for their portfolio management approach can be assigned.

Types of Active Management

Petajisto and Cremers findings include pure indexers plotted in the bottom, left-hand corner of the following chart, with less than 20% active share and low tracking error as the funds attempt to closely mimic
the benchmark against which they are indexed. A diversified stock picker can have high active share despite its low tracking error, because its stock selection within industries can lead to deviations from the
benchmark. In contrast, a manager betting on factors, such as value or quality, can generate a large tracking error relative to peers even without large deviations from benchmark holdings. Taking active positions
in individual stocks and in factors, concentrated stock pickers combine the two approaches and generally
PAG E 3

2 014 Fu n d E v a l u a t i o n G r o u p , L LC

3%

F O C U S T O P I C B R I E F / A P R I L 2 014

have greater than 80% active share. Closet indexers score low on both dimensions of active management
while claiming to be active managers. They generally exhibit active share levels in the 20% to 60% range.3
The specific tracking error levels classified as high and low vary by asset class across the manager categories.

ActiveShare

High
Diversified
StockPicks

Concentrated
StockPicks

Closet
Indexing

Factor
Bets

Low
Pure
0 Indexing
0
Low

High
TrackingError

Source: Petajisto and Cremers, 2009 study

E VO LUTI O N O F AC TIVE SHAR E

19 8 0 -2 0 0 9

In terms of performance, concentrated stock

pickers (high active share, high tracking error)
were the best performers each year and across
the sample period, followed by diversified stock
pickers (high active share, low tracking error).
Closet indexers tended to underperform their
benchmarks after fees and transaction costs.
Low active share managersboth pure and
closet indexershave become increasingly
prevalent in the U.S. equity fund universe
over the past 30 years. Prior to the 1990s,
most mutual fund assets were truly active.
By 2009, funds classified as highly active,
with 80% or higher active share, represented
only approximately one-fourth of U.S. equity
mutual fund assets under management. Pure
indexers, with active share less than 20%, have
grown from almost nothing in 1980 to one-fifth
of mutual fund assets at the end of 2009. This
illustrates the growth of passive investment
over the past two decades. Closet indexing,
however, has become even more popular than
pure indexing, accounting for approximately
one-third of all mutual fund assets at the end
of 2009.4

The reason an increasing number of managers

hug their benchmark varies. For some, fund
size comes into play. As funds surpass $1 billion
in assets, statistically speaking, it is harder to
maintain high levels of active share. Looking
Source: Petajisto, 2010 study
across the market capitalization spectrum,
additional challenges are presented. Small and
mid cap funds tend to have higher levels of active share than their large cap counterparts. This is due not
only to the larger number of companies in the smaller cap benchmarks, but also to the fact that smaller
cap benchmarks tend to be relatively flat, while large cap benchmarks tend to be top-heavy, with the
largest companies representing a disproportionate amount of the benchmark. The top 50 holdings in the
small cap Russell 2000 Index represented just over 11% of the Index as of February 28, 2014, whereas
the top 50 holdings in the large cap Russell 1000 Index represented over 41% of the Index.5 Large cap
managers who want to purchase any of the largest market cap stocks in their benchmark will subsequently
see their active share tumble. For others, closet indexing is an appealing prospect, especially when market
volatility increases. Closet indexers have little chance of meaningful outperformance, but they are unlikely
to perform so poorly that they will be firedand they will collect active management fees. The growth of
defined contribution plans is linked to another structural reason for the decline in active share, as plan sponsors
sometimes attempt to manage the risk of underperforming the benchmark to help mitigate the potential for
litigation from plan participants. This can pressure fund managers to ensure benchmark-like returns.
PAG E 4

2 014 Fu n d E v a l u a t i o n G r o u p , L LC

F O C U S T O P I C B R I E F / A P R I L 2 014

Conclusion
We believe that opportunities exist for managers with unique strategies and competencies to add value
to a portfolio, beyond that of a purely passive representation of the market. Though active share is not
a tool that can predict returns in an absolute sense, research has demonstrated that managers with
high active share have greater opportunities for outperformance. As one would expect, deviations from
the benchmark can introduce risk, so manager selection is critical in the actively managed portion of
an investors portfolio. Given the cost premium assigned to actively managed strategies, investors must
understand the fees they are paying and how they relate to the amount of active management they
receive. Active share provides an indication of a managers conviction level and willingness to manage
beyond a benchmark, but may not be indicative of a managers stock picking skill. Simply being dierent
from the benchmark is not enough to outperform; a manager needs to be dierent in the right ways. We
recommend allocating to equity managers who exhibit exceptional skill through a repeatable process,
and show strength across FEGs six tenets for manager analysis: conviction, consistency, pragmatism,
investment culture, risk management, and active return. Though the overall tenets are qualitative in
nature, there are many quantitative measureslike active share and tracking errorthat continue to be
valuable components of our manager due diligence process and help our clients invest with lions rather
than dogs.

1
2
3

4
5

Neuman, Scott. Zoo in China Swaps Dog for Lion, Hopes No One Notices. The Two-Way. National Public Radio, 15 Aug. 2013.
Max, Sarah. Is Your Fund Manager Active Enough? Barrons. 12 Jan. 2013. Online reprint accessed 14 Jan. 2013.
Cremers, Martijn K. J., Antti Petajisto. How Active Is Your Fund Manager? A New Measure That Predicts Performance. International Center for
Finance Yale School of Management, 2009.
Petajisto, Antti. Active Share and Mutual Fund Performance. NYU Stern School of Business, 2010.
Russell

PAG E 5

2 014 Fu n d E v a l u a t i o n G r o u p , L LC

F O C U S T O P I C B R I E F / A P R I L 2 014

D I S C LO S U R E S
This report was prepared by Fund Evaluation Group, LLC (FEG), a federally registered investment adviser under the Investment Advisers Act
of 1940, as amended, providing non-discretionary and discretionary investment advice to its clients on an individual basis. Registration as an
investment adviser does not imply a certain level of skill or training. The oral and written communications of an adviser provide you with information about which you determine to hire or retain an adviser. Fund Evaluation Group, LLC, Form ADV Part 2A & 2B can be obtained by written
request directly to: Fund Evaluation Group, LLC, 201 East Fifth Street, Suite 1600, Cincinnati, OH 45202, Attention: Compliance Department.
The information herein was obtained from various sources. FEG does not guarantee the accuracy or completeness of such information provided
by third parties. The information in this report is given as of the date indicated and believed to be reliable. FEG assumes no obligation to update
this information, or to advise on further developments relating to it. FEG, its affiliates, directors, officers, employees, employee benefit programs
and client accounts may have a long position in any securities of issuers discussed in this report.
Index performance results do not represent any managed portfolio returns. An investor cannot invest directly in a presented index, as an investment vehicle replicating an index would be required. An index does not charge management fees or brokerage expenses, and no such fees or
expenses were deducted from the performance shown.
Neither the information nor any opinion expressed in this report constitutes an offer, or an invitation to make an offer, to buy or sell any securities.
Any return expectations provided are not intended as, and must not be regarded as, a representation, warranty or predication that the investment will achieve any particular rate of return over any particular time period or that investors will not incur losses.
Past performance is not indicative of future results.
Investments in private funds are speculative, involve a high degree of risk, and are designed for sophisticated investors.
This report is prepared for informational purposes only. It does not address specific investment objectives, or the financial situation and the
particular needs of any person who may receive this report.
All data is as of February 28, 2014 unless otherwise noted.

PAG E 6

2 014 Fu n d E v a l u a t i o n G r o u p , L LC

F O C U S T O P I C B R I E F / A P R I L 2 014

INDICES
Barclays Capital Fixed Income Indices is an index family comprised of the Barclays Capital Aggregate Index, Government/Corporate Bond Index,
Mortgage-Backed Securities Index, and Asset-Backed Securities Index, Municipal Index, High-Yield Index, and others designed to represent the
broad fixed income markets and sectors within constraints of maturity and minimum outstanding par value. See https://ecommerce.barcap.
com/indices/index.dxml for more information.
The CSFB Leveraged Loan Index is designed to mirror the investible universe of the $US-denominated leveraged loan market. Loans are added
to the index upon issuance if they qualify according to certain criteria.. See www.credit-suisse.com for more information.
The Dow Jones-UBS Commodity IndicesSM are composed of exchange-traded commodity futures contracts rather than physical commodities. By
tracking commodity futures rather than commodity spot prices (meaning the prices quoted for immediate payment and delivery of physical
commodities), the indexes are investable benchmarks, meaning they can generally be replicated using futures contracts. The weightings for
each commodity included in DJ-AIGCI are calculated in accordance with rules that ensure that the relative proportion of each of the underlying
individual commodities reflects its global economic significance and market liquidity.
The FTSE NAREIT Composite Index (NAREIT Index) includes only those companies that meet minimum size, liquidity and free float criteria as set
forth by FTSE and is meant as a broad representation of publicly traded REIT securities in the U.S. Relevant real estate activities are defined as the
ownership, disposure, and development of income-producing real estate. See www.ftse.com/Indices for more information.
The HFRI Monthly Indices (HFRI) are equally weighted performance indexes, compiled by Hedge Fund Research Inc. (HFX), and are used by
numerous hedge fund managers as a benchmark for their own hedge funds. The HFRI are broken down into 37 different categories by strategy,
including the HFRI Fund Weighted Composite, which accounts for over 2000 funds listed on the internal HFR Database. The HFRI Fund of Funds
Composite Index is an equal weighted, net of fee, index composed of approximately 800 fund- of- funds which report to HFR. See www.hedgefundresearch.com for more information on index construction.
J.P. Morgans Global Index Research group produces proprietary index products that track emerging markets, government debt, and corporate
debt asset classes. Some of these indices include the JPMorgan Emerging Market Bond Plus Index, JPMorgan Emerging Market Local Plus Index,
JPMorgan Global Bond Non-US Index and JPMorgan Global Bond Non-US Index. See www.jpmorgan.com for more information.
The London Interbank Offered Rate (LIBOR) is a daily reference rate based on the interest rates at which banks borrow unsecured funds from
other banks in the London wholesale money market.
The Merrill Lynch high yield indices measure the performance of securities that pay interest in cash and have a credit rating of below investment grade. Merrill Lynch uses a composite of Fitch Ratings, Moodys and Standard and Poors credit ratings in selecting bonds for these indices.
These ratings measure the risk that the bond issuer will fail to pay interest or to repay principal in full. See www.ml.com for more information.
Morgan Stanley Capital International MSCI is a series of indices constructed by Morgan Stanley to help institutional investors benchmark
their returns. There are a wide range of indices created by Morgan Stanley covering a multitude of developed and emerging economies and
economic sectors. See www.morganstanley.com for more information.
Russell Investments rank U.S. common stocks from largest to smallest market capitalization at each annual reconstitution period (May 31). The
primary Russell Indices are defined as follows: 1) the top 3,000 stocks become the Russell 3000 Index, 2) the largest 1,000 stocks become the
Russell 1000 Index, 3) the smallest 800 stocks in the Russell 1000 Index become the Russell Midcap index, 4) the next 2,000 stocks become the
Russell 2000 Index, 5) the smallest 1,000 in the Russell 2000 Index plus the next smallest 1,000 comprise the Russell Microcap Index. See www.
russell.com for more information.
S&P 500 Index consists of 500 stocks chosen for market size, liquidity and industry group representation, among other factors by the S&P Index
Committee, which is a team of analysts and economists at Standard and Poors. The S&P 500 is a market-value weighted index, which means
each stocks weight in the index is proportionate to its market value and is designed to be a leading indicator of U.S. equities, and meant to
reflect the risk/return characteristics of the large cap universe. See www.standardandpoors.com for more information.
The S&P Global Property Indices define and measure the investable universe of publicly traded property companies. With more than 450 constituents from more than 35 countries, the index is ideal for a range of investment activities, including benchmarking active funds and setting
the foundation for passive funds. See www.standardandpoors.com for more information.
Information on any indices mentioned can be obtained either through your consultant or by written request to information@feg.com.

R E S - 4 027 4 -24 -2014

PAG E 7

2 014 Fu n d E v a l u a t i o n G r o u p , L LC

F O C U S T O P I C B R I E F / A P R I L 2 014

Research Team
CHERYL A. BARKER / Senior Research Analyst / Global Equities
NOLAN M. BEAN, CFA, CAIA / Managing Principal / Investment Philosophy and Portfolio Construction
KEITH M. BERLIN / Senior Vice President / Director of Global Fixed Income and Credit
CHRISTIAN S. BUSKEN / Senior Vice President / Director of Real Assets
KEVIN J. CONROY, CFA, CAIA / Senior Research Analyst / Hedged Strategies
GREGORY M. DOWLING, CFA, CAIA / Managing Principal / Director of Hedged Strategies
SUSAN MAHAN FASIG, CFA / Managing Principal / Director of Private Capital
JEFFREY D. FURST, CFA, CAIA / Senior Research Analyst / Hedged Strategies
BRIAN A. HOOPER / Senior Research Analyst / Global Equities
GREGORY D. HOUSER, CFA / Senior Vice President / Capital Markets
JAY R. JOHNSTON / Research Analyst / Real Assets
PETER R. KISTINGER / Research Analyst / Real Assets
MARK KOENIG, CFA / Senior Vice President / Director of Quantitative Analysis
J. ALAN LENAHAN, CFA, CAIA / Managing Principal / Director of Hedged Strategies
DAVID L. MASON, CAIA / Senior Research Analyst / Hedged Strategies
CHRISTOPHER M. MEYER, CFA / Managing Principal / Chief Investment Officer
PAUL J. NEUMANN / Research Analyst / Hedged Strategies
MICHAEL J. OCONNOR, CFA / Senior Research Analyst / Global Fixed Income and Credit
MICHAEL J. OYSTER, CFA / Managing Principal / Portfolio Strategist
WILLIAM B. PHELPS, CAIA / Senior Fund Operations Analyst / Managed Solutions
GARY R. PRICE / Managing Principal / Head of FEG Managed Portfolios
SAMUEL A. RAGAN / Research Analyst / Global Equities
G. SCOTT TABOR / Senior Research Analyst / Private Capital
DANIEL I. TIRPACK / Research Analyst / Private Capital
NATHAN C. WERNER, CFA, CAIA / Senior Vice President / Private Capital
RYAN S. WHEELER, CAIA / Director of Fund Operations / Managed Solutions

Fund Evaluation Group, LLC / 201 Eas t Fif th St. / Suite 160 0 / Cincinnati, O hio 45202 / P: 513.977.4 4 0 0 / F: 513.977.4 430 / w w w.fe g.com

PAG E 8

2 014 Fu n d E v a l u a t i o n G r o u p , L LC