8
Active share among U.S. equity funds has received
a lot of attention since Martijn Cremers and Antti
Petajisto published their first paper on the topic in
2009
. Their work quantified the degree to which
managers differentiate their funds from their bench-
marks, as active share measures the percentage
of a fund’s holdings that differ from the benchmark’s
weightings. This made it easier for investors to spot
closet indexers who charge active-management fees.
Cremers and Petajisto’s research also found that
sufficiently high active share is perhaps a necessary—
but not necessarily sufficient—precondition for
outperformance. Our tests have not found a link
with outperformance.
Yet, actively managed foreign-stock funds have so far
escaped this scrutiny. That’s partly because foreign-
equity data isn’t as reliable as domestic-stock data.
Because of the existence of multiple share classes
for many multinational companies, active share is
likely overstated to a small extent for many foreign-
stock funds. Take consumer staples giant
Unilever
.
Its shares trade as
ADR
s in the United States, as well
as in the Netherlands and in London. If the share
class held in a fund portfolio doesn’t match the one
represented in the index, the stock may count as an
active position relative to the benchmark.
This issue likely will be corrected as foreign-stock
databases become more robust. For the time being,
though, we’re treating this as a first look at foreign-
equity active share, sticking to broad themes with
details to be filled in later.
Common Traits With U.S. Cousins
As an initial inquiry, we looked at the actively man-
aged large-cap foreign-stock funds in the Morningstar
500
. This group includes
31
funds, with just more
than half of them in the foreign large-blend category.
We calculated their active share based on their most
recent portfolio and relative to the analyst-selected
benchmark for their respective category. What stood
out is how closely the results align, at least anecdot-
ally, with what we’ve seen so far among actively
managed U.S. stock funds.
In fact, some of the most active foreign-stock funds
share
DNA
with some of the bolder U.S. stock funds.
This certainly applies to foreign large-blend fund
Longleaf Partners International
LLINX
. It tied with
foreign large-blend
Janus Overseas
JAOSX
for our
group’s highest active share,
98
.
5%
. This is similar to
the results for large-blend sibling
Longleaf Partners
LLPFX
, which has one of its category’s highest active
shares relative to the S
&
P
500
Index,
96
.
0%
.
Such lofty differentiation owes to both funds’
extremely concentrated portfolios. Each holds just
20
stocks or so, with more than
50%
of assets in the
top-
10
holdings alone. For Longleaf Partners Interna-
tional, this includes insurance company
Fairfax
Financial Holdings
FFH
, which is run by Warren
Buffett acolyte Prem Watsa.
One other trait the siblings share is high tracking
error. Tracking error measures the volatility of a fund
that is not explained by movements in the benchmark.
Whereas active share notes differences with the
benchmark at the individual position level, tracking
error measures so-called factor bets, such as over-
weighting a given sector relative to the index or hold-
ing lots of cash. As a general principle, highly con-
centrated funds tend to have both high active share
and high tracking error. With
25%
currently in cash
and a nearly
40%
position in industrials, Longleaf
Partners International illustrates both.
Such bets give the fund a fairly high five-year tracking
error of
7
.
4%
relative to our group’s
5
.
5%
median.
But that’s not high enough to place among the group’s
boldest players. That distinction belongs to Janus
Overseas, with a
13
.
7%
five-year tracking error, and
First Eagle Overseas
’
SGOVX
10
.
2%
. With more
than
100
holdings, First Eagle isn’t nearly as concen-
trated as Longleaf, but its portfolio is just as eclectic.
Active Share for Foreign Funds
Morningstar Research
|
Kevin McDevitt
