2
had a
53%
success rate—that might not sound really
high, but when you consider how many funds get
merged away, that figure is actually quite high
and compares favorably with a
38%
success rate for
the lowest-turnover quintile in U.S. equity funds.
The success rate declined steadily with each quintile
on down to
26%
for the quintile with the highest
brokerage costs compared with
31%
for turnover.
Thus, a low brokerage-cost fund was twice as
likely to succeed as a high brokerage-cost fund. By
contrast, using turnover barely helped at all.
Outside of domestic-equity funds, brokerage costs
didn’t move the needle very much. In balanced funds,
the bottom quintile lagged while the other quintiles
were close together. In international-equity and
sector funds, it didn’t move the needle. It didn’t show
much impact in bonds, but that’s as expected as bond
trading and commissions work differently than stocks.
While brokerage commissions worked quite well
in this test, I wouldn’t put them in the top echelon of
pre-dictive value with expense ratios, Morningstar
Ratings for funds, and manager records because we
have fewer available time periods to test.
Why Brokerage Costs Might Link to
Trading Costs
So, if brokerage costs are good signals for the much
larger piece of the pie known as trading costs, let’s
look at why that might be the case. One reason is that
brokerage costs are kind of like tolls for trading. If
you pay a lot of tolls, then it stands to reason you are
doing a lot of driving.
Greg Kadlec, Roger Edelen, and Richard Evans hit
upon another reason for the link in their study, “Scale
effects in mutual fund performance: The role of
trading costs.” In the disclosure of brokerage commis-
sions, funds must reveal how much went to soft
dollars. Soft dollars are an amount over and above
the basic commission that funds pay in exchange
for research. They could be paying for research on a
stock or macroeconomic trends or for access to
management or focus groups set up specifically for
their benefit, such as a group of doctors discussing
their views of a new drug.
Brandywine
BRWIX
even
used soft dollars to pay for a part-time analyst.
Kadlec, Edelen, and Evans found that performance
was worse (implying higher trading costs) where
funds used soft dollars.
Presumably, the reason is that asset managers nego-
tiate soft-dollar deals before they make the trades
and therefore may have to trade more than they would
otherwise just to meet the soft-dollar targets. A
trade designed purely to hit soft dollars is of course
not likely to improve returns.
Finding Funds With Low Brokerage Costs
As mentioned, you can find all the Morningstar
500
funds’ brokerage costs at
mfi.morningstar.com
. I’ll
call out a few below and share some basic principles
on the types of funds that tend to have low
brokerage costs.
Nearly all the index funds we tested had low
brokerage costs. In fact, they tend to have quite low
trading costs. We can infer this from the fact that
most index funds lag their benchmarks only by their
expense ratio. If they had significant trading costs,
they would lag by more. There have been cases with
higher-turnover indexes where the funds did, in fact,
lag by a wider margin.
So, let’s look at some of the actively managed funds
that charge low brokerage costs.
Morgan Stanley Focus Growth
AMOAX
, run by
Dennis Lynch, had one of the lowest figures in
our database at just one fifth of
1
basis point. This
makes some sense as Lynch runs a low-turnover,
large-cap-focused portfolio.
Jensen Quality Growth
JENSX
also has low
brokerage costs. It, too, is a focused, large-cap fund
with low turnover—ideal for low trading costs.
The same can be said for
Dreyfus Appreciation
DGAGX
,
Mairs & Power Growth
MPGFX
, and
Sequoia
SEQUX
, which also have brokerage costs of
around
1
basis point a year.
Harbor Capital Appre
ciation
HACAX
is also in that group, though it trades
a bit more with a
41%
turnover rate.
Brokerage Costs Signal Funds’ True Cost Hurdle
Continued From Cover
