435 / 708
“Who cares what a fund charges? If a fund has great
returns, then I know it was able to overcome its fee.”
I see comments like this one just about every time
I write about the predictive power of fees. There’s a
certain logic to it, but the problem is one of persis-
tence. Fund expense ratios don’t tend to change much,
whereas performance changes a lot. I’ve shown
how fees predict success, but this time I wanted to
go one step further and look at the combination
of expenses and past performance. Would you rather
own a low-cost fund with poor past performance
or a high-cost fund with great past performance?
Novice investors would typically say the latter, but the
former is the better choice. Hands down.
I went back
10
years and grouped funds into expense-
ratio quartiles relative to category and trailing five-
year-return quartile by category. (I used the past five
years because that’s the time period investors seem
to rely on most.) Then I combined the data so that
I had fee quartiles grouped by past returns. That left
me with
16
groups of funds: the cheapest funds
divided into quartiles by return, the second-cheapest
funds in return quartiles, and so on. See the table
on Page
2
for the whole picture.
At the end of
2005
, the U.S. equity funds in the
cheapest quartile but with returns in the worst quar-
tile had an expense ratio of
0
.
73%
and an average
five-year return of negative
1
.
92%
annualized. The
funds in the priciest quartile but with top-quartile
performance had an expense ratio averaging
2
.
17%
,
but returns were a robust annualized
6
.
95%
. So,
how did the next
10
years go? The funds that started
with low costs and poor returns went on to enjoy
a
7
.
41%
annualized return compared with
5
.
46%
for
those formerly high-performing pricey funds. In
fact, the cheap laggards produced the highest returns
of any of the
16
groups, while the pricey sports
cars had the worst of any group.
The return figures actually understate the difference
because high-cost failures were more likely to be
liquidated by the parent company because of subse-
quent periods of poor performance. This is why
I use success ratios in conjunction with returns.
What happened? Low costs delivered the goods and
high costs hurt performance. We read too much
into past performance, but the data shows we should
pay more attention to expense ratios. Expenses are
a constant whose impact compounds over time. Jack
Bogle calls it the “relentless rules of humble arith-
metic.” In addition, a smaller part was played by the
way that markets rotate favor so that lagging sectors
and strategies recover while strong sectors and strate-
gies revert to the mean. Thus, those funds with
lagging five-year returns often came back to beat the
funds that won the previous round.
I’d also call your attention to the success ratio. That’s
a figure that tells you what percentage of a group of
funds survived and outperformed over the next period
in question. Here we see a
38%
success ratio for
the cheap laggards versus just
9%
for the pricey funds.
This tells me that the cheap funds not only outper-
formed but also were much more likely to survive than
the pricey return champions. Only
9%
of those seem-
The Battle of Costs
vs. Performance
Fund Reports
4
Harbor Capital Appreciation
Matthews Asian Growth & Income
PIMCO High Yield
Vanguard Tax-Managed Capital
Appreciation
Morningstar Research
8
Are Large Caps Overpriced?
The Contrarian
10
Cheap Medalists Due for a Rebound
Red Flags
11
Watch Out for Downgrades
Market Overview
12
Leaders & Laggards
13
Manager Changes and News
14
Portfolio Matters
16
The Usual Suspects Might Not
Help When Rates Rise
Tracking Morningstar
18
Analyst Ratings
Income Strategist
20
Funds That Use Everything but the
Kitchen Sink to Produce Income
Changes to the 500
22
FundInvestor 500 Spotlight
23
Follow Russ on Twitter
@RussKinnel
RusselKinnel, Director of
ManagerResearch and Editor
FundInvestor
October 2016
Vol. 25 No. 2
Research and recommendatio s for the s riou fund investo
SM
Continued on Page 2