![Show Menu](styles/mobile-menu.png)
![Page Background](./../common/page-substrates/page0766.png)
12
•
Fund Family Shareholder Association
www.adviseronline.comDO YOU REALLY WANT TO
hand
your portfolio off to a machine? Think
really hard about that.
Much is being made these days
of the rise of the robo-advisers, those
automated and semi-automated finan-
cial advice and portfolio manage-
ment services that Charles Schwab,
Betterment, Wealthfront and Vanguard,
among many, are now offering.
Vanguard, in particular, has been
transitioning billions of dollars of
investor assets from passive alle-
giance to its funds, to paid clients of its
Personal Advisor Services
group. No
doubt, you’ve noticed that Vanguard
is promoting its service ceaselessly.
Of course, what would you expect?
With something above $20 billion in
the service, they’re pulling in north of
$60 million in fees. That pays for a lot
of country club dues for Vanguard’s
executive class on top of the salaries for
its advisers.
I’m hoping, however, that Vanguard
will take some of that money and spend
it on their robos rather than sending it
to the C-suite, because if my experi-
ence with just one of their services,
the Vanguard Portfolio Analytics Tool,
is any indication, the computers run-
ning the show over there are in need
of some serious human intervention.
(And those salaried advisers Vanguard
has hired to help you aren’t the ones to
do it, because they’re liable to get their
hands slapped, or much worse, if they
take a position that doesn’t mesh with
Vanguard’s.)
Here’s my story, which started out
very innocently as an exercise in com-
paring my
Growth Model Portfolio
to
one of Vanguard’s “core” ETF port-
folios. To be clear, I wasn’t using the
Personal Advisor Services
, but rather,
a tool Vanguard offers on its website.
You may not know it, but Vanguard
offers up model portfolios made up of
ETFs in 11 different allocations rang-
ing from 100% stocks to 100% fixed
income, with incremental changes of
10%, so there’s a 90%/10% model,
an 80%/20% model and so on. These
model portfolios use Vanguard’s “core”
ETFs, Russell index ETFs, S&P index
ETFs and CRSP index ETFs for the
equity side of the ledgers. That’s 44 dif-
ferent model portfolios to choose from.
I figured I’d just test-drive
Vanguard’s Portfolio Analytics Tool
based on its most basic and most heav-
ily recommended ETFs used in the
“core” portfolios. The Vanguard Core
100/0 portfolio allocates 68.6% of
assets to
Total Stock Market ETF
,
29.4% to
Total International Stock
ETF
and 2.0% to a money market fund.
Essentially it’s 100% stocks. With the
Growth Model Portfolio
holding just
3% or so of its assets in my cash substi-
tute,
Short-Term Investment-Grade
,
this seemed the appropriate pairing.
So, I punched in the ticker symbols
and allocations for the
Growth Model
Portfolio
and selected Vanguard’s 100/0
portfolio for a comparison. It couldn’t
have been simpler.
But the output was anything but
simple. And much of what I got was
completely random. On top of which,
the report that appears on your screen
is not the report that appears when you
ask for a downloadable PDF document.
Let me give you just a few examples of
the silliness that is this “tool.”
Let’s start at the top. Of course
there’s the obligatory comparison of
expenses, 0.09% versus 0.36%, and pie
charts showing allocations. And then
there’s a line graph comparing the past
three years or so of performance for the
ETF portfolio and my
Model Portfolio
.
I say “three years or so” because the
graph ran for a 37-month period, which
seems rather random. Anyway, I was
thrilled to see that the computer showed
the performance of the
Model Portfolio
was 18.75% per annum versus 14.23%
for the Vanguard portfolio.
The
Model Portfolio
runs a little
more expensive on measures like price/
sales and price/book, etc. and has a
smaller allocation to emerging markets
than Vanguard’s portfolio. Nothing out
of the ordinary or surprising. I did
find the Modern Portfolio Theory stats
showing a negative alpha of -0.91 for
Vanguard’s portfolio versus a positive
2.91 for my
Growth Model Portfolio
’s
a bit strange. Partly it’s because the two
portfolios are compared to slightly dif-
ferent benchmarks, which you wouldn’t
know if you didn’t click on the “bench-
mark” link at the bottom of the table.
Vanguard’s index funds actually “lose”
alpha to straight indexes because of
their expenses. The portfolio managers
in the
Growth Model Portfolio
, how-
ever, add alpha, or in simple terms, they
add stock-picking smarts.
Then there’s the diagram show-
ing how evenly distributed the port-
folio’s “stock sectors” are allocated.
The Vanguard ETF model was actu-
ally slightly askew with an overweight
to cyclical stocks, while the
Growth
Model Portfolio
was spot-on evenly
allocated.
Now, here’s where things begin to
ROBO ADVICE
Rise of the Machines, and of Errors
The Seven Questions
VANGUARD ALSO OFFERS a simple, “7-question tool” that they say will give you a “custom invest-
ment plan.” I tried that one as well. As an aggressive investor with a long time horizon, I was eager
to see how they’d allocate me, given that they say, up front, that the only funds they recommend
are
Total Stock Market Index
,
Total International Stock Index
,
Total Bond Market Index
,
Total International Bond Index
, the
STAR
LifeStrategy
funds and
Prime Money Market
.
Surprise: My recommended plan (remember, this is a custom plan) was to allocate 60% of my
money to Total Stock Market Index and 40% to Total International Stock Index. Let’s just say I
didn’t click on the “Invest Now” button.