馃攳
The Problem With Small Cap Stocks - YouTube
Channel: Ben Felix
[0]
- You probably won't be
surprised if I tell you
[2]
that small cap stocks in
general have outperformed
[5]
large cap stocks over the longterm,
[7]
even on a risk adjusted basis
[9]
with risk measured as standard deviation.
[12]
This is a well-documented
around the world and over time.
[15]
Small caps were the first
market anomaly that challenged
[18]
the capital asset pricing
model and resulted
[21]
in today's multi-factor
models for asset pricing.
[24]
I think that most investors accept this
[26]
and the more ambitious
DIY investors out there
[29]
might use small cap ETF's
[30]
to get exposure to the size
effect in their portfolios.
[34]
That sounds like a good
idea based on the data;
[36]
however, there is a big problem.
[39]
More recent research has shown
that there is no evidence to
[42]
support the existence of the size effect,
[45]
not now and not ever.
[47]
I'm Ben Felix, Associate
Portfolio Manager at PWL Capital.
[51]
In this episode of Common Sense Investing,
[53]
I'm going to tell you why
you should think carefully
[55]
about owning small cap ETF's.
[57]
(techno intro music)
[68]
Before I start the episode,
I want to tell you that
[70]
I've started a weekly podcast
called the Rational Reminder,
[73]
co-hosted by my colleague,
Cameron Passmore.
[76]
It's a reality check on sensible investing
[78]
and financial decision
making for Canadians.
[81]
If you enjoy my Common
Sense Investing videos,
[84]
I think that you will
really like the podcast.
[86]
I've put links to the Rational Reminder
[88]
in the description below.
[89]
Factor investing or smart beta is the
[92]
biggest trend in investing right now
[94]
and small stocks are one of
the most prominent factors.
[97]
Given its reputation, most would believe
[100]
that the small cap premium
is strong and robust.
[104]
The data tell a different
story, size is much weaker
[107]
than the other well-known
factors like value,
[110]
profitability, and momentum.
[112]
The size premium was first
documented by Rolf Banz in 1981.
[116]
Banz was a student of Eugene Fama.
[119]
He used the data from 1936 through 1975
[122]
when he initially
identified the size effect.
[125]
Recreating the Banz study over the
[127]
same time period with today's data shows
[129]
that the size premium is only
statistically significant
[132]
at the 10% significance threshold,
[134]
but not the more commonly
accepted 5% threshold.
[137]
In other words,
[139]
the size premium was not
statistically significant
[141]
from zero over the data period
[143]
that it was originally discovered.
[144]
Banz and most other researchers
use the crisp database
[148]
from the University of Chicago
to research security prices.
[152]
That data is continuously
being fixed and updated.
[155]
In 1997, Tyler Shumway's paper,
[158]
"The Delisting Bias in Crisp Data",
[160]
detailed how he corrected
the data for delistings,
[163]
which had not previously been captured.
[165]
Most of the delistings
were for negative events,
[168]
and mostly for smaller companies,
[170]
so the correction made the data
[172]
for small caps look much worse.
[174]
With the updated data for
U.S. stocks and expanding
[177]
the research to other countries,
[179]
there are many time periods
and geographic regions
[181]
where the size effect has been small
[183]
and statistically insignificant
[185]
since Banz published his paper.
[188]
Over longer periods of time,
[189]
like from 1927 through 2017 in the U.S.,
[193]
the size premium does become
statistically significant,
[196]
but compared to other well-known factors,
[198]
it's T-stat, or a measure
for statistical significance,
[201]
is low as is its premium.
[204]
Beyond the problems with the robustness
[206]
of the size factor, in general,
[208]
there are substantial issues
[209]
when it comes to implementation.
[211]
Much of the size effect that
we do observe is concentrated
[214]
in the smallest stocks in the market.
[217]
The smallest stocks are
notoriously expensive to trade
[220]
for a fund tracking index,
[222]
making building live index portfolios
[224]
that capture the size effect a challenge.
[226]
So far, small cap stocks in
general must sound pretty bad.
[231]
Fortunately, I do have some good news
[233]
which I will follow up
with some more bad news.
[236]
All of these issues with the size
[238]
effect apply to small
cap stocks as a whole,
[241]
but we know that size
is not the only factor.
[244]
In 1993 and again in 2015,
[247]
Eugene Fama and Kenneth
French documented that
[249]
small cap growth stocks failed
to deliver the size premium
[253]
and underperformed their
large cap growth counterparts.
[257]
More recently, it has also been documented
[259]
that the profitability
factor explains most
[262]
of the under performance
of small cap growth stocks.
[265]
It happens that most
small cap growth stocks
[268]
have low profitability, explaining
their under performance.
[272]
When we take small cap stocks
[273]
and sort them by relative price,
[275]
so value versus growth, and profitability,
[278]
so low versus high profitability,
[280]
we find that the performance
of small cap growth stocks
[283]
with low profitability are responsible
[286]
for the under performance
and statistical unreliability
[289]
of the asset class as a whole.
[291]
When we remove those stocks from the data,
[294]
the size premium comes back to life.
[297]
The premium becomes consistent over time
[299]
and across geographies.
[301]
It is no longer concentrated
in the smallest stocks,
[303]
it becomes larger in terms of
the premium that it delivers,
[306]
and it becomes statistically significant.
[309]
For example, from 1975 through 2017,
[313]
the universe of small cap stocks
[314]
in the U.S. returned 14.93% per year.
[318]
After removing the small growth
low profitability stocks,
[321]
the return increases to 16.5% per year.
[325]
All of this has been
documented by many researchers.
[328]
One of my favorite papers on the subject
[329]
has a title worth sharing:
[331]
"Size Matters If You Control Your Junk".
[334]
The paper refers to small cap growth
[336]
low profitability stocks as junk.
[339]
Hurray!
[340]
We saved the size effect,
[342]
but we are not yet in the clear.
[344]
Everything that we've talked
[345]
about so far has been about the data.
[348]
You do not care about the data,
[349]
unless it can be applied
to your portfolio.
[352]
Let's think about an ETF listed in Canada
[355]
that a Canadian investor might use to
[356]
get small cap exposure.
[358]
XSU from iShares fits that bill.
[361]
It tracks the Russell 2000 index
[363]
which is a cross section
of U.S. small cap stocks.
[366]
Based on the Morningstar
Style Box Analysis,
[369]
this ETF consists of
mostly small cap stocks
[372]
and some mid caps with the small caps
[374]
approximately equally split
across value, blend, and growth,
[379]
but with a bit of a tilt toward growth.
[381]
If you remember what we said
[382]
about small cap growth stocks dragging
[384]
down the returns of the
asset class as a whole,
[387]
this is not good news
for you as an investor.
[389]
How bad is it?
[391]
Let's look at the returns
[392]
of the Russell 2000 index over time
[394]
compared to a small cap index
from dimensional fund advisors
[397]
which removes the small cap growth
[399]
low profitability stocks.
[401]
From 1979 through 2017,
[404]
the Russell 2000 index returned
11.73% per year in USD,
[409]
while the dimensional U.S. small cap index
[411]
returned 14.37% per year.
[414]
Over that same period, the U.S. market
[416]
as measured by the Russell
3000 index returned 11.95%.
[421]
So the Russell 2000 trailed the market
[423]
before considering the higher fees
[425]
and costs that a small cap
index fund typically incurs,
[428]
not to mention the additional
risk of small cap stocks.
[431]
Again, picking on XSU,
[433]
it has an MER of 0.36%
compared to 0.07% for XUU,
[440]
which is iShares total U.S market ETF.
[443]
If XSU were delivering
[445]
on the higher expected
returns of small caps,
[447]
it might be worth the fee,
[449]
but without eliminating small
cap growth low profitability,
[452]
there is little value
in adding a relatively
[454]
expensive small cap ETF to your portfolio.
[457]
While we're here, we
may as well look at XCS,
[460]
which is the iShares
Canadian small cap fund
[462]
tracking the S&P/TSX small cap index.
[465]
Morningstar shows that it consists
[467]
of all small cap stocks with
a bit of a tilt toward value,
[470]
but still a sizeable amount in growth.
[472]
I only have data going
back to February 2000.
[475]
The S&P/TSX small cap index returned 5.03%
[479]
per year from February 2000 through 2017.
[482]
While the dimensional Canada
small cap index returned 9.97%,
[486]
and the S&P/TSX composite returned
[489]
6.28% per year on average.
[491]
Again, we see that without
the sort for growth
[494]
and profitability in small
caps, there is little point
[497]
in attempting to capture the size effect.
[499]
There are some U.S. listed
ETFs that might do a better job
[503]
of capturing the size effect.
[504]
For example, Vanguards VBR
is a small cap value fund
[508]
with no exposure to
small cap growth stocks.
[511]
While this is more attractive
than what we've seen so far,
[514]
it does still leave much to be desired.
[517]
To illustrate what I mean,
[518]
we can compare the Vanguard
small cap value ETF to the DFA
[522]
U.S. small cap value fund,
which is a U.S. mutual fund.
[527]
DFA is a fund company that specializes
[529]
in small cap value strategies.
[531]
We can compare the holdings
of these funds to get an idea
[534]
for the differences in
how they're structured.
[536]
VBR has an average market
cap as that September 2018
[540]
of 3.6 billion while
the DFA fund has a much
[543]
smaller average company
size at 1.8 billion.
[547]
This means that the companies
in the DFA fund are on average
[550]
about half the size of the
companies in the Vanguard fund.
[553]
VBR also has less overall value exposure
[556]
with a price to book ratio
[557]
of 1.87 compared to 1.32 for the DFA fund.
[562]
Lower price to book indicates
deeper value exposure.
[565]
Over the period spanning January
2000 through December 2017,
[570]
a period where small caps
generally performed well,
[573]
the DFA small cap fund
outperformed the Vanguard fund
[576]
by 54 basis points per
year on average after fees.
[580]
This is what we would expect
[582]
due to the smaller average company size
[584]
and deeper value exposure.
[587]
This does not make the
DFA fund the better.
[590]
Keep in mind that smaller companies
[591]
and deeper value means that
you were taking more risk.
[595]
The DFA fund is riskier,
[597]
but it is intentionally
capturing as much of
[600]
specific types of risks as possible
[602]
in order to increase expected returns.
[605]
This is what you would hope for
from a small cap value fund.
[608]
I'm telling you this
because a DIY investor
[610]
does not typically have
access to DFA funds.
[613]
If you do not have access to DFA,
[616]
you do not have the same level
of access to the size effect.
[619]
I questioned the efficacy
of using small cap ETFs,
[623]
even the half decent ones like VBR,
[625]
to effectively capture the size effect.
[627]
There is little question
that the size effect exists
[630]
in the data when we correct
[631]
for small cap growth low
profitability stocks.
[634]
However, finding a product that
[636]
effectively makes this correction
[638]
on implementation is
exceptionally difficult,
[641]
especially for a DIY investor
without access to DFA.
[646]
As I have said before,
[647]
most DIY investors are
likely to have the best
[650]
long-term investment
experience using low costs
[653]
broadly diversified
market cap weighted ETFs.
[656]
Have you tried to get exposure
[658]
to small cap stocks in your portfolio?
[660]
Tell me about it in the comments.
[662]
Thanks for watching.
[662]
My name is Ben Felix of PWO Capital
[664]
and this is Common Sense Investing.
[666]
I will be talking
[667]
about a new common sense
investing topic every two weeks,
[669]
so subscribe, click the bell for updates.
[672]
If you enjoy my Common Sense
Investing video series,
[675]
don't forget to check out
the Rational Reminder podcast
[678]
which is available on
Libsyn, iTunes, Google Play,
[680]
Spotify, and Stitcher. Links in the notes.
[684]
(techno outro music)
Most Recent Videos:
You can go back to the homepage right here: Homepage