Why primary research is overrated.
Charlie Munger would be proud.
Askeladden Capital
–
Primary
Research Whitep
aper
–
August
2016
Questions/thoughts/comments? thehappyanalyst@askeladdencapital.com
1
Why primary research is overrated.
Samir Patel
–
Founder/
PM,
Askeladden Capital
2016
-
08
-
23
Heuristics
heuristics heuristics
: we use them because we have to. We can
’
t reasonably start from first
premises every
time we expres
s a thought or ask a question
; we
’
d never get anywhere
. Most of the time,
we don
’
t have to worry about how exactly
the internal combustion engine underneath the hood
functions
or
how
Google
’
s search
algorithm
gets us
closer to what we want t
o know; we can skip all
those mental steps
that (usually)
don
’
t matter
, just ass
ume they
’
re true and
logical
and our car will start
when we turn t
he key
,
leaving us free to focus
more on the
mental processes
that (usually) do
make a
difference
.
Except
when we go too long without asking those questions, we
often end up with nonsensical results
that everyone agrees to be true. F
o
r
example: in the age of social and mobile and cloud, why the heck
are we still working
at cramped desks
in
expensive
co
-
located
urban
-
core office space that it
’
s societally
ineffi
cient and expensive to maintain?
One of the core
assumptions
underlying modern investing is
the notion that
you have to have an analytical
edge
. Indisputable, of course, which is exactly why I
’
m about to dispute it: even when
I probably have an
analytical edge, I
’
m just gonna assume that I don
’
t, and live with that.
Surely you
’
ve
heard one or
more versions of the following:
If it
’
s in the news, it
’
s in the price.
What
do you see that the market doesn
’
t?
Do you have a view on next quarter
’
s comps?
These are all reasonable quest
ions to ask, if you accept the premise that an analytical edge is what leads
to outperformance, ex
cept: that premise doesn
’
t look very true.
Primary research is the flavor du jour of the investing world: it started with extensive
write
-
ups
on VIC
and SumZero and Seeking Alpha, cascading to its inevitable conclusion when Bill Ackman spent
countless hour
s and who knows how many million dollars finding out literally everything one could
possibly know about the busine
ss oper
ations of pyramid scheme (I mean, sorry, multi
-
level
-
marketing
endeavor) Herbalife. And it
’
s funny, we a
ll knew that
Herbalife was a scummy
organization whose
fundamental business model was
“
let
’
s see
how many people can we rip off
before lunch
go go go
,
”
but there
’
s a big
difference
between
being an unethical business and being a legal pyramid scheme. And all of
the
primary research in the world, most of which just confirmed what all of us could guess with 95%
confidence just from having
eyes and ears and
functioning gray matter, couldn
’
t save Ackman from
whiffing on the core question under
lying the investment thesis:
will the
FTC put Herbalife out of business.
This isn
’
t about Herbalife, though, nor is it about Valeant (where
Ackman
’
s
primary research also
missed the core question
about whether PBMs
and payors
could be
suckered
into the sunset
), or
Micron
a
nd Fannie Mae (hi Einhorn!). Nor is it about an utterly fantastic write
-
up I read by a moderately
well
-
known New York fund, in
which they
interviewed dealers in Brazil and scoured the ends of the earth to
Askeladden Capital
–
Primary
Research Whitep
aper
–
August
2016
Questions/thoughts/comments? thehappyanalyst@askeladdencapital.com
2
determine that
one of the two duopolistic
providers of center
-
pivot irrigation was an absolute gem of a
company through
and through. And they were right, on all of that
, but still got blown up on their
thesis, because they whiffed on the one question that was really important:
the right starting point to use
for valuation in an incredibly cyclical industry (ag), and
the right long
-
term growth rate. I spent all of a
few hours reviewing annual reports from the 1990s, comparing them to today,
and determining that
aforementioned fund
’
s pr
ojections were wildly, wildly optimistic. Notwithstanding that they
“
knew
”
the
company
in absurdly more depth than I did, and I had no analytical edge whatsoever
–
I
was literally just
reading
the
publicly
-
available information that is
everyone
’
s starting point
–
I
quickly d
etermined that
not only was that fund overly optimistic, but so was the market.
And then over here on the other side of the primary research equation, everyone
’
s favorite value
investor
–
Warren
Buffett
–
does
his due diligence on the back of an envelope, or perhaps on a
Dairy
Queen
napkin if he
’
s feeling really fancy
, and we all know what he and Munger have done over the past
however many decades
.
Meanwhile,
h
is intellectual
protégé
Allan Mecham sits in an office above a taco
shop in Salt Lake City, reading 10
-
Ks and apparently never even bothering to call ma
nagement, and he
’
s
blown the market to bits
for, oh, a decade and a half.
The plural of “anecdote” is not “data,” of course, but
I do find it tremendously interesting that
everyone is
trying to gain an analytical edge and yet apparently in the collective never doing so
, judging by the poor performance
of most
long/short
funds.
You
’
ll find plenty of value investors willing to go to bat halfway on this
–
i.
e., advocating a non
-
quantitative approach to
valuation, where they prefer conservative guesstimates to
three
-
statement DCFs
with decimal
-
point precision
–
but
you
’
ll find very few people willing to stand up
and say,
“
primary research is overrated.
”
Because saying that sounds a lot like,
“
I
’
m lazy and I don
’
t
want to do much work,
”
and when you
’
re in an industry that
by and large optimizes for AUM rather
than for returns, you
’
re usually pretty concerned about the optics of what you said.
But the interesting thing about lying to other
people
is you eventually start to believe it yourself: and
clearly the majority of investors have internalized the idea that a) they n
eed an analytical edge to
outperform, and b) they somehow have one, despite the fact that
c) literally everyone else is trying t
o do
the same thing and, in the collective, not outperforming. So perhaps there are some people who truly
do have an analytical
edge, and can
utilize that to outperform, but in the aggregate it
’
s not a bet that
seems worth making.
What
’
s much more interesting, to me, is turning that premise on its head:
let
’
s start by assuming that I ne
ver
have an analytical edge.
What would that mean? That would mean that while I often do
“
deep research
”
(such as facility tours, talking to people in the industry, and so on), I view those as data po
ints rather
than determinants. Okay, this is one person
’
s opinion, or one experience dining at this restaurant
–
but
everyone has their own agenda, their own schema, their own
unique set of circumstances that clouds
their objectivity. One angle does not an investment thesis make, and the aggregate of
self
-
conducted
channel checks can often be summarized in terms of Yelp or Glassdoor reviews, or competitors
’
publicly
-
released financial statements, and so on.
That is to say, granularity is not always valuable: when chatting with a prospective LP, I brought up the
example of restaurant
chain
Fogo de Chao
(FOGO), of which I am a modest shareholder. The thesis,
reductionistically, is that the company is trading at an attra
ctive st
eady
-
state free
-
cash
-
flow yield and has
a long runway for accretively reinvesting that cash flow into new restaurants. New unit
economics are
fantastic, the concept already has
the best of both worlds
–
national
brand recognition and validation
Askeladden Capital
–
Primary
Research Whitep
aper
–
August
2016
Questions/thoughts/comments? thehappyanalyst@askeladdencapital.com
3
with a modest footprint
and plenty of room for expansion
–
and
over to the side you have some macro
no
ise around labor costs and
soft restaurant comps, but that
’
s all plus
-
or
-
minus stuff that really doesn
’
t
detract from the long
-
term thesis. The prospective LP was seeking to understand why I would spend
time working on
new names rather than drilling down deeper into Fogo. The answer? What would you
have me do?
Spend time and money visiting every new
restaurant when I could just watch Yelp reviews
instead? There is no new information
to garner here: the make/break question on the thesis is simply
can management competently put new boxes on the ground.
If the answer is yes, the stock works
over my
investment horizon; if the answer is no, then it doesn
’
t.
And again, this isn
’
t about Fogo, but rather the bigger point of forest/trees:
particularly when you
’
re a
kid running a fund out of your childhood bedroom (like me!)
there is always someone out there
who has
more resources than you. And that argument diminishes a bit when you
’
re at $100 million in AUM, or
$500, but it still applies
–
assuming
you
’
re the axe on a stock i
s incredibly dangerous.
But because it
makes a good pitch to prospective investors (look at all this incredible primary research we do!
), you
fool yourself into believing that
’
s how to outperform
–
even
though people who do that don
’
t
outperform.
So if analytical edge
is a mistaken assumption, what
’
s the right one? Behavioral
edge. Making good
decisions. Being epistemologically humble. Take the no
-
modeling
, ignore
-
the
-
macro
approach
on step
furth
er: don
’
t just look for really wide disparities to reasonably conservative fair values, but look for
really wide disparities to
level
-
of
-
analysis
-
required.
As Buffett and Munger say,
“
there are no points for
difficulty.
”
Just as it
’
s probably not a good investment thesis if you have to model it down to the
decimal point, it
’
s probably also not a good investment thesis if it hangs on the verdict of one
equipment dealer in su
per
-
rural Brazil.
If you study
the great value investors in depth
–
in
particular, Charlie Munger
–
what
emerges is the
concept of mental models. Most of the
world isn
’
t rocket science, and being an effective inves
tor (or
more broadly, an effective human being) isn
’
t necessarily
about being smarter or having some complex
secret
“
edge.
”
It
’
s about thinking about things the right way and making the right decisions over and
over and over again, in a room (/world) full
of people who don
’
t even try. People who
are too busy
worrying about what other people think of them,
who are too busy running away with premises they
didn
’
t create and nev
er bothered to stop to examine. Those people probably do a very good job of
squeezing out whatever analytical edge there is, bu
t they
’
re not even trying to make good decisions
–
it
’
s
just not on their radar screen.
Sa
ying
“
I
think about things a lot and
try to make good decisions
”
isn
’
t an investable pitch to
prospective LPs
, so it
’
s not very popular
–
but
I don
’
t care about being investable; I care about earning
returns. So it
’
s what I do.
Don
’
t take me the wrong way: I think primary research can be valuable
, and often engage in
it
.
Facility
visits in particular are super helpful.
But I think it
’
s not an end in and of itself.
It
’
s valuable when you
use it as an opportunity to ask the right questions, on the right stories, at the right times. It
’
s valuable
when you can get it for free (i.e., free
-
riding on data sources like Yelp or
Glassdoor or I suppose even
Twitter and Google Trends, and certainly on investing
-
idea sites like those I mentioned.) But the
fundamental problem of all organizations is
resource allocation, whether time or money: and my time
and money is better spent putting myself in the best positions to make
good decisions,
rather than
searching for a mythical analytical edge I
’
m never going to find.
Next
More from Askeladden Capital Management
The most important insight of the day
Get the Harvest Daily Digest newsletter.
