Return Expectations

March 24, 2010

Fat Startups, BMI and The Lorax

So, there I was hanging out in Denver airport again, reading the back-and-forth between Ben Horowitz and Fred Wilson on fat versus thin startups.  They’re both cats that I respect and I think they both make carefully reasoned, nuanced arguments that have far more depth than a quick summary can articulate. 

But all this talk of heft got me thinking about the Body Mass Index; those hip to the BMI calculation know that it's a measure of weight relative to height.  Doctors tend to be fans because it's a decent indicator of body fat which correlates with increased risk of morbidity and mortality.

And BMI tends to work pretty well, except when it doesn't.  For example, serious athletes tend to have more muscle per unit of height than the rest of us and thus have unusually high BMIs: Arnold Schwarzenegger's BMI clocks in at 33 (the "ideal weight range" runs from 18.5 to 24.9).  Even George Clooney clocks in at 29, just shy of the "obese" cutoff of 29.9.

The beauty of muscle, though, is that it's so much more metabolically active than fat.  That's why one can be heavy, but fit; the two are not mutually exclusive.  And  neither, necessarily, are capital efficiency and cash abundance in the startup world.  Indeed, while I do find myself more firmly in Fred's "lean" camp, I am sympathetic to the notion that sometimes a
company with a lot of dollars on hand can pivot more quickly as business
conditions change, or scale faster to discourage new
entrants or build competitive advantages
that give them a leg up on existing rivals.  Companies can be well-funded but hungry.  I get it.

But here's my question: sure, bigger companies can earn bigger exits, but if they burn more cash on the way to a big outcome, isn't it almost inevitable that return multiples will be depressed?  Indeed, there can be some positive externalities associated with a big outcome, but as an LP — the money behind the money — what I care most about is multiple on investment.  I'm focused on the numerator and the denominator.

And then I started thinking about some of the cleavages of interest that may arise from fatter startups.  Most directly, the dollar value of carry to a GP of 2x outcome on a $100 million investment is greater than that of a 5x on $10 million.  I know which outcome I would rather have, but I'm not sure that a random GP's answer would be the same.  Then you get into non-economic motivations: having big outcomes can help build a brand, almost regardless of the profits.  Few people know what multiple Sequoia, for example, generated on Cisco's IPO twenty years ago, It could be a 100x, or it could be a 2x.  Most likely, it's somewhere in between, but either way it's an awfully spiffy logo to have on the web page.  Entrepreneurs, too, can get a lot of jazz from landing a big round of funding; it's an endorsement of their idea and the vote of confidence that makes for good press releases and generates respect.

Further, aiming to build a large enterprise fundamentally changes a startup's optionality profile.  No longer are a wide range of exit scenarios compelling; suddenly, a happy outcome for all becomes limited to one of the handful of large exits that take place in a given year.  With expectations of a modest number of IPOs (which I fear is a structural thing,) and a few dozen M&A transactions in excess of $250M, all those fat startups are like so many well-prepped high school seniors looking to land a spot at Top College.  It's just tough arithmetic.  It may be a fun ride for the entrepreneurs and GPs, but it can be nerve-racking for us LPs at the tail end of the whip.

And the risks are different, as well.  "Fatter" startups imply either fewer portfolio investments or larger funds (or perhaps even imbalanced funds where a few large bets compete for GP time with many small ones).  None of these are particularly positive outcomes for fund investors. 

I could go on, but it would start to stress me out.  And one of my sources of stress is that entrepreneurs and GPs are pretty well represented in these discussions, but we LPs tend to be distant observers.  Who speaks for us? I wondered out loud.  And just at that moment a little man appeared out of the breeze:

[With apologies to Dr. Seuss]

“Mister!” he said with a sawdusty sneeze,
“I am the Lorax.  I speak for LPs
I speak for LPs for LPs have no tongues.
And I’m asking you sir, at the top of my lungs” –
He was extremely upset as he warned retribution –
“What’s that THING they have done with our cash contribution?”

“Look, Lorax,” I said, “it's easy to get crotchety.
GPs deploy billions; it sometimes still bothers me.
In fact,” I continued, “some think it’s a lottery.
A lottery whose tickets can be a new social network,
A diversion to give cube-dwellers ways to shuck work.
But it has other uses.  No, we’re not out of the woods.
You can use it to serve ads or sell virtual goods.
Or crowdsource a date or find homes in new ‘hoods.”

The Lorax said,
“Sir! You are crazy from drinking the potion.
There is no one on earth who would buy that fool notion!”

But the very next moment we were shown to be wrong.
For, just at that moment, a ‘tween came along.
And she thought that the club we’d just ridiculed was great.
She happily subscribed for thirty-three ninety-eight.

I laughed with the Lorax.  “You poor stupid guy!
You never can tell what some people will buy.”

“I repeat,” cried the Lorax,
“I speak for LPs!”

“All GPs are top quartile,” I told him.
“Those marks are a tease.”

July 15, 2008

Tyranny of the Relativists

There’s a great riff in John McWhorter’s book, The Power of Babel: A Natural History of
Language, that describes how expressive force in language diminishes over
time.  His poster child is the word
“terrible” which once meant “causing dread or fear or terror,” but now is used
as a mild negative modifier (e.g.,
“I’m such a terrible putter that my wife got me personalized golf balls that
say, ‘three putt.’ ”) 

Similarly, the language of business is littered with
expressions that slouched from overuse: phrases like “best of breed” went from
insightful to hackneyed over the course of mere years.
Of course, private equity has its very own once-powerful phrase that’s
meandered to meaninglessness: Top Quartile. 
Now we all know how Top Quartile got to be the gold standard: a couple
of cool graphs in the Book of David some McKinsey studies, an
endorsement by an influential pension fund or two and Top Quartile was the
place to be. 

Everyone wanted to be in the Top Quartile, and, soon enough,
most people were – if you accepted their pro forma returns based on aggressive
multiples of EBBS (earnings before, ahem, bad stuff) or included estimates of
what an acquirer might pay for a start-up once it had perfected its swell
product and revenues had started their inevitable ramp.

The problem with that kind of thinking is that it obscured
the true aim of private equity: return enhancement (relative to the opportunity
cost of equity capital, i.e., some
appropriate public benchmark).  Success
was no longer about getting adequately compensated for taking incremental
risks; it was now about beating the next guy (or more specifically, the next
three guys).  

And that’s the paradox, of course, of any peer-group
measure: when new entrants deploy capital, their extra dollars will likely
depress returns across the spectrum – since the marginal dollar tends to set
the price of assets – but the cohort of “winners” (with victory defined as
being Top Quartile) grows.  The club gets
less elite as people enter from the bottom.

Of course, measuring private performance against a public
benchmark can be pretty tricky and is fraught with tough questions like: what’s
the appropriate evaluation horizon, how do you equalize for appraisal effects,
should you beta-adjust? to name a few. 
As a result, it’s tempting to fall back on an easy measure like peer
group rank.

A while back, I started asking VCs what they thought the top
quartile breakpoint for venture over coming vintage years might be assuming
reasonably “normal” public market conditions. 
In my informal and unscientific survey, more than half of those I asked
guessed that the top quartile line would be at or below zero.  Surely, some firms will do fantastically
well, but if the VCs I sampled are right, it will be really amazing that dozens
and dozens and dozens of firms that tinker in one of the riskiest corners of
the financial world will be able to call themselves successful even though they
will have lagged the public market – not to mention cash – by many hundreds of
basis points.  What’s more amazing is
that merely by surviving, these firms will earn billions of dollars in fees in
aggregate and the folks who funded them will dance a jig for having done a
great job.

It’s enough to remind one of an old story: two campers are
awakened by the sounds of a bear sniffing at their tent.  While one of the campers starts panicking,
the other calmly starts putting on his sneakers.  “Why are you putting on your shoes?  There’s no way you can outrun that bear,”
whispers the first camper.  “I don’t need
to outrun the bear,” replies the second camper, “I just need to outrun you.”

That kind of thinking may work for relativists, but the
important return-enhancing role that PE plays in institutional portfolios
demands that we instead ask: for all the extra risk and illiquidity we incur,
if you can’t outrun the bears (and the bulls) why bother?

May 27, 2008

Beware of PE Managers Speaking Greek

So here’s a game I’m playing a lot lately: dueling
Greek.  No, I’m not finally fulfilling my
father’s long-dead wish that I attend Greek School.  It’s just that a lot of PE managers are
slinging the Greek around.  Seven or
eight years ago it was really rare for an LBO pro or VC to describe themselves
as an “alpha” manager, but it seems commonplace today.

Of course, I understand what people are getting at.  They’re suggesting that they’re after
improved risk-adjusted return.  But
that’s where it gets dicey: how do you articulate risk in a PE context?  And what’s the appropriate level of return
per unit of risk?

When I’m feeling saucy, I ask these self-proclaimed alpha managers: “ok, then,
what’s your beta?”  At that point, the
GP probably thinks that he’s got a smart aleck on his hands, but the question
is sincere.  Really, it is!  (I’m not one of those Dollar and a Dream
guys.)

As we all remember, alpha is the excess return of a
portfolio relative to the return predicted by a portfolio of similar risk
(expressed as beta).  Said more elegantly (note the gratuitous equation inserted to gain
credibility with the quants):

Portfolio alpha = Portfolio Return – (Risk Free Rate + Portfolio Beta * Equity Risk Premium)

So assuming a long-horizon nominal equity risk premium around
6.5% and an average nominal risk free rate of about 4.5%, a portfolio with a
beta of one should have a return of about 11% and anything above that is alpha,
sweet alpha.  [We could quibble about
time-varying premia and risk free rates, but the generic point is valid,
no?]  Here’s the catch: I just don’t
believe that PE betas are anywhere close to one; on the buyout side, higher
levels of debt implicitly raise beta while the venture guys have incredible
volatility of outcomes.

Fortunately, most folks decide to play along with the spirit
of the question and we typically get into good discussions of risk appetite,
tolerance, and management.  It’s always a
stimulating chat, but people rarely answer my question directly.  To be a sport, a GP once argued – with a grin
– that the beta had to be less than 1 because most people thought the
correlation of PE to the public markets was about 0.65.  Without skipping a beat, he acknowledged that
appraisal effects and stale prices made that number totally meaningless.

So, what is the beta of PE? 
Is it 1.25 (kinda like Vanguard’s small-cap Explorer fund)?  Or more like 2.0?  If it’s the former, you’ve got to generate
net returns greater than about 13% to have positive alpha; if it’s the latter,
your bogey is closer to 18%. 
[Interestingly, the time-worn PE goal of 500 bps of excess return
probably implies a beta of about 1.75, which is
about the beta exhibited by some public managers with concentrated portfolios].

At this point – like Chevy Chase
playing Gerald Ford – I exclaim, “I was told there would be no math here,” so
why sling numbers around in search of a pedantic point?  Because PE supposed to be a return enhancer
relative to public market alternatives (the opportunity cost of risk
capital).  I’m finding that when people
say “alpha,” they typically just mean “top quartile,” and that’s not alpha at
all.  Maybe “top quartile” will equal
positive alpha over time – I have my doubts – but right now it’s just a worn
marketing slogan.  Don’t get me wrong, I
love firms who can outperform their peer group, but unless we can figure out
how to find the folks who are generating true risk-adjusted excess
return, we need to keep asking ourselves the question: why bother? 

May 6, 2008

Justify My Love

My buddy Peter – a very smart cat – was a pure math major in
college. Once, I asked him what the
difference between pure and applied math was and he told me with an impish
grin: “the applied math guys know how to add . . .” Of course, Peter went to Brown University, a
very funky place, so I’m sure that he could’ve done an interpretive dance about
the Lebesgue Outer Measure and still gotten a passing grade on his senior
thesis (just kidding . . . feel the love, Providence!)

I, on the other hand, studied history in college which means
that I’m good with trivia at cocktail parties, but that’s about it. Every now and again, though, I get to
thinking about arithmetic; specifically, the arithmetic of the venture business
and I wonder if we’re all closer to the pure math end of the spectrum than the
applied end.

VC math should be pretty straightforward: send a dollar out
to a portfolio company and hope it comes back with a few of its friends. Do that often enough and you’ve got a good
fund-level return.

Unfortunately, the LPs who invest a Dollar and a Dream have
prevented the shakeout that we were all talking about in 2002 from happening
and there continue to be too many iffy $500 million “early stage” funds out
there. Now I’ve got nothing against $500
million funds in particular. Despite my
seed-stage and smaller-fund bias (I like being "long idiosyncrasy and short
momentum"), we’ve got a few investments in that size stratum and think those specific
guys have some distinctive advantages.

Here’s where it gets dicey for the masses, though (and I’ll
make some gross simplifying assumptions): if you’re an LP and investing in an
run-of-the-mill $500 million fund hoping to get a 3x net return, that fund has
to generate $1.75 billion in returns ($1.25B in profit less 20% carry equals two
turns of profit). Of course, that’s just
the capital that accrues to the firm’s ownership stake. Since a lot of firms end up owning only
10-15% of their companies at exit, you’ve typically got to gross the $1.75
billion up by a factor of between 6.67 and 10. That suggests that those firms need
to create between $12 and $17 billion of market cap just to get a 3x
fund-level net return to their LPs. Caliente!

Let’s unpack that box a bit more: at the $15 billion midpoint of the exit range
above, a firm that invests in 25 early-stage companies will have to get, on
average, $600 million exit valuations for each and every one of them. That’s a pretty daunting number when you
consider that the typical M&A valuation has hovered in the high
double-digit millions for quite some time.

Of course, such a batting average would be unprecedented
(this is a slugging percentage business, after all), so if you assume that a
quarter of the companies generate all the returns while the other three
quarters collectively return the cost basis, each of those 6 home run companies has to enjoy an exit valuation
of $1.67 billion (roughly what Google paid for YouTube). That’s livin’ la vida loca!

The situation above is exacerbated by the fact that not all
firms invest 100% of their capital because they reserve up to 15% of capital for
fees. Also, you could make the argument
that the firms most likely to earn the above returns will charge premium
carries, making the hurdle higher for compelling net returns. To be fair, firms have a few levers to pull –
maintaining higher ownership percentages
(!) in companies and recycling capital – that can make the challenge less
daunting. They could also deploy less
capital per company, but that’s tough to do with a larger fund.

Like I said, though, I do still believe that some firms will
be the exceptions that prove the rule; some will be good while some others will
be lucky.

In the meanwhile, a lot of LPs will be serenading their GPs
with the line from that old Madonna song (cue the sensuous and moody bass line):
“I’m just wanting, needing, waiting for you to justify my love. Hoping, praying for you to justify my love .
. .”

April 2, 2008

A Dollar and a Dream

Anyone remember Curtis Sharp? A long time ago, Curtis won a $5 million New York Lottery jackpot and he spent the rest of the 1980s sashaying around New York in bespoke suits and bowler hats attracting crowds wherever he went. Curtis was the Man. Charismatic and impish, Curtis was the Toast of the Town.

I once saw Curtis at a Yankee game and the aura around him was palpable. He was Electric. People just wanted to touch Curtis. Maybe some of that luck would rub off.

That afternoon, a few rows behind us, a drunk kept yelling at no one in particular: “You can’t win if you don’t play!” I thought that the guy was pretty creative for exhorting the listless Yanks (it was 1989, after all) to get in the game by using the Lotto marketing slogan within earshot of Curtis. I appreciated the confluence.

I think of Curtis often . . . around my shop, I’ve started calling Curtis the patron saint of LPs who invest in venture capital funds. Don’t get me wrong, I love VC. We continue to find extraordinary people doing venture investing in extraordinary ways.

It just strikes me that when you push some LPs to articulate why they have outsize venture commitments when history shows that only the smallest slice of the business has rewarded the faith, they throw out New York Lottery slogans with Ivy League veneer. They use words like “optionality,” and “asymmetric payoff.” They might as well be mimicking the catchphrases from the Lotto TV commercials: “Hey, you never know,” or “All you need is a dollar and a dream.”

They just want to be Curtis: The Guy That Beat The Odds.

I hope that everyone makes loads of money and I never root against anyone. That’s bad karma. I do find myself asking, though, how many LPs who invest in venture firms are committing the fallacy of composition: Some people have made fabulous returns in venture, therefore venture will provide fabulous returns. But hey, as they say: you can’t win if you don’t play.