Venture Capital, Public Information, and the Law of Unintended
Consequences
If you follow the newspaper business pages, or take
a business journal, in the last year you may have noticed
stories on new revelations of the dark secrets of venture
capital management. If you flipped past, dismissing this
as part of the general media schadenfreude over the high
technology let-down, it's time to listen up. What you
don't know can hurt you, assuming you pay taxes.
Venture capital (VC) is considered a game for the sophisticated
investor. The money that goes into a venture capital fund
is raised from wealthy individuals, corporations, and
financial institutions. Institutions come in two flavors.
Private institutional investors are entities like insurance
and annuity companies, diversified financial managers
or corporate pension funds. Public institutions include
the endowments of public universities, and the retirement
trust funds for state and other governmental employees.
Generally, the annual and quarterly results and reports
of venture funds are not a matter of public record. Unlike
common mutual funds, the assets of a venture fund are
not listed on any exchange, and there's no easy way to
get a share value of the fund, or any of its private company
investments. When fund returns have been disclosed, they
are usually a combination of multiple funds in an institution's
portfolio, or even combined with other types of assets.
Many public institutional investors are arguably subject
to the Freedom of Information Act (FOIA). In the past
year or so, newspapers and some others have attempted
to gain access to the more detailed investment results
and valuations of specific funds. These attempts include
cases in Texas, California, and Massachusetts. Some of
the public institutions targeted with these requests have
settled, disclosing at least part of the results of their
investments, for instance CalPERS, the California public
employees retirement fund, and the investment arm of the
University of Michigan. Some selection of these results
has been published with great ado by the investigative
reporters behind the requests, to at least mild interest
by readers. So where's the problem? Surely the public
interest is being served?
Or is it? Some of the requesters for venture information
go beyond newspapers and others with an arguable interest
in putting the informatiion in the public record. For
instance, one frequent FOIA flyer is a California resident,
Mark O'Hare. What interest does a private individual have
in sueing for this information? He is using it to create
a private, for profit data bank. While there are existing
databases of venture capital information, these are based
on the voluntary release of information by funds and invested
companies.
How about the quality of analysis of data from the newspapers
requesting and sueing for VC information? Here's the San
Jose Mercury News' Matt Marshall:
CalPERS' online report... through September 2002,
shows Arrowpath drew $825,000 from CalPERS and its investments
[are] still worth $723,318. The difference is mostly
the fees that went to Arrowpath managers since 2001.
That's a little over $100,000.
Heinous, eh? 15% of the capital squandered on managers
in one year. But take a lookat the actual CalPERS commitment
to this fund. That's a matter of $5,000,000. Management
fees are based on committed capital, not on the capital
drawn to date (the $825,000). And 2% (on the low end of
management fees) of $5,000,000 is... $100,000. This is
a reporter who either does not understand how funds work,
or is being disingenous to sensationalize the story.
The actual evaluation of venture returns takes a bit
more sophisticated analysis than that. This is a very
modest introduction to the topic. If you looked at the
CalPERS and U of M results, one thing will jump out at
you: The investment returns (IRRs) are all over the place.
If watching your NASDAQ portfolio gives you alternate
adrenalin highs and heartburn, you don't want to be investing
in venture. From year to year (vintage), and between funds,
and through the history of each fund, there's very high
degree of variance in results. The up years of the Internet
bubble, and the downs of the bust, have added a stronger
cyclic element to that variance in the last decade, but
VC has always been volatile and it always will be.
By definition, there is no ready market for the shares
of the private companies in which venture funds invest.
We hang onto them until the wished-for liquidity event
(exit) of going public, or being bought out by another
company. Setting a value on these assets in the meantime
is more an art than a science, but you can be assured
that the valuations are both related to the NASDAQ averages
and even more volatile. The investors' shares in the fund
itself are likewise illiquid. If they are saleable at
all, it will be at a deep discount to another qualfiied
investor. Once you are in a fund, all you can do is grit
your teeth and hope it's one of the funds and vintage
years that win.
You may also be assured that you will take losses the
first few years. The annualized percentage returns for
maturing funds (5+ years from inception), are a sort of
polite fiction. This isn't a CD or T-bill that quietly
and evenly accrues interest over time. In venture, bad
news arrives first, hopefully followed by good news in
the long run. Some call this the law of the J-curve, and
the reasons for it are fairly simple:
Brand new startup companies are by definition risky,
and some fraction of investments in them will turn out
to be bad choices. The wise fund manager will kill and
write off the bad ones as soon as the trouble is apparent.
Meanwhile, the good companies take time to build their
products, reach their markets, grow, and exit at a profit.
Management fees, the largest part of the overhead costs
of a fund, are charged evenly through its life based on
its total size (committed capital), so that expense is
also front loaded compared to the eventual profits as
good investments mature. So not only do you get to grit
your teeth, you're on a roller coaster ride that almost
always starts out with a swoop toward the ground.
Why do public institutions invest in venture capital
funds, if it's such a dodgy business? Simple: Over the
long run, averaged over multiple funds and vintage years,
venture capital has a higher rate of return than most
other asset classes. This higher average return is the
market's quid pro quo for the investors' acceptance of
the drastic ups and downs in this high variance asset
class. A prudent manager of a long term portfolio such
as a pension fund or endowment won't put a high percentage
of their value into venture capital, because the valuation
and returns are too volatile year to year, and the asset
is illiquid. But a good manager will have some fraction,
often 10 - 15%, spread among multiple VC funds to enhance
the long term return on their assets. They may drink their
Maalox during the down years, but the institutional managers
can console themselves that time and the law of averages
are on their side.
A good investor in venture capital funds should understand
this proposition and be ready to withstand bad years in
order to participate in the good ones. As a negative example,
corporate investors are notorious for their inability
to invest through the whole cycle. They tend to enter
the VC markets when they are already bouyant. Since dropping
valuations and low returns can make things too hot politically
for their in-house management sponsors, they often drop
out at the bottom of a market, either ceasing investment
or liquidating their portfolio entirely. This makes the
average corporate less desirable as an investor to fund
managers, and pretty much guarantees that they miss the
rewards of market upturns, which are supposed to be their
compensation for being in a volatile asset.
So here's the first unintended consequence of disclosure
of VC returns by public entities: They will start to act
more like the corporate investors. The newspapers are,
after all, going to need to get a return on their investment
of time by running more stories, and there is the public
interest to be served:
We can now parse all of the data available to us
-- a firm's results, the fees made by its managers,
the donations the managers made to the political campaigns
of CalPERS' elected officials (link mine) -- and decide
whether we think CalPERS' investments are made fairly
and based on merit alone. - San Jose Mercury News
The next unintended consequence is that public entities
will have more trouble placing their money into venture
capital funds. Venture capital management firms prize
investors who are loyal when times are bad. They dislike
investors who create extra burdens of reporting and entanglements
with regulation and politics. Compared to private institutional
investors, public entities will become a denigrated class.
This is not speculation. Managers from firms such as Crescendo,
Storm Ventures, Charles River and Sequoia have publicly
stated that they will be less likely to take money from
public entities, or will avoid them entirely. Sequoia
has carried through on its statement, evicting the University
of Michigan. The upshot is that public entities will end
up in funds whose managers have had lower performance
in the past, since these funds have less choice in investors.
Another bias toward lower performance.
This is not the end of it. Some of the information requests
to public entities have gone beyond quarterly total return
numbers. For instance, the San Jose Mercury News also
pushed to receive the specific valuations set on private
companies by each fund in the CalPERS portfolio. This
is anathema to venture fund managers. It is considered
valuable proprietary information of both the fund and
the invested company. Exposure of valuations can drastically
affect the ability of a company to negotiate for further
financing. Disclosure that its funders have written down
the value since investing can weaken a company in the
eyes of both new investors and prospective customers.
Disclosure of a high valuation can also botch a deal,
by indicating that a mutual agreement on investment terms
is unlikely. For the infant start-up company, stealth
as well as speed are often the only weapons against incumbent
competitors.
A venture fund subject to this type of reporting will
become the victim of understandable discrimination by
entrepreneurs. The start-up executive with a good track
record, who has a choice of investors, is not going to
accept exposure to this risk and will simply avoid such
funds. Since fund managers cannot accept the downward
bias on performance this implies, at this point public
entities will be kicked out of the venture asset class
entirely. They then lose access to the greater returns
over time that have been provided to their overall portfolios.
And the final unintended consequence is the loss of the
portfolio yield that it was in the 'public interest' to
protect against undue influence in the first place. The
first order losers? The retirees, students and universities
that are the beneficiaries of the public entity portfolios.
By Tim Oren
