With limited partners clamoring to enter reputable venture funds and oversubscribed vehicles becoming the norm even as venture capitalists raise colossal pools of money, it might seem silly to ask whether venture capital investing is worthwhile.
Investors in venture partnerships, however, have to consider the question, given the industry’s long-term commitments, its illiquidity and the difficulty of comparing the internal rates of return of VC funds with the performance of other asset classes.
First is the issue of determining just how risky venture capital truly is compared to other types of investments such as public stocks or treasury notes. Matching venture IRRs to an index such as the S&P 500 won’t suffice because general partners tend to hold portfolio companies at cost until repricing takes place at subsequent financings or liquidity is reached. Meanwhile, the public stock market bounces up and down, frustrating investors who want to make real-time comparisons between their public and private holdings.
The “stale price problem,” as Harvard Business School Professor Josh Lerner calls it, was not such a big deal when venture capital was a mere 2% or 3% of a pension or endowment’s asset allocation. But as private equity has become more popular and institutions have boosted allocations to 10% or more, the necessity of making an apples-to-apples comparison between venture capital and public investments – and the difficulty in doing so – has become more troublesome.
Almost two years ago, Dr. Lerner and his Harvard colleague, Paul Gompers, published an article in The Journal of Private Equity suggesting a way to compute venture capital performance based on using a formula to determine the value of a venture fund’s portfolio and comparing it with the public market’s performance over a fixed period.
The complex formula took into account the change in the value of public market companies comparable to those in the VC’s portfolio; the change in relative prices of private and public equity markets as a whole; the change in profitability of a portfolio company; the degree of leverage of the enterprise; and the change in other relevant characteristics of the company, such as the receipt of patents.
Although the approach would be useful to limited partners interested in weighing venture investments against stock holdings, it has yet to be adopted by the industry, and, as a result, IRRs remain the key means of assessing a venture fund’s performance.
Learning to Live With the Lumpiness
Wilshire Associates Vice President Erica Bushner likens static venture capital valuations to real estate, an appraised asset class. From an accounting standpoint, venture capital offers high returns with relatively low volatility, and venture valuations are more stable than the public markets, she says. L.P.s also know how valuations are being computed because they are spelled out in partnership agreements.
Sovereign Managing Principal Katie Cattanach says investors have to learn to live with the “lumpiness” of VC performance data. Nevertheless, investment officials at public pensions, endowments and foundations must report some kind venture capital performance data to their boards, and they must use some benchmark to judge their portfolios.
The $97.5 billion California State Teachers’ Retirement System (CalSTRS), for example, wants its venture portfolio to at least beat the median IRR for each vintage year and would prefer to reach the top quartile. According to CalSTRS’ consultant, Pathway Capital Management Inc., through April of this year the pension achieved a 12% return on 1989 funds while the median was 9.3% – CalSTRS notched a 32% return on its 1990 funds, soaring past the 18% median. And for 1992 vehicles, its funds were returning 38% while the median was 19.3%, says Chief Investment Officer Pat Mitchell.
CalSTRS’ former private equity Investment Officer Solomon Owayda (VCJ, February 1997, page 7) and previous consultant Abbott Capital Management managed to keep returns for the pension’s VC portfolio at the upper quartile, Mr. Mitchell says. But CalSTRS is not only interested in how its portfolio stacks up against other venture funds, it also cares how its VC investments measure against the stock market. CalSTRS wants VC returns to beat the Russell 3000 Index by 5%, a premium the pension believes is merited because of the illiquidity and long-term nature of venture investing, Mr. Mitchell explains.
CalSTRS has missed the 5% mark by a long shot. Through April of this year, one-year returns for the Russell 3000 were 17.5%, while CalSTRS’ net VC returns were 18.4%.
Over three years, the Russell Index yielded a 26% return, but in that same time period the pension’s venture investments beat the index by less than half of a percent, turning in 26.4%.
The picture is a bit brighter over five years, however, with CalSTRS’ VC holdings notching 26.3%, as the Russell Index came in at 24.9%.
Mr. Mitchell says CalSTRS still aims for the 5% mark, but even without reaching it he thinks venture investing has been worthwhile for the pension, simply because venture returns are beating the public markets, even if slightly.
Venture capital investors such as CalSTRS who acknowledge the potential value of the asset class also recognize the inefficiency of the private market – a trait that can create potentially great returns and widely varying performance among funds. Therefore, choosing venture firms that are best able to exploit that inefficiency – rather than be exploited by it – can be a daunting task, but it is crucial to building a successful venture portfolio.
Not All Managers Are Equal
In an effort to determine how important manager selection is in various asset classes, Wilshire studied the spread in returns in a variety of sectors between the mid-1980s and the mid-1990s by top-quartile performance and median managers. Among public bond managers, the spread was a mere one-half of 1%; the figure jumped to 1.3% for public equities managers. The difference among international equities managers was 2.1%, and for venture capital, the spread was 7.6%.
The chasm between top quartile and median VC managers illustrates how important it is for L.P.s to identify top-quality managers and steer clear of the rest.
By the end of 1998, cumulative IRRs for all existing venture funds were all over the map. The top performer boasted a 472.3% return, and the worst 72.4%. The upper quartile IRR – the 75th percentile – sat at 18.2%, while the median came in at 8.4%. The lower quartile – the 25th percentile – notched a mere 1% return. One should keep in mind, however, that the 747 funds examined by Venture Economics Information Services, a sister company to Venture Capital Journal, include funds very early in their development, which could be expected to yield low or even negative returns. Skimming off those early-stage funds and studying returns by vintage year portrays a more meaningful picture, and startling differences remain between different performers.
For example, Venture Economics listed 25 venture funds raised in 1992. The upper quartile return through 1998 for 1992 funds was 23.8%, while the median was 14.1% and the lower quartile 8%.
Also worth noting is the difference between the averages, capital-weighted averages and medians for each vintage year’s funds. The capital-weighted averages give proportional consideration to each fund. The pooled average simply treats all returns as though they came from one giant fund. The median marks the midpoint along the spectrum of returns earned by each fund of a vintage year.
For 1992 funds, the average return was 23.3%, while the capital-weighted return was 26.7%. Indeed, cap-weighted averages were higher than averages for every vintage year from 1980 through 1998, except 1996 and 1998, suggesting that bigger funds performed better. That does not mean, however, that the size of a fund caused stronger performance. Presumably, firms with longer track records raise bigger funds than newer or less successful firms, and solid performers are rewarded with bigger funds if they continue yielding high returns.
By the end of last year, almost three-fourths of the funds raised in 1992 were yielding returns below the average, again underlining the importance of careful manager choice.
Take Your Pick – If You Can
When it comes to venture fund selection, not all limited partners are equal. Those with sizable pools of capital and long-standing relationships with top-tier firms have a better shot at getting into the funds of top-quartile firms than newcomers with smaller pools and lesser-known names.
Several advisers and general partners suggest that newcomer L.P.s consider funds-of-funds as a way of getting into better-name vehicles while gaining diversification and strongly recommend against plunking an entire allocation of, say, $10 million into one or two funds.
Nevertheless, August Capital Chief Financial Officer and Partner Mark Wilson cautions that investors must carefully select their funds-of-funds managers, making sure they have access to the firms the budding VC investor wants.
Bessemer Venture Partners G.P. Chris Gabrieli also advises institutional investors to be more receptive to first-time funds with firms whose partners have venture capital experience. He is not convinced that investors in older funds necessarily reap better rewards, pointing to Benchmark Capital as a highly successful younger firm. Benchmark’s first fund invested in eBay Inc., thr online auction company that held one of last year’s most successful venture-backed IPOs.
August’s Mr. Wilson warned, however, with the faster pace of fund raising, firms are returning to market to raise their third fund before their second fund shows any meaningful returns, forcing investors to judge performance only through the first fund. The problem with that is the first fund would have been invested about eight to 10 years earlier, and factors that contributed to that vehicle’s success – personnel and strategy, for example – might have changed. Investors must be alert to detect such shifts.
In any case, even younger firms might not want inexperienced investors. Benchmark, whose general partners worked for well-regarded firms before forming the group in 1995, attracted investors from their previous firms. Unfamiliar investors were scrutinized to determine whether they would be long-term players in the asset class, and Benchmark did not welcome first-time VC backers, says General Partner Andy Rachleff. The firm prefers limiteds from charitable foundations because they take a hands-off approach and tend to be long-term investors. Most top-flight firms avoid state or gatekeeper money, he adds, in part because they often try to lower fees and carries. Mr. Rachleff declines to discuss specifics about Benchmark’s carry structure, except to say the firm has had “a premium carry” from its first fund.
The Benchmark G.P. suggests new investors consider the fund-of-funds route or hire an investment professional from another pension, endowment or foundation with personal connections to top-tier firms.
Like the founders of Benchmark, Russ Aldrich left one firm, Redleaf Ventures, to start a new one, Guide Ventures. Having worked on one Redleaf fund, Mr. Aldrich split from the group earlier this year to raise a Northwest-focused vehicle using the same seed- and early-stage hands-on approach that Redleaf employs. Guide Ventures II L.P. (see page 25) is aiming for $25 million and had notched about $10 million by press time. At two years old, Redleaf I, the $15 million fund Mr. Aldrich worked on and still manages, is too young to show significant returns, making Mr. Aldrich’s fund-raising efforts considerably more difficult than those of established firms.
Still, Mr. Aldrich would be leery of accepting money from a newcomer to VC, in part because of the unproved results of his investment approach. “Call me conservative, but I want people who understand the risks associated with a venture fund, who understand the variance in the model,” he says.
It’s Who You Know
Despite many G.P.s’ discomfort with newcomers to venture investing – and a market that lets the venture firm pick and choose among L.P.s – first-time venture investors still can gain entry into big-name vehicles.
Situated in the heart of Silicon Valley, Santa Clara University decided to invest in venture capital about two years ago and asked acquaintances in the financial world to list the top 10 venture firms, university President Paul Locatelli recalls.
Then the school investigated the general partners of those firms to find who among them were Santa Clara alumni, spouses of Santa Clara graduates or who knew people affiliated with the university.
Personal contacts, including Santa Clara alumni Jim Strand at Institutional Venture Partners and Nancy Schoendorf at Mohr, Davidow Ventures, opened doors for Rev. Locatelli to meet with both firms. Rev. Locatelli, who used to be a finance instructor, had personally known Mohr Davidow’s Bill Davidow and veteran VC Don Lucas, Sr., with whom the university had invested. In addition, a Santa Clara trustee happened to be an adviser to Sequoia Capital and brought the university and that firm together.
The personal connections allowed Rev. Locatelli to sell Santa Clara to the venture firms, and he assured them the university would return as an investor, so long as fund performance levels merited a continued relationship.
Santa Clara has invested in IVP, Sequoia and Mohr Davidow with Mr. Lucas, and in a small fund managed by Aspen Ventures, a firm started by Santa Clara graduates. Only one firm has denied the school access, but the group promised to try to wedge Santa Clara into a side fund. Rev. Locatelli anticipates the endowment will back one more VC group and will count on the subsequent funds of existing partnerships to flesh out Santa Clara’s 3% to 5% allocation to venture capital.
Not every newcomer to the asset class is as fortunate as Santa Clara in getting into big-name funds. But lesser-known and younger funds must also raise money. Sovereign will consider investing in a fund even if it has low venture returns, if it has performed relatively well in an unpopular sector, such as biotechnology, says Dr. Cattanach.
In analyzing all funds, Sovereign wants to uncover the secret to a firm’s success. Was a fund’s IRR saved by one home-run investment out of a stable of 14, asks Sovereign Analyst Gary Ratliff, who says the adviser prefers to see consistent success over a portfolio’s deals, rather than one or two companies carrying the rest on their coattails.
Wilshire’s Ms. Bushner says what marks the kiss of death for her while choosing an investment is when, in conducting due diligence, she runs into a series of cross complaints from a partner and a previous employer that she cannot completely sort out.
Defending the “Wild West”
Harvard’s Dr. Lerner says less experienced or less successful venture firms can look to corporate investors for backing because they tend to be more interested in forming strategic partnerships than in seeking high returns. Such VC firms also could align with strategic limited partners, such as banks, and offer favorable terms.
Finally, firms with mediocre returns should be able to explain what went wrong and describe to potential L.P.s how the group has changed its investment strategy to ensure better performance, the professor says.
Despite concerns about the risk of various venture funds and the difficulty of comparing their returns with those of other investment classes, Bessemer’s Mr. Gabrieli sees VC as a relatively safe bet because even poor VC performers generally break even.
For funds raised between 1980 and 1993, the bottom-performing fund for every year except 1986 had a negative IRR at the end of 1998, but lower-quartile funds had positive IRRs for every vintage year except 1981 and 1991, according to Venture Economics. Hence, when it comes to capital preservation, Mr. Gabrieli sees VC as remarkably reliable.
The general perception about Mr. Gabrieli’s profession, even among relatively wealthy, sophisticated individual investors, he says, is that it’s “just completely crazed, cowboy risky.” But the Bessemer G.P. continues to put his money beyond his faith – his personal portfolio is overwhelmingly in private equity and contains no public stocks, he notes.
Cumulative Vintage Year Performance as of 12/31/98
U.S. Venture Capital Funds (only)
Calculation Type: IRR
Vintage Capwtd Pooled Upper Lower
Year Num Avg Avg Avg Max Quart. Median Quart. Min
1980 19 14.5 22.5 20.0 31.8 18.8 13.6 9.3 -1.9
1981 21 6.8 9.5 8.4 25.4 12.7 8.8 -0.2 -5.9
1982 29 3.3 4.4 4.5 13.5 8.5 4.2 0.1 -19.1
1983 58 6.2 7.0 7.9 41.5 10.6 5.8 1.5 -11.4
1984 68 4.9 5.5 6.1 25.5 11.4 4.3 1.1 -18.4
1985 47 7.0 9.5 10.6 28.3 14.9 7.9 1.6 -41.5
1986 43 7.9 12.1 13.8 25.6 12.1 6.4 3.4 0
1987 66 8.4 13.1 13.9 41.8 17.2 7.9 0.8 -37.8
1988 48 11.3 18.2 18.0 42.9 19.2 7.7 2.2 -9.7
1989 56 12.6 17.7 19.1 56.1 22.0 10.0 0.8 -19.3
1990 23 17.6 24.9 28.2 75.2 29.8 10.6 0.7 -8.8
1991 16 14.8 20.1 21.0 62.2 24.4 14.8 -0.8 -11.2
1992 25 23.3 26.7 31.3 102.6 23.8 14.1 8 -6.4
1993 40 19.6 25.9 30.2 83.4 27.9 10.5 3.7 -6.6
1994 37 21.7 26.4 30.9 99.7 34.6 20.0 4.1 -23.6
1995 33 30.3 37.1 39.3 252.3 38.2 17.3 7.6 -7.9
1996 29 42.5 41.7 44.6 472.3 41.2 12.9 0.6 -19.1
1997 42 13.1 16.1 20.9 156.7 29.7 0.6 -7.5 -61.8
1998 16 2.5 -16.8 37.5 262.1 -1.3 -10.7 -20.5 -72.4
Source: Venture Economics Information Services/NVCA