Is the US venture capital model broken? Does it need to be appreciably smaller? Does it need to be appreciably different?
We are sceptical of claims that the VC model is broken or needs to be radically changed. As our historical analysis indicates, the level of commitments to and the investment pace of the US VC industry since 2002 have both been consistent with the historic averages.
At the same time, the returns to VC funds appear to have been roughly equal to those of the overall stock market. This does not suggest to us that there is too much money in US VC, nor does it indicate to us that the VC model is broken. Instead it appears to represent the more or less natural evolution of a relatively competitive market.
In fact, the only real difference was the unusual and unexplained paucity of VC-backed initial public offerings, averaging only slightly more than 50 a year between 2004 and 2007.
The small number of IPOs from 2004 to 2007 came despite the robust stock market over that period and despite the large number of companies that received VC funding over the previous five to 10 years. By comparison, in all but one year during the 1990s, there were more than 100 VC-backed IPOs. In five of the 10 years, there were more than 150. Then, in the recession/bear market of 2001 to 2003, the number of VC-backed IPOs dropped below 50 each year. But this was not unusual for a down-market – a similar pattern had occurred in the bear market from 1989 to 1991.
It is not yet clear why there were so few IPOs. Some blame the increased costs imposed on companies by legislation. Some blame increased litigation risk and the concomitant increase in directors’ and officers’ and other insurance. Some blame inattention from investment banks that were able to make more money from other activities.
And some blame the scarcity on the fact that too many similar companies were funded during the dot.com boom, competing so fiercely that consumers received most of the benefits.
But it is important to keep in mind that an IPO is not the only way for a VC to exit an investment. VCs also exit by selling their portfolio companies. Nevertheless, the increase in merger and acquisition exits did not offset the decline in IPOs.
As a result, we suspect there is more upside than downside for the VC vintages of 2001 to 2007. According to informal sources, new legislation is probably less costly and more manageable than it was in 2005 and 2006. There are
more boutique investment banks with incentives to market IPOs. And, as we mentioned earlier, recent reports suggest there is now a larger pipeline of IPO candidates.
As we write this, commitments to US VC partnerships appear to be historically low in 2009. In 2009, Private Equity Analyst reported commitments of about $13bn (€10bn) to US VC funds.
Compared with the value of the stock market at the beginning of 2009, commitments are only 0.111% against the historical average of 0.138%. Measured relative to the stock market at the end of the year, the 2009 commitments are even lower, at 0.086%, compared with the historical average of 0.125%. All indications are that commitments are likely to continue to be low into 2010 and possibly beyond.
Based on the historical relationship between commitments and performance, the low level of commitments suggests that returns to the 2009 and 2010 vintage years are likely to be relatively strong.
And there are other grounds for optimism about VC. The most compelling is the transformation of the US corporate research and development system. The central corporate R&D laboratory was a dominant feature of the innovation
landscape in the US for most of the 20th century. While the concept of the centralised laboratory originated in the German chemical industry, US corporations had adopted it with enthusiasm by the 1950s. These campus-like facilities employed thousands of researchers, many of whom were free to pursue fundamental science with little direct commercial applicability. Among the best-known were Bell Laboratories (with 11 Nobel laureates) and IBM Central Research (with five).
Beginning in the early 1990s, however, American corporations began fundamentally rethinking the role of these centralised research facilities. Reflecting both a perception of disappointing commercial returns and intensified competitive pressures, US companies undertook a variety of changes to these facilities. Notable among them were paring the size of central research facilities in favour of divisional laboratories and relying much more heavily on what has been termed "open innovation" – alliances with and acquisitions of smaller firms. To economists, however, these changes are not surprising.
Observers such as Michael Jensen have contrasted the incentives within corporate research facilities unfavourably with those offered by venture capitalists. Jensen suggests that had higher-powered incentives been offered, some of the poor performance of research-intensive firms would have been avoided. Consistent with this argument, Samuel Kortum and Josh Lerner found that venturebacked firms were about three times as efficient in generating innovations as corporate research.
This transformation suggests the demand for venturebacked firms is likely to increase in the medium and longer term. The model of growing companies for full or partial acquisition by larger firms – which has been standard practice for many years in the computer networking business, for instance – is likely to be a growing segment of venture activity in the years to come.
And given the fact that corporate research spending, both in the US and globally, is many times the magnitude of venture capital investment, the size of the opportunity is likely to be substantial.
The US VC model has been enormously successful over the past 30 years. During that time, the industry has consistently received commitments and invested at a pace of oughly 0.15% of the value of the overall US stock market.
Of course, there has been some variation in commitments nd investments around that mean-a variation that can be traced in large part to the recent returns of the industry.
As a general rule, higher returns have typically attracted more capital from investors. But the greater capital has put downward pressure on returns, which in turn has resulted in smaller capital commitments. And as less capital has predictably led to increased returns, we have seen another increase in capital commitments and investment-and hence the beginnings of a new cycle.
We see little that makes us believe the VC model has changed or is broken. As far as we can tell, we are now leaving a period with slightly above average capital and average to slightly below average returns for a period of well below average capital. We would not be surprised to see this followed, perhaps quickly, by a period of above average returns.
The quote reproduced here is extracted from an upcoming article entitled It Ain’t Broke: The Past, Present, and Future of Venture Capital, to be published in the Journal of Applied Corporate Finance, Volume 22, No 2, Spring 2010, a Morgan Stanley publication.