Updates > Journal

Journal

Date

October 19, 2012

Author

Eric Ball (Class 16)|Eric Ball (Class 16)

Share To

What Can Venture Capitalists Learn from Academics?

Journal

Eric R. Ball, Class 16

Most business executives and academic researchers essentially exist in separate countries without a common language. With a foot in both worlds, I am working to bridge this divide and bring useful, actionable insights from academia to the executives who can use them. My business career has been varied, leading up to my current position as Senior VP of Finance at Oracle Corporation and personal venture investor, and earlier in my career I spent time earning my doctorate and teaching at three universities.This month, The Kauffman Fellows Press is releasing a book I co-wrote with Joseph LiPuma (from EM-Lyon in France) entitled Unlocking the Ivory Tower: How Management Research Can Transform Your Business. The book addresses the question, "What can executives and entrepreneurs learn from academics about running their business?" Faculty are rewarded primarily for writing peer-reviewed academic journal articles. When academics do consult for business or write for a broader practitioner-audience, they typically focus on a single idea coming from their own research, which can result in one article being unnecessarily stretched into a full-length book. On the other side, executives may respect academic thinking but consider it irrelevant to their rough-and-tumble set of daily challenges. Managers also lack the time to sort through a large academic literature, which can be both physically and linguistically challenging for them to access, to glean the fraction of content providing actionable insights for running their business. This is the situation that Joe and I seek to change.In this article, I ask the same question for venture capitalists: What subset of academic literature has results that are interesting, relevant, and actionable for practicing VCs? With input from several prominent finance faculty at the University of Chicago and Harvard, I have selected recent venture articles from academic literature to summarize here. As with any "greatest hits" collection, the selection is somewhat subjective.1As Joe and I have done in our book, for each reviewed article I provide a takeaway thesis, a summary, and a brief discussion of what it means for practitioners. I start with Kaplan, Sensoy, and Stromberg's2 investigation of whether venture capitalists will maximize return on investment more by focusing on the management team or on the business plan, for candidate startups seeking funding (this dilemma is commonly referred to as the "horse versus jockey debate," and has prominent adherents on both sides). I then turn to Nanda and Rhodes-Kropf's3 description of how venture capitalists do, and should, change their investing approach and goal depending on the overall macro level of funding of startups in general. Lastly I use an article I cowrote with Hsin-Hui Chiu and Richard Smith4 to address a debate about whether VCs can predict stock market movements when selecting times to take their portfolio companies public, or whether they instead simply react to recent stock market movements. I end with much briefer descriptions of six additional venture articles published in the past four years.

Steven Kaplan, Berk Sensoy, and Per Stromberg, 20095

"Should Investors Bet on the Jockey or the Horse? Evidence from the Evolution of Firms from Early Business Plans to Public Companies."

Thesis

At the margin, to maximize return investors should place more weight on the business than on the management team.

Summary

Kaplan, Sensoy, and Stromberg study 50 venture-backed startup companies to identify how company characteristics evolve over the life cycle of the startup. They find that the lines of business pursued remain surprisingly stable, while management turnover is substantial and related to the formation of assets that are easily transferable (i.e., alienable assets). The authors then replicate most of their results with a separate sample of 106 nonfinancial startup IPOs from the year 2004.Kaplan et al. measure seven broad characteristics of a sample of startups over time, from first business plan to three years after IPO: financial performance, line of business, points of differentiation, nonhuman capital assets, growth strategy, top management, and ownership structure. They seek to provide a systematic description of the early life and evolution of companies, address the horse-versus-jockey debate (business-versus-management) among venture capitalists, and explore the implications of data about the evolution of early-stage firms for academic theories of the firm (in particular, for theories that emphasize the difference between employees and nonhuman assets).In terms of the horse-versus-jockey debate, the authors note that Tom Perkins of Kleiner Perkins focuses on a company's technological position. Don Valentine of Sequoia assesses the market for the product (which prompted him to invest in Cisco despite a perceived weak initial team, because of a large and growing market). Arthur Rock, on the other hand, emphasizes the quality of the management team.6The authors' key findings include the following:

The sample companies experience dramatic growth in revenue, assets, and market value (even when they do not become profitable within three years of their IPO).
The core businesses of the companies are remarkably stable. Only one firm (2%) in the initial sample (and 7% in the followup sample) changes its core line of business, which "suggests that the firms' business idea or line of business is fixed and elemental at an early stage in a firm's life" (p. 77). This finding runs counter to the "widespread belief that it is common for firms to change their business lines" (p. 83). Kaplan et al. find that companies may expand their target market but do not change their line of business.
The stated importance of expertise declines over time. Half of companies stress management knowledge in their initial business plan, but only 16% do so by the time of the IPO.
The human capital (top management) of the companies changes substantially over time. For CEOs at the initial business plan, 72 percent remain CEO at IPO and 44% are still CEO three years post-IPO. For the next four top executives, 50% remain on the team at IPO and only 25% by the third year post-IPO. Further, relatively few departing executives go on to found new firms—a higher percentage repeat their experience working for a newer firm.
Firms with a higher percentage of assets that can be easily transferred (i.e., alienable assets), relative to physical assets or intellectual property, initially tend to have significantly higher management turnover. This finding suggests that specific human capital is more important early in a company's life, before transferable assets have formed.

Kaplan et al. show the importance to the venture investor of picking a good business. In particular, they note the following:

It is important to note that the results do not imply that good management is not important. The large equity incentives VCs provide to new management suggest that good management is valuable. However, the results suggest that poor or inappropriate management is much more likely to be remedied by new management than a poor or inappropriate business idea is to be remedied by a new idea. (p. 79)

Finally, this article has implications for academic theories of the firm. Some theories7 assume that a company will organize around nonhuman capital assets. Kaplan et al. find that such assets form early and remain fairly stable; as such, they represent the "glue" that holds companies together while the management team is more malleable. This finding is consistent with the resource-based view of the firm initiated by Wernerfelt8 and extended by Rajan and Zingales,9 whereby such non-employee resources are critical sources of strategic value (though in very early stages the founder may also be a critical resource). The stability of young companies' lines of business also supports the natural selection theories of Hannan and Freeman,10 who argue that the creation and replacement of management teams is more useful than the adaptation of those teams.

Discussion

The debate between VCs who emphasize the technology (e.g., Tom Perkins) or product market (e.g., Don Valentine), and those who emphasize the quality of the founding management team (e.g., Arthur Rock) is exactly the kind of disagreement that can be resolved with a disciplined analysis of data. I applaud Kaplan et al. for taking an important real-world issue and doing just that. Of course, both the business plan and the management team are relevant.11 Nonetheless, this article provides evidence that the importance of the founding management team tends to fade over time while the business plan exhibits more permanence. The implication for VCs is to tilt our venture due diligence a little more in favor of the business itself.

Ramana Nanda and Matthew Rhodes-Kropf, 201112

"Investment Cycles and Startup Innovation."

Thesis

Venture-backed companies who receive their initial investment in hot markets are less likely to go public, but those who do go public have a higher valuation. The flood of capital in hot markets creates investment in riskier and more innovative startups.

Summary

A variety of academic articles have explored the cyclical nature of venture investing, and seek to explain why there are "hot markets." Some suggest that investors have a herd mentality13 or that they lose discipline as "me-too" ventures get funded in heated markets.14 Others suggest that investors may rationally react to improved investment opportunities.15 In a different article published the same year, Nanda and Rhodes-Kropf16 argue that abundant capital allows investors to experiment more effectively, shifting the startups born in periods of abundance to a set that is more risky and innovative (and not necessarily better or worse).In this paper, Nanda and Rhodes-Kropf examine how the environment in which a startup is first funded is related to its ultimate outcome. They examine startups funded from 1980 to 2004, from the time of first financing to their exit or bankruptcy. They find that startups receiving funding in quarters when many other startups were also funded were less likely to go public (and more likely to go bankrupt) than those funded in quarters with less activity. However, among those startups successful in getting to IPO, those funded in active periods had higher valuations and a higher number of patents. Getting funded in boom times creates a wider variance of outcomes (both more negative and more positive), suggesting that such startups are more innovative.The authors then examine the behavior of investors who invest across different environments, and the results indicate that the risk tilt is not being driven by new, uninformed investors in hot markets. Rather, long-term investors are changing how they individually invest.They also investigate whether the correlation is driven by hot markets as a response to different investment opportunities involving a more novel supply of candidate companies, or whether the abundance of capital changes the demand by investors for the type of company they wish to finance. To that end, Nanda and Rhodes-Kropf seek a variable that leads to excess money but that is unrelated to investment opportunities. They note that the supply of capital is influenced by the success of prior investments by investors, such that future fundraising depends on the past incidence of IPOs. Knowing the fundraising occurring at a particular point in time makes it possible to predict investment activity 2-3 years later. Since IPOs tend to be from companies that are at least four years old, these IPOs are unlikely to be systematically related to the arrival of new opportunities. The authors use the number of IPOs 3-4 years in the past as an instrumental variable for the number of investments made in a given quarter. This instrument captures the part of investments that are due to increases in capital unrelated to investment opportunities for venture funds. The results suggest that the excess capital in a hot market actually changes the type of startup that venture investors will fund, toward companies that are more innovative and have a wider possible range of outcomes.Finally, the authors determine that the companies funded in hot markets tend to generate a higher number of patents and citations than companies funded in other times. This finding suggests that the companies are actually more innovative, and not simply riskier. In addition to explaining venture investing behavior across the cycle, their results suggest that such cycles play a central role in diffusing and commercializing technologies in the economy. That is, "financial market investment cycles may create innovation cycles" (p. 4). The policy implication is that such cycles are positive, and not a phenomenon to be discouraged.

Discussion

This paper indicates that even the most experienced individual venture capitalists change their approach to invest in riskier and more innovative startups at times when more capital is being deployed to startups in general. This lack of evenness in financing over time may not be a problem, as it has historically been viewed, but rather may play a beneficial role in the creation of new technologies. Financial market cycles may produce innovation cycles.As an objective effort to describe what VCs appear to be doing, this paper is not written as a normative piece to tell VCs what they should do. To the extent that VCs are rational economic agents, however, Nanda and Rhodes-Kropf's work highlights how it may be rational to reevaluate one's risk tolerance in quarters when VCs find themselves competing for a fixed supply of deals among an expanded supply of capital from competing VCs. In baseball parlance, it may make sense to hit for singles when capital is constrained and to hit for home runs (with more strikeouts) when capital is abundant.Some VCs stick to a fixed approach for evaluating deals. In the United States, the 2012 MoneyTree report shows that venture deal funding has fluctuated, from a peak of $105 billion in 2000 to less than $20 billion in 2003, rising to $32 billion by 2007, dipping to $23 billion by 2010 after the financial crisis, and recovering to $30 billion in 2011.17 I can say that my own venture investing risk tolerance has not moved with these fluctuations. These results suggest that it makes sense to reconsider one's own risk tolerance depending on how much money limited partners in the aggregate are allocating to venture in a particular period.

Eric Ball, Hsin-Hui Chiu, and Richard Smith, 201118

"Can VCs Time the Market? An Analysis of Exit Choice for Venture-Backed Firms."

Thesis

When selecting the time for IPOs for their portfolio companies, venture capitalists react to stock market runups but cannot predict future stock price movements.

Summary

In this article, Chiu, Smith, and I examine what factors influence the choice of exit for startups between going public and being acquired.19 Similar to Nanda and Rhodes-Kropf's20 description of cyclicality in venture financing, this article explores similar cyclicality in IPO markets. IPOs tend to occur in waves, so that the probability of one company exiting by going public depends on how many other companies have recently gone public.Some researchers have concluded that venture capitalists have some ability to determine when the market is at a peak, and take their portfolio companies public before a downturn. Lerner21 finds a negative correlation between the probability that a biotech startup goes public and the return to a biotech sector stock index in the three months following the IPO, and interprets this negative correlation as evidence that decision-makers anticipate a market decline. Schultz22 demonstrates that, even in an efficient market, underperformance of stock returns after equity offerings should be expected. In his model of "pseudo-market timing," companies are more likely to go public after the market has increased above a trigger price, so that investors appear to have timed the market with clustered offerings at market peaks even if they cannot predict market behavior.Figure 1 shows the quarterly numbers of IPOs and acquisitions in the United States, demonstrating that IPO activity goes through hot and cold waves while acquisitions tend to exhibit steady growth. The proportions of IPO and M&A exits vary significantly from year to year. For example from 1995–2000, IPOs represented more than 50 percent of exits, whereas in 2008–2009 they accounted for fewer than 5 percent of U.S. exits.23 If investors are seeking the highest return, this change suggests that the relative value of exiting through one means or the other fluctuates materially over time.

Figure 1. Number of Exits from Venture-Backed Firms via IPO and Acquisitions and the NASDAW Index by Quarter.Reprinted with permission from Ball et al., "Can VCs Time the Market?", p. 36.

In this article, we examine the exit choice of 8,163 venture-backed U.S. startups from 1978–2009, and correlate this choice with the performance of the broad stock market and sector stock performance for the year before and the year following exit. We test a market-timing hypothesis against the alternative possibility that exit choices are driven by capital market conditions. We find support that startups go public in reaction to prior stock market runups, rather than in prediction of a downturn.Besides pre-exit and post-exit stock market performance, we also find other economic variables with a statistically significant impact on the probability that exit choice takes the form of IPO over acquisition: recent change in gross private domestic investment, recent change in the number of patents granted, recent change in consumer confidence, recent one-year treasury bill interest rate, recent change in cash flow to equity funds, proportion of equity capital among capital raised, and the pre-money valuation of the startup.One surprising result is that the propensity to select an acquisition exit and the level of acquisition exit activity are negatively related to subsequent market returns, but this result fades in models that control for market conditions. This finding may indicate that acquisition activity increases when other investment opportunities for acquirers diminish. The negative subsequent returns suggest that the market is slow to recognize this lower value of growth activities.

Discussion

A large debate within the academic finance community since the 1960s has been whether markets are efficient. That is, can investors use existing information to outperform other investors in predicting the future price of stocks. Market efficiency does not mean that a stock or market is priced accurately, but it does mean that the odds are equal that the price is too high as that it is too low. I tend to agree with Ken French24 who acknowledges that the recent bubbles and crises show that significant mispricing can occur, but emphasizes that mispricing does not mean easy profits and recommends investors should act is if prices are right. However, I also agree with Jay Ritter25 who cites major outlier events like the stock market bubble of the late 1990s as evidence that "the market gets the little things right, but sometimes gets the big things wrong" (p. 2).Lerner's work suggests that sophisticated investors like venture capitalists might be better able to predict the stock market (at least in the biotech sector where startups have a longer life before exit and more flexibility in selecting their IPO date).26 The analysis summarized above looks at multiple sectors (biotech and others) across a longer time horizon and finds no evidence that venture capitalists are better at forecasting stock markets. The statistical evidence is consistent with venture capitalists behaving somewhat like Pavlov's dogs, taking their portfolio companies public any time the stock market rises and often doing so in groups. We also discover evidence that macroeconomic variables play a role in exit choice, identifying several macro variables which, whenever they become high relative to a longer-term average, tend to be associated with increased IPO activity.Any individual venture capitalist must take market conditions as a given, and typically thinks of the IPO window as being "open" or "closed." This analysis provides some insight into what economic and stock market factors may be influencing whether the market is open, which may help VCs evaluate whether they want to bring a portfolio company public at a particular point in time—regardless of whether many or few other startup companies are also going public.

Conclusion

Kurt Lewin captured the contribution that academic analysis can make to practitioners when he said that "there is nothing more practical than a good theory."33 However, rigorous academic analysis only impacts practice to the extent that it is communicated. I would add solid empirical analysis as a second type of research that can inform practice. Academic advances do eventually find their way to practitioners, but mostly through placement in textbooks read by MBA students who are not in a position to set policy for their firms until twenty years later. My hope is that we can shorten this lag.Communication of useful theory and analysis is the motivation for this first summary of faculty venture research to appear in the Kauffman Fellows Report. I invite readers to provide feedback (how useful is this type of summary to you in your jobs?) as well as suggestions for other research that you find relevant to practice. We have an opportunity to conduct our own experiment to determine if monitoring the literature can improve our returns on venture investing over the next decade.

Eric Ball

Eric is Senior VP and Treasurer for Oracle, a $37 billion provider of optimized and integrated business hardware and software systems. Eric was named in 2011 as one of the "100 Most Influential People in Finance." He leads a team of 22 in managing the company's worldwide treasury operations, capital markets activity, investments, and venture portfolio. Eric serves on the Boards of Oracle Japan and the Richard Lucas Medical Foundation, and has worked at Flextronics, Cisco Systems, Avery Dennison, and AT&T. He is an investor in eight startup companies. Eric has a BA from the University of Michigan, an MA/MBA from the University of Rochester, and a PhD from the Drucker-Ito School of Management. He lives in Menlo Park with his wife and two sons.

1 Some selections were unanimous, while others arose from our unabashedly allowing voting for one's own work. I welcome disagreements and suggestions for what should have been included—if a review of academic literature becomes a recurring contribution to future issues of the Kauffman Fellows Report, future articles will hopefully be based on votes from more faculty and practitioners regarding the academic contributions most relevant to active venture investors.2 Steven Kaplan, Berk Sensoy, and Per Stromberg, "Should Investors Bet on the Jockey or the Horse? Evidence from the Evolution of Firms from Early Business Plans to Public Companies," Journal of Finance 44(2009): 75115, doi: 10.1111/j.1540-6261.2008.01429.x.3 Ramana Nanda and Matthew Rhodes-Kropf, Investment Cycles and Startup Innovation, Harvard Business School working paper 12-032, (2011).4 Eric Ball, Hsin-Hui Chiu, and Richard Smith, "Can VCs Time the Market? An Analysis of Exit Choice for Venture-Backed Firms," Review of Financial Studies 24, no. 9 (2011): 31053138, doi:10.1093/rfs/hhr042.5 See note 2 for full bibliographic details.6 Arthur Rock's view is captured by David Gladstone and Laura Gladstone: "You can have a good idea and poor management and lose every time. You can have a poor idea and good management and win every time" (Venture Capital Handbook [New Jersey: Prentice-Hall, 2002], 9192); Tom Perkins's (and probably Don Valentine's) view is better captured by Warren Buffett: "When a management team with a reputation for brilliance tackles a business with a reputation for bad economics, it is the reputation of the business that remains intact" (Jerry Useem, "America's Most Admired Companies," Fortune, 7 March 2005, para. 1).7 e.g., Oliver Hart, Firms, Contacts, and Financial Structure (Oxford: Clarendon Press, 1995); Oliver Hart and John Moore, "A Theory of Debt Based on the Inalienability of Human Capital," Quarterly Journal of Economics 109, no. 4 (1994): 841879; Bengt Holmstrom, "The Firm as a Subeconomy," Journal of Law, Economics, and Organization 15 (1999): 74102, doi: 10.1093/jleo/15.1.74.8 Birger Wernerfelt, "A Resource Based View of the Firm," Strategic Management Journal 5(1994): 171180.9 Raghuram Rajan and Luigi Zingales, "The Influence of the Financial Revolution on the Nature of Firms," American Economic Review 91, no. 2 (2001), 206212.10 Michael Hannan and John Freeman, "Structural Inertia and Organizational Change," American Sociological Review 49, no. 2 (1984), 149164.11 In my personal venture co-investing with a prominent Silicon Valley investor, we have in fact tended to prioritize the management team, while acknowledging the common need to replace the founding CEO as the business evolves.12 For full bibliographic details, see note 3.13 David Scharfstein and Jeremy Stein, "Herd Behavior and Investment," American Economic Review 80, no. 3 (1990): 465479.14 Udayan Gupta, Done Deals: Venture Capitalists Tell their Stories (Boston: Harvard Business School Press, 2000).15 Paul Gompers, Anna Kovner, Josh Lerner, and David Scharfstein, "Venture Capital Investment Cycles: The Impact of Public Markets," Journal of Finanancial Economics 87(2008): 123; Lubos Pástor and Pietro Veronesi, "Rational IPO Waves," Journal of Finance 60, no. 4 (2005): 17131757.16 Ramana Nanda and Matthew Rhodes-Kropf, "Financing Risk and Bubbles of Innovation" (Harvard Business School working paper 11-103, 2011).17 PricewaterhouseCoopers and the National Venture Capital Association (based on data from Thomson Reuters), "Historical Trend Data," The MoneyTree™ Report, 2012 Q2.18 For full bibliographic details, see note 4.19 This article expanded and further developed my 2008 doctoral dissertation in management (with Richard Smith as committee chair), Eric Ball, Does Market Timing Affect the Exit Choice for Venture-Backed Startups? (Doctoral dissertation, Claremont Graduate University, Drucker-Ito School of Management, 2008).20 Nanda and Rhodes-Kropf, "Investment Cycles and Startup Innovation."21 Josh Lerner, "Venture Capitalists and the Decision to Go Public," Journal of Financial Economics 35, no. 3 (1994): 293316.22 Paul Schultz, "Pseudo-Market Timing and the Long-Run Performance of IPOs," Journal of Finance 67(2003): 17951828.23 The Thomson VentureXpert database shows that from Q1-1995 to Q1-2000, there were 1,051 IPOs compared with 951 acquisitions. This is in contrast to Q1-2008 through Q4-2009, during which there were 19 IPOs compared with 612 acquisitions.24 Eugene F. Fama and Kenneth R. French, "Q&A: The Limits of Arbitrage," Fama/French Forum, 14 July 2010.25 Jay Ritter, "Introduction," in Recent Developments in Corporate Finance, ed. J. Ritter (Williston, VT: Edward Elgar Publishing, 2003).26 Lerner, "Venture Capitalists."27 Gompers et al., "Venture Capital Investment Cycles."28 Steven Kaplan, Mark Klebanov, and Morten Sorensen, "Which CEO Characteristics and Abilities Matter?" Journal of Finance 47(2012):9711005.29 Michael Ewens and Matthew Rhodes-Kropf, "Is a VC Partnership Greater than the Sum of its Partners?" (Harvard Business School working paper 12-097, 2012).30 Robert Harris, Tim Jenkinson, and Steven Kaplan, "Private Equity Performance: What Do We Know?" (Fama-Miller working paper, 2012).31 Paul Gompers, Anna Kovner, Josh Lerner, and David Scharfstein, "Performance Persistence in Entrepreneurship," Journal of Financial Economics 96(2010): 1832, doi:10.1016/j.jfineco.2009.11.001.32 Marco Da Rin, Thomas Hellmann, and Manju Puri, "A Survey of Venture Capital Research," in Handbook of the Economics of Finance, vol. 2, ed. G. Constantinides, M. Harris, and R. Stulz (Amsterdam: North Holland, 2010).33 Kurt Lewin, Field Theory in Social Science: Selected Theoretical Papers (London: Tavistock, 1952), 169.

Journal

What Can Venture Capitalists Learn from Academics?

Eric R. Ball, Class 16

Steven Kaplan, Berk Sensoy, and Per Stromberg, 20095

Thesis

Summary

Discussion

Ramana Nanda and Matthew Rhodes-Kropf, 201112

Thesis

Summary

Discussion

Eric Ball, Hsin-Hui Chiu, and Richard Smith, 201118

Thesis

Summary

Discussion

Further Reading

Paul Gompers, Anna Kovner, Josh Lerner, and David Scharfstein, 200827

Steven Kaplan, Mark Klebanov, and Morten Sorensen, 201228

Michael Ewens and Matthew Rhodes-Kropf, 201229

Robert Harris, Tim Jenkinson, and Steven Kaplan, 201230

Paul Gompers, Anna Kovner, Josh Lerner, and David Scharfstein, 201031

Marco Da Rin, Thomas Hellmann, and Manju Puri, 201032

Conclusion

Eric Ball

Related

Doing Venture Capital In The Age Of Product-First Companies

New Trend In Venture Capital: The Discovery Round

The State of Technology and Culture in Africa

Kauffman Fellows Summit Comes to Canada