The Certification Role of Large Block Shareholders in Initial Public Offerings: The Case of Venture Capitalists

Article excerpt

INTRODUCTION

There is considerable theory and evidence that initial public offerings (IPOs), on average, experience positive initial returns and that the market reacts negatively to seasoned equity offerings. This kind of market reaction is a direct implication of Leland and Pyle (1977) and Myers and Majluf (1984). Although Welch (1989), Allen and Faulhaber (1989), and Grinblatt and Hwang (1989) have presented signaling models that predict that intrinsically higher value firms can convey their private information through IPO underpricing to deter mimicking by lower value firms, there are several reasons that make the first party actions suspect. For example, Gale and Stiglitz (1989) show that initial public offering signaling models break down when issuers are allowed to sell equity more than once. If potential investors cannot distinguish among the quality of new equity offerings and place an average value on all issues, the lemons problem described by Akerlof (1970) occurs.

To provide an effective signaling mechanism for an equity offering, an issuer can hire or lease the reputation from a third party specialist. Booth and Smith (1986) model the certification role provided by investment bankers to reduce information asymmetry in new equity offerings. DeAngelo (1981), Beatty (1989), Titman and Trueman (1986), Johnson and Miller (1988), and Carter and Manaster (1990) examine how auditors and underwriters help resolve the asymmetric information inherent in the IPO process. Megginson and Weiss (1991) find that the initial return to venture-capital (VC)-backed IPOs is lower than that of non-VC-backed IPOs. Their evidence suggests that the presence of large block shareholders may provide a complement to underwriter reputation in reducing IPO uncertainty. In addition, Barry, Muscarella, Peavy, and Vetsuypens (1990) show that venture capitalists specialize their investments in firms to provide intense monitoring service. Their monitoring functions are recognized by the market through lower underpricing. Moreover, Gompers (1995a) finds that the monitoring function provided by venture capitalists is valuable both to investors and entrepreneurial companies. I distinguish this study from the earlier works by focusing on VC reputation. Using the equally weighted value of standardized venture size and age as a proxy for reputation, I find that IPO underwriting costs are associated negatively with VC reputation. The evidence is consistent with Booth and Smith's (1986) certification hypothesis.

Venture capitalists are active investors that are insiders in their portfolio companies. Usually they hold substantial equity positions, perform monitoring functions, and serve on company boards of directors. The success of a VC firm depends not only on the ability to identify and nurture its portfolio companies that are likely to be successful, but also depends on the ability to unwind its equity position so that returns from entrepreneurial effort are maximized. When a portfolio company goes public, VC firms usually play an important role in helping the company choose managing underwriters.(1) Venture capitalists may participate in meetings with representatives of underwriters and discuss the timing and terms of the offering. With equity at stake, venture capitalists have an incentive to take their portfolio companies public when the portfolio companies' equity valuations are high.(2) Thus, VC-backed IPOs provide a unique opportunity to explore the certifiCation role of large block shareholders in new equity offerings.

Consistent with the findings of Barry et al. (1990) and Megginson and Weiss (1991), my results show that VC backing reduces the magnitude of underpricing and that such backing significantly lowers the underwriting spread. In addition, I find that the level of IPO underpricing and underwriter spread are associated negatively with the level of VC reputation. The results are consistent with the implications of the certification hypothesis. …