DATE: MAY 2, 1999


Introduction and the Evolution of Netscape

Implicit in the underwriting process of an Initial Public Offering (IPO) is an unusual conflict of interest that investment bankers face. On the one hand, investment banks, acting in the best interest of the issuing firm, should attempt to bring an IPO to market at the highest possible price. In doing so, they would also maximize the revenue that they realize from the offering. On the other hand, if investment banks set the offering price too high, they may not be able to place the issue; leaving them with unmarketable securities that must be sold at a loss on the secondary market (in the case of a firm commitment offering) .
Netscape approached the private equity market four times prior to their IPO. The firm was conceived from $3 million of private equity capital from Jim Clark. In 1994, Clark contributed an additional $1.1 million of equity capital. Subsequent to that, the venture capital firm of Kleiner, Perkins, Caufield & Byers contributed $5 million of capital. However, the largest round of private financing, $18 million, occurred in 1995 from Adobe Systems and five other media companies.
The private financing obtained by Netscape had numerous advantages. First, the terms of private debt and equity could be customized for individual investors and for Netscape itself. Netscape also benefited by avoiding costly registration and floatation costs. By avoiding registration, Netscape did not have to reveal confidential or proprietary information to the capital markets. Finally, private placements are advantageous in so far as financing terms are much simpler to renegotiate relative to public issues.
While private placements have significant advantages for issuing firms, the limited investor base for such financing is a major drawback. The illiquidity of private placements is an important concern for Netscape and other closely held firms. That is, private investors must be rewarded with a premium above the risk-adjusted return on invested capital to compensate for the relative illiquidity of their investment.
Subsequent to its fourth round of issuing private equity, Netscape began the process of preparing for a 1995 IPO. The advantages of public equity are clear:
? Issuing firms are provided with improved access to capital markets
? Shareholders benefit from improved liquidity
? Original owners can diversify their assets
? Monitoring and information are provided by external capital markets
? The issuing firm’s credibility with customers and stakeholders is improved

While public ownership has numerous advantages, there are significant drawbacks as well:
? Public equity offerings are extremely expensive
? Implicit and explicit costs of dealing with shareholders are high
? More information is revealed to competitors than would otherwise be the case it the issuing firm remained private

? Costs associated with reduced operating flexibility due to market pressures are significant

Bringing an IPO to Market

The IPO process begins with the filing of the registration statement to the SEC. The registration statement includes the issuing firm’s audited financial statements and a complete description of the firm’s business - products, prospects, and possible risks. The underwriters are legally responsible for ensuring that the registration statement discloses all relevant and material information about the firm. The underwriter is also responsible for forming the underwriting syndicate, marketing the stock, and allocating shares among syndicate members .
Marketing the issue typically involves a “road show”, in which the firm’s management and underwriters attempt to sell the IPO to institutional investors. Road shows also serve as a market barometer that enables the underwriter to gauge the demand for an issue. Thus, the road show will influence the offer price issue, the number or shares, and the allocation of shares to particular investors. This is called the book-building process, and it begins upon the SEC’s approval of the registration statement.
The key to success in a book-building effort lies in the use of a strategic pricing and allocation policy designed to offset the investor’s incentive to understate his or her interest in an IPO. By committing to favor investors who provide strong indications of interest with relatively large allocations of underpriced shares, the investment bank can simultaneously limit the distortion of investor’s incentives in bidding and so increase the level of proceeds the issuing firm can expect to generate from its IPO .

In the process of book building, Netscape’s lead underwriter, Morgan Stanley, proposed that Netscape increase the offering price of $14 per share to $28 per share due to the extent of over-subscription for the offering. This, in part, reflects the effectiveness of the book-building process. Investment banks that excel at book-building must have strong relationships with the most sophisticated investors in IPOs. Banks gain very little from marketing an issue to investors who have neither significant public information nor the market power to influence the demand decisions of other investors .
Investment banks, such as Morgan Stanley, that underwrite more and larger offerings can negotiate more favorable pricing terms for their IPOs because they have greater opportunities to “bundle” offerings. More specifically, Morgan Stanley was able to command a price above Netscape’s initial suggested price range because institutional investors want to secure large allocations of Morgan Stanley’s future offerings. Finally, an investment bank’s commitment to stabilize the price of its offerings subsequent to their issuance enhances their reputation as an underwriter and is closely related to the success of a book-building effort .
Theoretically, road shows provide a forum for institutional investors to probe the issuing firm’s management for its assessment of the firm’s prospects. However, in practice the firm rarely provides information beyond what has been disclosed in the prospectus. This is because the SEC may require revisions to the prospectus if the substance of a public statement is viewed as material (material in the sense that it could affect the value of the firm). Once such a statement is included in a prospectus, the issuer is liable for its accuracy .
The preliminary prospectus represents the outcome of the investment banker’s due diligence effort. Under Section 11 of the Securities act of 1933, the information provided in the prospectus also serves as the foundation for civil liability arising from omissions of material facts. After the preliminary prospectus has been circulated among potential investors, these participants are asked to provide non-binding indications of interest in the issue. Thus, the “book” is built from these expressions of demand and, based on the information in the book, the terms of the offering are finalized .
Despite the apparent opportunities for investment banks to overprice IPOs, empirical evidence suggests that most IPOs are underpriced (see table on following page). In the case of Netscape, although the initial offering price was revised from $14 per share to $28 per share, the $58.25 closing market price on the first day of trading indicates significant underpricing.

The Underpricing of IPOs

In a firm commitment offering, the cost of having unsold shares is borne directly by the underwriter. This provides sufficient incentive for investment banks to underprice IPOs. However, according to Ritter (1987) best efforts issues also tend to be underpriced. Clearly, the potential loss of reputation that an underwriter would bear in the event of a failed issue is an important driver for underpricing.
Baron (1982) analyzed a potential conflict of interest between underwriters and issuing firms arising from their differing incentives and the underwriter’s superior information with respect to market conditions. The underwriter’s incentive is to set the offering price low enough to ensure that all of the shares can be sold without much effort and without subjecting the underwriter to excessive risk. The underpricing of IPOs facilitates an easier and less risky underwriting process.
A study in 1989 by Muscarella and Vetsuypens suggests that Baron’s assertion is not complete. Muscarella and Vetsuypens examined the IPOs of 38 investment banks that took themselves public, and hence, did not suffer from differing incentives or information asymmetries as postulated by Baron. These investment banks underpriced their own issue by an average of 7%.
Tinic (1988) examined the underpricing phenomenon in the context of the potential liability with respect to investor lawsuits brought on by poor performance subsequent to an IPO. Tinic found that the degree of underpricing was much higher after 1933, following the introduction of the Securities Act of 1993. This supports the contention that on average underwriters underprice IPOs.
Some firms go public with the intent of placing a larger, more seasoned offering at some point in the future. In these instances, it is customary to underprice the initial offering so that investors will be more likely to buy a firm’s seasoned offering after making favorable risk-adjusted returns from the IPO .
Because the information they provide to investment banks will affect the ultimate issue price, investors have an incentive to misrepresent their true opinions of an IPO. Investors may want to appear pessimistic about the issuing firm in the hope of receiving a share allocation at a favorable price. This is often cited as a reason for IPO underpricing. However, Benveniste and Spindt (1989) concluded that the manner in which investment banks price and allocate share in an IPO when they use a book-building process lends itself to obtaining credible information from large institutional investors. More specifically, they assert that IPOs are priced and shares are allocated as follows:
? Investment banks underreact to information provided by investors when they price IPOs

? Investors who provide more favorable information are allocated more shares
Therefore, to receive a greater share allocation, investors must truthfully reveal their opinion of an IPO.
Hanley (1993) tested this hypothesis empirically. Her results indicated that when investors expressed strong demand for an issue, as was the case with Netscape, the investment bank tends to price the issue above the price range suggested in the prospectus. In the case where the opinions expressed by investors corresponded with the initial expectations of an investment bank, the issuance price of an IPO tended to be in the price range suggested in the prospectus. Finally, when investor demand was relatively low, IPOs were generally priced below the expected price range.

Market Timing and IPO Pricing

A “Hot Issue” market refers to a period over which a large number of firms undertake an initial public offering. If the hot issue period is demand driven, companies contemplating IPOs should consider delaying their offerings as competition for funds may cause them to receive lower prices than they would receive by waiting. Conversely, if the hot issue period is supply driven, IPOs could be better served by being undertaken immediately to capitalize on the greater supply of available funding. Whether demand or supply driven, the IPO underpricing trend persists.
Since hot issues are underpriced on average, excess demand for IPOs typically exists. If new issues are not underpriced, informed investors will not put in an order for an IPO. Consequently, uninformed investors will obtain their full allotment of shares, and will on average lose money. This is what is commonly referred to as the “winners curse.”
Investment bankers, wanting to broaden the appeal of their IPO shares, may therefore systematically underprice the issue to induce uninformed investors to buy them. It follows then, that riskier IPOs are more prone to the winner’s curse and tend to be more underpriced on average, than less risky IPOs.

The Netscape IPO and Valuation

On August 8, 1995, the eve of Netscape Communications Corporation’s IPO, the board of directors convened to review a proposal by Morgan Stanley – the lead underwriter of the issue. Morgan Stanley was recommending that the issuance price for Netscape’s shares be increased 100%, from $14 per share to $28 per share. This recommendation was in response to the overwhelming oversubscription for Netscape’s shares, which has already caused the underwriters to increase the share offering from $3.5 million to $5 million shares.
The valuation included herewith provides a basis for the $28 per share price of Netscape upon its IPO. As was discussed above, The August 9, 1995 closing price was $58.25 – nearly double its issue price. In retrospect, the IPO was significantly underpriced. However, the valuation assumptions necessary to justify a $28 per share price are as follows:
? The risk free rate is the prevailing medium-term treasury rate prevailing in August 1995

? The equity risk premium is 6% (consistent with new research this decade indicating that risk premiums have been steadily declining from historical norms subsequent to 1992)

? Prevailing corporate tax rate at the time was 40%

? The firm’s unlevered beta is 0.72 (using Microsoft’s beta as a proxy)

? Steady state growth from 2005 into perpetuity is 4% (less than the nation’s historical GDP growth)