Printer Friendly

SWORD financing of innovation in the biotechnology industry.

This paper analyzes a financing arrangement known as stock warrant off-balance-sheet research and development ("SWORD") that has been used recently in the biotechnology industry. A SWORD, as the industry calls this arrangement, consists of a set of contracts that defines both the relationships among the parties to a SWORD and the property rights accruing to each party. The primary objective of this paper is to show how a SWORD can promote innovation in a manner not possible when more conventional financing arrangements are used. Since a SWORD is designed to facilitate R&D, the various contracts are described first. Next, the rationale for considering a SWORD to be a form of (off-balance-sheet) project financing is provided. Then, a comparison with conventional capital budgeting is used to show how a SWORD provides flexibility in making investment and financing decisions that can enable a biotech firm to engage in R&D that otherwise might not be undertaken. However, as carefully crafted as a SWORD is, with the intent to balance the interests of all parties, it cannot eliminate the favorable informational asymmetry possessed by biotech managers. This asymmetry is shown below to confound the interpretation of the benefits of a SWORD. Another objective of this paper is to examine the stock market effects of the first SWORD announcements to determine whether investors have responded positively or negatively to this arrangement.

Innovation is important to biotech firms because it provides the competitive edge necessary for survival. However, uncertainty about the commercial viability and regulatory approval of new products and technologies makes innovation a very risky undertaking. Also, conventional internal financing of innovation is generally not possible because biotech firms tend to be small with meager profits and few cash resources. Total assets average $3.5 million for the smallest firms in the 1989 Ernst and Young biotech survey (60% of the respondents) and $13.1 million for the "midsize" firms (22% of the respondents). Only 15% of the firms report positive cash flow from operations in 1989, and 38% indicate that cash on hand will support current spending levels for less than 12 months. Biotech firms are forced to turn to external sources for considerable amounts of funds for innovation: on average, biotech firms plan to raise $55 million by the end of the decade in new external financing.(1)

Financing pressures have often led biotech firms to perform contract research for the larger, better-financed pharmaceutical companies. However, proprietary control over any developed products or technologies is transferred to the pharmaceutical sponsor, who then brings the products to market. Even though such an arrangement solves the funding problem, it can be a barrier that keeps the biotech firm from developing its own manufacturing and marketing capabilities. The biotech firm could be successful in innovation yet still be relegated to the role of a research house. Understandably, biotech managers report that R&D funding is the most crucial business issue they face.(2)

Since 1988, several biotech firms have used a SWORD to set up a new, separate organization: (i) that owns the property rights to the R&D; (ii) whose financial statements have no impact on those of the biotech parent; and (iii) whose financing comes from a public units offering. Each unit consists of one share of the new venture common stock that can be called by the parent, and one warrant to purchase one share of common stock of the parent firm.(3) In essence, each biotech firm has created its own research sponsor and has retained the option to buy out this sponsor by calling the new venture common stock.(4)

Since the new venture funds are not commingled with assets in place, the parent's existing shareholders avoid dilution of value if the innovation fails; a failed equity-financed R&D project could lead to substantial dilution. However, because new investors are less informed about the R&D than the biotech managers and possibly face adverse selection, they are unlikely to participate without a signal about the value of the R&D. The warrants signal that the existing shareholders are willing to share the value of the parent's assets in place even if the innovation fails. This signal is costly because the warrants also lead to equity dilution if exercised. Nevertheless, as shown below, a SWORD produces less dilution than does conventional capital budgeting.

Perhaps the ultimate test of the SWORD arrangement is that presented by the stock market. While the small sample examined here is not statistically reliable, it does provide an opportunity to measure a SWORD's usefulness. The stock market reaction generally is positive for each SWORD, allaying fears that the potential for adverse selection renders the SWORD valueless. This positive reaction is consistent with that found by Chan, Martin, and Kensinger |3~, who provide similar evidence for the R&D decisions of high-technology firms, and McConnell and Muscarella |8~, who document positive announcement effects for unexpected increases in capital expenditure decisions. Thus, issuing equity is not always an indication of the adverse selection problem identified by Myers and Majluf |11~. Even in the Myers and Majluf framework |11, p. 194~, large returns from a highly successful investment benefit both current and new shareholders. Asquith and Mullins |1~, echoed by Smith |13, p. 7~, state that a positive stock price effect is consistent with a favorable information effect associated with an investment by the firm. The findings of Masulis and Korwar |7~ and Mikkelson and Partch |9~ regarding capital expenditures provide some support for such a positive information effect.(5) In light of the findings of these other studies, the interpretation offered here for the positive announcement effects is that a SWORD is a viable technique for financing valuable biotech R&D.

The remainder of this paper is organized as follows. Section I provides details of the contracts that make up the SWORD. Section II evaluates the SWORD in terms of project financing, while Section III compares the SWORD to conventional capital budgeting. The problem of asymmetric information is discussed in Section IV, as is the impact of adverse selection on the parent's stock price. The stock market reaction is analyzed in Section V, and Section VI provides some concluding comments.

I. Organizational Design

This section describes the basic features of a SWORD, and Exhibit 1 displays some information about the SWORDs examined here. (These SWORDs produced four parent and new venture combinations: ALZA and BioElectric Systems, Inc. (BES); Centocor and Tocor, Inc.; Immunex and Receptech Corp.; and Genzyme Corp. and Neozyme Corp.(6)) Appendix A provides some background on the parents and the new ventures. Next, the contracts that define a SWORD are described. They are: (i) the technology license agreement, (ii) the services agreement, (iii) the development contract, (iv) the license option agreement, and (v) the financing arrangements.

(i) Technology License Agreement. Through this agreement, each new venture is granted, by the parent, a worldwide, exclusive, royalty-free license in perpetuity, to use the technologies associated with the development of the products. Each parent has completed some basic research on the technologies and transfers this technology to the new venture. The value of each new venture stems from the property rights associated with the license, and the new venture maintains a perpetual claim on the future stream of operating income generated by the technologies.

(ii) Services Agreement. The parent provides management and administrative services to the new venture throughout the development process. The new venture has minimal need for technical and operating managers because, in a sense, it is nothing more than a financial intermediary: the funds from the units offerings are invested in short-term securities and paid over time to the parent for R&D expenses, as specified in the development contract.

(iii) Development Contract. Each parent performs the actual development at existing facilities. The new venture benefits from the parent's research expertise without having to secure and train technical specialists or construct duplicate laboratory facilities. The parent will undertake the patent approval process for products or technologies that appear to be commercially viable.

(iv) License Option Agreement. This agreement contains a provision similar to the exercise provision embedded in a financial option. The parent is granted a (real) option by the new venture to acquire a worldwide license to manufacture, use, and market any developed products in exchange for the royalty payments described in Panel B of Exhibit 1. The new venture really represents a bundle of real options since the parent can exercise its right to license or sublicense any developed products or technology on a product-by-product basis. Centocor, in order to maintain its exclusive license, also must make a $15 million "milestone payment," in cash or Centocor common stock, for each individual product that receives FDA approval. Immunex similarly must pay a "license option exercise price" in cash or stock of $29 million on or before January 31, 1993, per development program.

(v) Financing Arrangements. Each SWORD unit contains one share of callable common stock and one warrant to purchase one share of the parent's common stock. Panel C of Exhibit 1 describes the timing and amount of the purchase option (i.e., call) prices, paid in cash or common stock of the parent at the discretion of the parent, and Panel D provides some details about the warrants. The parent can reacquire the property rights to the developed technologies and products only by calling the new venture shares. If competitors, large or small, were to purchase the new venture's shares, the parent could call the stock and dismantle the new venture.(7) The warrants protect the new venture investors from having the parent expropriate wealth by refusing to call the new venture shares. By exercising their warrants, the new venture shareholders can share in the non-new-venture aspects of the parent's income stream while continuing to receive royalty payments. Even if their stock is called, the new venture shareholders can continue to participate in the returns from the innovation by exercising their warrants. Since both current and new venture investors maintain claims on the parent, their incentives appear to be aligned in a SWORD.

In total, a SWORD is a set of contracts that cut across both the real and financial sectors. Exhibit 2 illustrates the relationships generated by both conventional capital budgeting and a SWORD. In Panel A, a firm using conventional capital budgeting undertakes a new investment by incorporating it internally and finances it by selling new securities to investors. Panel B of Exhibit 2 shows that a SWORD is more complex: (i) during the R&D phase, property rights are transferred to the new venture through the technology license, and funds flow from investors through the new venture and on to the parent via the services and development agreements; (ii) during the commercialization phase, the parent receives a license option on the R&D outcomes and pays royalties to the new venture on successful products; and (iii) the warrant and callable new venture common stock allow both the parent and new venture shareholders to maintain claims on the R&D outcomes.

Building from this description of a SWORD, the following two sections explain why a SWORD is useful. The off-balance-sheet feature of SWORD financing is examined next in the context of project financing. The subsequent section shows how a SWORD can encourage a biotech parent to undertake innovation that might be foregone under conventional capital budgeting.

II. Relationship to Project Financing

When investment is undertaken internally by a sponsoring organization, new funds are commingled with those generated by assets in place. Project financing produces a separate legal entity, so no commingling occurs. Because a SWORD's new venture is a corporation that owns the rights to the technology and holds the R&D funds, a SWORD is a form of project financing. However, a SWORD is financed totally by equity, while project financing typically uses high levels of debt. Mao |6~ and Shah and Thakor |12~ point out that by segregating funds, creditors can more effectively monitor a project. The consequent pricing effect of the monitoring leads the firm to choose relatively high proportions of debt financing for projects that are inherently high risk. For Mao |6~, the creditworthiness of the project and the secured, prior-claim position of their claim are of chief concern for a project's creditors. He suggests that project financing is suitable for large, capital-intensive projects such as pipelines and oil refineries. Such projects involve tangible assets that: (i) serve as collateral, (ii) embody tested technology, and (iii) generate operating cash flows. Shah and Thakor |12~ examine capital structure implications, allow for cost differentials between information revelation by managers and monitoring by creditors, include the benefits of debt-related tax shields, and downplay the importance of bankruptcy costs. Shah and Thakor show that higher project risk leads to higher degrees of leverage.

Tax shields also are a driving force in the use of debt in project financing. However, since the ability to service debt is not reliable, these tax shields must be foregone in a SWORD. Given the long lead times for product development and regulatory approval, a debt-financed new venture might default on interest payments at some point, even if the R&D ultimately is successful. Bankruptcy proceedings, if initiated, would be unlikely to enable creditors to recover all of their original investment. The collateral is basically a depleting supply of cash, since the development and services agreements assure that a SWORD has little need for tangible, income-producing assets. In a SWORD, bankruptcy costs are avoided because only equity financing is used. Indeed, debt financing runs counter to the intent of a SWORD. Funds are meant to be disbursed to the parent for R&D expenses and are not meant to be returned to the investors as interest payments.

A SWORD's proper role can be seen most clearly if investment opportunities are viewed as a continuum of activities from research through development to capital expenditures. At one end of the continuum, Mao's project financing is like a capital expenditure decision in which tangible assets are chosen from a menu of existing, tested technologies. The purpose of a SWORD is to add new items to the menu; a SWORD project has yet to reach the stage where capital expenditures are needed. The new venture owns the property rights (in perpetuity) to intangible assets (i.e., technological innovation) that have been obtained in exchange for royalty payments that will occur only after the technology becomes "tested." Such intangible assets provide little security for potential creditors regardless of the priority of the associated claim.

At the other end of the continuum, a SWORD seems to be appropriate only on the development side of R&D. Because research ideas are even less tangible than product development, a research-oriented SWORD would be two steps from the manufacturing and marketing stage. Notably, in the SWORDs examined here, each parent had completed some basic research before creating the new venture. The parent, perhaps in conjunction with the SWORD investors, possibly could use leveraged project financing to commercialize the output of any successful innovation. Thus, a SWORD appears to carve out a niche as a special, all-equity form of project financing for R&D that is nearing the implementation stage.

III. Relationship to Conventional Capital Budgeting

A biotech firm could undertake innovation internally through conventional capital budgeting. Using a "real option" framework (Myers |10~), this section shows that SWORD financing provides more flexibility to the biotech parent than does conventional capital budgeting. Myers breaks the value of the firm into the present value of assets in place (Va) and the value of growth opportunities (Vg):

V = Va + Vg. (1)

Vg is a real option to make future (discretionary) investment in the real sector.(8) Vg will vanish if not exercised because real options are firm-specific or are traded in thin and imperfect secondary markets. For a firm with N shares outstanding, the stock price, P, before innovation, is Va/N. If the real option is undertaken as a capital budgeting TABULAR DATA OMITTED project and M new shares are issued,(9) the resulting stock price is:

P|Prime~ = (Va + Vg)/(N + M). (2)

The new shareholders now have a claim on assets in place equal to Va*(M/(M + N)). Dilution will be avoided only if Vg equals or exceeds Va*(M/N). If a SWORD is used and the warrants also represent M new shares, with one warrant buying one share, the stock price is:

P|double prime~ = |Va + Vg + E - X - O~/|N + M~ (3)

where E = the warrant exercise price; X = either PO, the purchase option price to call the new venture shares, or R, royalty payments to the new venture from the license option agreement; and O = other payments to the new venture such as milestone payments. Either PO or R is paid; both will not be paid at the same time. Equations (2) and (3) clearly indicate that Vg has a direct bearing on whether the current shareholders will avoid dilution of wealth.

Exhibit 3 compares the stock price effect of capital budgeting and a SWORD for various levels of innovation success. The benchmark price is always P = Va/N. P|double prime~ is also compared to P|prime~ in the same row. In the exhibit, any superiority the SWORD has (i.e., higher stock price) stems from greater flexibility. Exhibit 3 shows that M new shares always are issued under conventional capital budgeting and only when Panel C is obtained ex post does P|prime~ exceed P. However, managers using the SWORD can decide ex post to: (i) in Panel A, allow the real option to expire; (ii) in Panel B, either allow expiration or license the real option and pay royalties (R) and other payments (O); or (iii) in Panel C, either license or reacquire the property rights by calling the new venture shares (paying PO). If the warrants are not exercised, this decision is made with no increase in shares outstanding.

Even if the warrants are exercised when they are "in-the-money" (i.e., P |is greater than~ E), the payment to the firm of E per warrant limits dilution relative to capital budgeting; P|double prime~ is generally greater than P|prime~ in Exhibit 3. Comparing the first and third columns of Panel A shows this most directly. It is not obvious from Exhibit 3 why the warrants would be exercised if the innovation is worthless, but the new venture shareholders do maintain a claim on the parent's assets in place that is independent of the outcome of the R&D. A SWORD's flexibility comes with a cost -- dilution -- that occurs when the warrant is exercised: in the second column of Panel A, P|double prime~ is identical with the benchmark price P, but in the third column, P is higher.

If conventional capital budgeting were the only investment vehicle available, most biotech firms would be unlikely to undertake innovation independently. Krasker |4~ shows that straight equity issues lead to dilution of value and that this dilution increases as the size of the offering increases in relation to the size of the issuing firm. Since the planned spending on R&D reported above is very large relative to the size of most biotech firms, Krasker's results suggest that the stock price decline would be so severe that many biotech firms would be unable to raise these amounts through straight equity offerings. Thus, biotech firms often have become research houses for larger companies. While conventional capital budgeting is like an ex ante commitment to the untested technology, the SWORD's license option agreement permits the commitment to be made ex post. With a SWORD, the biotech firm has greater control over the developed technology and its own future survival, and can undertake innovation that might otherwise be foregone.

IV. Adverse Selection

The previous section has shown that the biotech parent can benefit greatly from a SWORD. This section examines whether these benefits will necessarily be reflected in the stock market reaction to the announcement of a SWORD, especially when informational asymmetries are taken into account.

New venture investors face a potential adverse selection problem because the parent's managers possess information about Vg at the outset while the new venture investors observe only the actual (ex post) Vg. Nothing in the SWORD contracts can assure potential investors that the parent is not moving the least viable and/or most risky projects off its balance sheet. Fearing such behavior, new investors would consider the new venture shares to be of limited value and would value the units through the warrants. Current shareholders also would be affected by the informational asymmetry since the SWORD has become an equity offering that will be completed when the warrants are exercised. In the Myers and Majluf |11~ analysis, equity financing leads to a stock price decline because managers selecting equity financing are signalling that the real option has less-than-favorable outcomes. Because a SWORD does not eliminate the underlying informational asymmetry, the potential for adverse selection suggests that a SWORD would lead to a decline in the parent's stock price.

The warrants are intended to: (i) align the incentives of the investors and managers, and (ii) signal that the new venture has value because the current shareholders are willing to share the value of assets in place. The warrants can be thought of as a modified version of the Leland and Pyle |5~ signalling mechanism. In Leland and Pyle |5~, the manager's willingness to maintain a claim on the investment outcomes certifies that investment's worth; here, a claim against the assets in place substantiates the value of the new investment. The signal must be costly if it is to have merit, and the dilution that occurs in the future when the warrants are exercised represents this cost. Thus, rather than being an indication of adverse selection, the warrants simply entail a cost that current shareholders must bear if they want the R&D to proceed.

The new venture investors may suspect that the premium reflected in the warrant exercise price leaves the SWORD units overvalued.(10) Ultimately, the new venture shareholders bear the residual risk of the innovation. As the prospectuses clearly point out: (i) the parent might allow any or all of the real options to expire; (ii) there is no assurance of successful or timely product development; and (iii) patent protection and regulatory approval are unpredictable. If adverse selection were a problem, these shareholders would have little recourse but to accept the unfavorable outcomes that would make both their new venture shares and their warrants worthless. The risks of innovation are unsystematic and specific to each biotech firm. Since a SWORD is offered to the investment public, the new venture risk can be diversified across many investors, and these investors can diversify the SWORD risk across their portfolios. Because the R&D outlay is so large relative to the parent's size, the parent cannot diversify this risk internally across projects. A SWORD can be viewed as a mechanism for allocating the risk of innovation through the financial markets to investors who are willing to accept it. A SWORD does not necessarily indicate that adverse selection exists.

However, since the potential for adverse selection at the time of the offering has not been eliminated, a SWORD may not be as beneficial as the previous sections indicate. An examination of completed offerings can reveal the factors, both positive and negative, that are important in the market's valuation of a SWORD. The discussion in this section points to two different views regarding the stock market reaction to a SWORD announcement. If a SWORD is the result of an adverse selection phenomenon, then the parent's stock price should decline when a SWORD is announced. Such a negative response would suggest that either the parent's future growth prospects are weak or that the SWORD arrangement is nothing more than a financial gimmick. In addition, if less-than-favorable outcomes are forecast, the parent's stock price variability should increase as the parent's survival value becomes more questionable. On the other hand, if a SWORD embodies valuable R&D that is made available to the investment public through the unit offerings, then the parent's stock price should increase. Also, as the parent's survival value is enhanced, its variability should decline after the SWORD announcement. A favorable stock market reaction would be consistent with the contention underlying this paper: a SWORD can provide the financing that allows the parent to undertake innovation that might otherwise be foregone.

Unfortunately, only the total impact of a SWORD announcement can be measured; separating the effects pertaining to the R&D plans from those pertaining to the financing aspects is not possible. Nevertheless, the stock market response does reflect investors' overall assessment of a SWORD's worth. The evidence provided next will assist in gauging how investors have reacted to the first SWORD offerings.

V. Stock Market Reaction to the Units Offerings

The purpose of this section is to determine the stock market reaction to the SWORD arrangement. This is accomplished through a statistical analysis of both the daily returns around the SWORD announcement and the variability of these returns before and after the announcement. The null hypothesis maintained below is that a SWORD produces no change, either in daily returns or variability. Against this null, statistically significant findings provide a basis for selecting between the characterizations of a SWORD presented in the previous section. The methodology and results are described in order next, then a discussion of the results is presented.

A. Methodology

The daily stock return reaction is evaluated using standard event-study methodology.(11) The small number of SWORD offerings to date make this section a kind of case study. Daily stock prices for each parent have been collected for 130 days before the new venture was incorporated and 130 days after trading began. Closing prices are used for ALZA, which is traded on the ASE. Centocor, Immunex, and Genzyme are OTC stocks, and closing bid prices are used. The results do not change if closing ask prices or averages of bid and ask prices are used. Day -130 to day -11 before incorporation is the estimation period for the standard errors, and day -10 before incorporation to day +10 after trading began is the announcement period.

Abnormal returns are measured by market-adjusted returns, which are calculated as the difference between the day t return for the parent and the day t return on the S&P 500 stock index. A daily return is defined as the natural logarithm of the ratio of day t's price to day t - 1's price. Cumulative abnormal returns are computed as the sum of the daily abnormal returns between days t|Prime~ and t.

The small sample size precludes the use of each day's dispersion of abnormal returns, so each parent's history of one- and two-day abnormal returns is used instead. Employing the single-factor market model produces similar (not reported) abnormal returns but generates negative intercepts. For the incorporation and SEC registration events, day -1 is when the announcement is actually made, TABULAR DATA OMITTED and day 0 is when these announcements appear in the press. Since the market knows beforehand when trading will begin, the actual trading day is day 0 and the previous day is day -1 for this event.

F-ratios are used to test the equality of the variances before, during, and after the announcement of each SWORD. The F-ratio is calculated as a ratio of variances and is distributed with n - 1 degrees of freedom in both the numerator and denominator. The pre- and post-announcement periods each contain 120 observations, and the announcement period contains 73, 110, 86, and 79 observations for ALZA, Centocor, Immunex and Genzyme, respectively. The pre-announcement variance is always the denominator.

B. Results

Exhibit 4 shows that the market reaction to the SWORD announcements is generally positive around both the registration date (RD) and the trading date (TD). The two-day abnormal returns are between 3.85% and 5.84% for ALZA, Centocor, and Immunex; except for Immunex's negative, insignificant abnormal trading day return, all the RD and TD one-day abnormal returns are positive and significant for these three firms. For Genzyme, the RD results produce the largest (positive) one- and two-day abnormal returns in the exhibit. However, the Genzyme TD abnormal returns are the most negative in the exhibit; this mostly is recovered on days TD + 1 and TD + 2. TABULAR DATA OMITTED Overall, Exhibit 4 indicates that the market has responded positively to the RD and TD events, suggesting that the SWORD R&D is perceived to be valuable.

The reaction on the incorporation date (ID) is positive only for Immunex, but at this point, investors have little information about the new ventures and no assurance that the innovation will actually proceed. Exhibit 4 also shows that over the whole announcement period, ALZA (3.82%), Immunex (37.55%), and Genzyme (12.53%) outperform the market. Even though Centocor performs poorly relative to the market over the announcement period (-16.60%) as its stock price declines from $20.25 to $17.50 per share, its price does increase to $24.50 by the end of the post-announcement period.

Exhibit 5 displays the results of the variance ratio tests. ALZA and Genzyme experience a significant decrease (at the 0.01 level) in variance both around and after the SWORD announcement. Centocor also exhibits a similar decline in variability, but the decrease is significant only during the announcement period. The variance increases for Immunex, and significantly so in the post-announcement period. At best, these results moderately support the contention that these SWORDs enhance their parents' survival value.

C. Discussion

Exhibit 6 provides a closer look at the behavior of the cumulative abnormal returns over the announcement period. The generally positive responses for ID, RD, and TD are apparent for each firm. Immunex displays the most continuous upward trend, while ALZA and Centocor fall sharply after increasing early in the announcement period. Genzyme experiences each type of movement: its CAR is 12.53% by TD + 10 but is approximately twice that amount in the middle of the announcement period.

Combined with the significant post-announcement increase in variability (Exhibit 5), the behavior of Immunex's CARs suggests that it has increased in riskiness while increasing in value. Under adverse selection, an increase in riskiness would be associated with a decrease in value that would restore the risk-return tradeoff. However, in the options pricing literature, greater volatility produces greater option value. Perhaps the market is signalling that large potential NPVs exist and forecasting that Immunex might not be able to successfully execute the innovation. That is, perhaps the market at the time of the SWORD offering viewed Immunex in total as an option. Indeed, Immunex received FDA approval to market Leukine in early March 1991, well over two years after the units offering.

Panel A of Exhibit 7 shows that each firm's stock price has increased since the SWORD offerings, so why the CARs in Exhibit 6 vary so much for ALZA, Centocor, and Genzyme is not clear. No extraordinary events were found to have occurred during their announcement periods. Perhaps the CARs merely reflect the volatility of these stocks: the annualized post-announcement period standard deviations range between 36% and 48%. Moreover, non-SWORD events can significantly affect the valuation of these firms, as the following discussion indicates.

ALZA and Genzyme are similar in that each: (i) experiences a significant post-announcement decline in variance (Exhibit 5); (ii) generates approximately 60.0% of its revenue directly or indirectly (from royalties) from product sales (Panel B of Exhibit 7); and (iii) faces much less dilution from the warrants than Centocor or Immunex. Exhibit 7 also shows that ALZA is the oldest and largest firm of the four with a far greater number of products having received FDA approval (11) and under development (40) than the other firms. Genzyme's Ceradase received FDA approval (and a seven-year monopoly via orphan drug status) in April 1991, not long after its SWORD offering. Thus, the positive announcement effects and decreased variability for these two firms are buttressed by the value of assets in place (product sales) and technological leadership (products under development). The market's positive assessment of the SWORDs for ALZA and Genzyme appears to be fundamentally sound and not driven by concerns about adverse selection.

Centocor and Immunex are different from the other two firms. Even though Exhibit 7 indicates that Centocor derives about half of its revenue from product sales, has two products approved for sale in Europe,(12) and markets five diagnostic products, Centocor to date (1993) has no FDA-approved products. While Immunex received FDA approval for Leukine in 1991 and has ten products in development, its 1990 products sales are the smallest (equaling only 15.8% of total revenues) and its warrants-related dilution (approximately 30% in Panel D of Exhibit 1) is the largest of the four parents. The positive valuation effects in Exhibit 4 suggest that the SWORD R&D is expected to be valuable for these two firms. However, the post-announcement variability effects in Exhibit 5 imply that the market has concerns about the survival prospects for Centocor and Immunex that have not been resolved by the SWORDs.

In addition to the patent difficulties described in footnote 6, Centocor announced on April 15, 1992, that the FDA had determined that Centocor lacked data needed for the approval of Centoxin. Centocor's stock promptly fell $12.75 to close at Exhibit 7's $18.50. On April 2, 1992, Immunex announced that it would post a loss for the quarter just completed (analysts had expected a profit of $0.09 per share) and that it had filed suit against Hoechst AG for unfair marketing practices. Hoechst and Immunex had jointly developed Leukine but are marketing it separately.(13) Immunex's stock plunged $10.50 that same day. If calculated before these events, the annualized growth rates for Centocor and Immunex would be roughly double those reported in Panel A of Exhibit 7.

This discussion underscores the crucial nature of the unsystematic risks found in the biotech industry and highlights the extreme importance of property rights and regulatory approval. While the evidence reported in Exhibit 4 indicates that the SWORDs are expected to produce valuable R&D for the four parents examined here, a prudent suggestion is that each future SWORD should be judged on its own merit. In other words, the possibility that SWORD investors face potential adverse selection cannot be eliminated a priori. Yet, the sagas of Centocor and Immunex turn more on riskiness than asymmetric information. Indeed, proper recognition of biotech risk can enhance the benefits and usefulness of the SWORD arrangement.

VI. Concluding Comments

This paper examines a new form of project financing in the biotechnology industry that can encourage firms to engage in innovation that might not be undertaken using conventional forms of financing. A SWORD is like a contract research sponsor and a financial intermediary: it retains the biotech parent to perform product development and channels funds from investors to the parent to pay for development expenses. Consistent with the results of the Ernst and Young survey, the real option allows the biotech parent to develop a competitive advantage through research, key people, and proprietary products. In addition, the biotech parent can choose which products it will manufacture and market after the R&D has been completed. The purchase option on the new venture shares and the licensing agreement provide the biotech firm with control over the commercialization of all developed technologies and products. The combination of callable common stock and warrants on the parent eliminates the danger that the parent will expropriate the wealth of the new equity capital contributors. The market has reacted positively to the four SWORD offerings analyzed in this paper.

The focus here is on biotech firms, but the application of the SWORD is not limited to this industry. SWORD financing is likely to be successful in situations where: (i) product development is technical in nature; (ii) financing is difficult because of the large risk inherent in product development or because of firm size; and, especially, (iii) control over manufacturing and marketing rights is crucial to the viability of the firm.

I would like to thank Brent Vaughan for his assistance in the early stages of the analysis. The anonymous reviewers, Joseph Black, Edward Chow, Wayne Lee, David Leonard, and, especially, Anjan Thakor made many helpful comments on earlier versions of this paper. All errors remain my own.

1 Burrill |2~. This is the fourth annual survey conducted by Ernst and Young. Questionnaires were sent to 1,095 firms that were identified as U.S. biotechnology firms. The survey was conducted in raid-1989, and 480 firms completed the questionnaire, giving a response rate of 44%.

2 Burrill |2, p. 41~.

3 Since two separate, independent organizations exist, it is not precise to call the original biotech firm the parent. Because there are specific contractual relations that bind the new venture to the biotech firm, the biotech firm will be referred to as the parent organization.

4 Research limited partnerships have been used in the biotech industry as a way to attract institutional investors without necessarily relinquishing control over products. While providing many of the benefits of equity financing, research limited partnerships offer no simple way for investors to alienate their positions during the life of the partnership. The new ventures examined here use publicly traded common stock that is accessible to the entire market and provides greater liquidity because investors have an exit mechanism throughout the life of the activity. Strategic alliances are also popular in the biotech industry, and while the ultimate control is not relinquished, it is shared by the members of the alliance. A strategic alliance may offer synergies or other benefits that would apply to particular situations, but such considerations extend beyond the scope of this paper.

5 The major conclusion of Asquith and Mullins |1~, Masulis and Korwar |7~, Mikkelson and Partch |9~, and Smith |13~ is that when a firm issues equity, its stock price declines. These authors' findings generally support the adverse selection problem highlighted by Myers and Majluf |11~.

6 At the time of this writing, three other SWORDs have been completed and one is nearing the offering stage: (i) in November 1990, Elan Corporation offered Drug Research; (ii) in May 1991, Genetics Institute offered SciGenics; (iii) in late January 1992, Centocor offered Tocor II; and (iv) Genzyme currently is preparing Neozyme II for public offering. For various reasons, none are included in the analysis here. Elan is based in Ireland, so stock market data is not readily available for this firm. On March 5, 1991, Genetics Institute lost a major lawsuit to Amgen Corporation over the U.S. patent for erythropoietin, a red blood stimulator. The stock market swiftly reacted to this event: on March 6, Genetics Institute common fell from $62.00 to $40.25 per share. On April 22, its common fell sharply again from $36.50 to $29.00. In 1990, Xoma Corporation filed suit against Centocor for patent infringement in the U.S. regarding Centoxin, and on October 28, 1991, a jury verdict upheld Xoma's claim, leading to the dismissal of Centocor's counterclaim. On December 10, 1991, the court ordered Centocor to pay Xoma equitable compensation, and on January 10, 1992, Xoma requested 40% of Centocor's net sales of Centoxin. This matter has yet to be completely settled. Since it is very difficult to disentangle the effects of such major events from those of a SWORD, these two events have been eliminated from the sample. Neozyme II is not included since the offering has yet to be completed.

7 An interesting question is whether a competitor could purchase units and then exercise the warrants to establish a position within the parent. The answer is yes, but since each parent is publicly traded, this threat can never be eliminated. If this possibility were judged to be a problem in the units offering, the parent could consider carefully the proportion of shares outstanding represented by the exercise of the warrant.

8 In each future state s, the real option will be worth the maximum of 0 or (Vs - I), where Vs is the payoff in state s and I is the initial investment, so that Vg can be viewed as the expected value of the real option across all future states. Myers and Majluf |11~ note that incorporating the time value of money in their formulation does not change anything essential, and this condition is assumed to apply to Equations (1) through (3) here.

9 Since the last section explains why a SWORD will be financed by equity, the appropriate comparison is to equity-financed capital budgeting.

10 To avoid an immediate transfer of wealth from current to new venture shareholders, the warrant exercise price cannot be less than the stock market price at the time of the offering. Panel D of Exhibit 1 shows that the exercise prices range from 10% to 30% higher than the stock price. This means that new venture investors will have a stake in the future growth of the parent.

11 Vaughan |14~ examines the effect of a SWORD on the parent's cost of capital.

12 Product approval in Europe proceeds on a country-by-country basis. Centocor's products are marketed in the United Kingdom, Germany, France, the Netherlands, Denmark, and Luxembourg.

13 Immunex claims that Hoechst is flooding Immunex's customer base with free samples in an attempt to gain market share for its version of Leukine. Immunex also claims that it is owed royalties on Leukine by Hoechst.


1. P. Asquith and D.W. Mullins, "Equity Issues and Offering Dilution," Journal of Financial Economics (January/February 1986), pp. 61-89.

2. G.S. Burrill, Biotech 90: Into the Next Decade, New York, NY, Mary Ann Liebert, Inc., 1989.

3. S.H. Chan, J.D. Martin, and J.W. Kensinger, "Corporate Research and Development Expenditures and Share Value," Journal of Financial Economics (August/September 1990), pp. 255-277.

4. W.S. Krasker, "Stock Price Movements in Response to Stock Issues Under Asymmetric Information," Journal of Finance (March 1986), pp. 93-105.

5. H.E. Leland, and D. Pyle, "Informational Asymmetries, Financial Structure, and Financial Intermediation," Journal of Finance (May 1977), pp. 371-387.

6. J.C.T. Mao, "Project Financing: Funding the Future," Financial Executive (August 1982), pp. 23-28.

7. R.W. Masulis and A.N. Korwar, "Seasoned Equity Offerings," Journal of Financial Economics (January/February 1986), pp. 91-118.

8. J.J. McConnell and C.J. Muscarella, "Corporate Capital Expenditure Decisions and the Market Value of the Firm," Journal of Financial Economics (May/June 1985), pp. 399-422.

9. W.H. Mikkelson and M.M. Partch, "Valuation Effects of Security Offerings and the Issuance Process," Journal of Financial Economics (January/February 1986), pp. 31-60.

10. S.C. Myers, "Determinants of Corporate Borrowing," Journal of Financial Economics (March/April 1977), pp. 147-175.

11. S.C. Myers and N.S. Majluf, "Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have," Journal of Financial Economics (June 1984), pp. 187-221.

12. S. Shah and A.V. Thakor, "Optimal Capital Structure and Project Financing," Journal of Economic Theory (March 1987), pp. 209-243.

13. C.W. Smith, Jr., "Investment Banking and the Capital Acquisition Process," Journal of Financial Economics (January/February 1986), pp. 3-29.

14. B.A. Vaughan, "Units Offerings as an Off-Balance-Sheet Financing Mechanism for High-Risk R&D," Working Paper, Santa Clara University, June 1990.

Appendix A. The R&D Ventures

ALZA Corporation, Centocor, Inc., Immunex Corporation, and Genzyme Corporation are publicly traded firms with 1990 revenues of $109 million, $65 million, $35 million, and $55 million, respectively. ALZA develops advanced pharmaceutical products incorporating drugs in therapeutic systems that control the quantity, rate, and duration of drug release. ALZA has approximately 40 products in various stages of development, and client sales in 1987 are estimated to be $350 million worldwide. Centocor develops, manufactures, and markets innovative products for human health care utilizing monoclonal anti-body technology. It currently sells products useful in the detection or monitoring of ovarian, pancreatic, and breast cancer; it is developing and clinically evaluating products for diagnostic imaging procedures, principally focusing on cardiovascular disease; and it is conducting research on the use of monoclonal antibodies for the treatment of autoimmune diseases. Immunex has engaged in pioneering research in soluble cytokine receptors that show promise in fighting autoimmune and infectious diseases, and its Leukine received FDA approval in April 1991. Genzyme is developing products in four major business areas: (i) biotherapeutics, (ii) diagnostic products, (iii) pharmaceuticals and fine chemicals, and (iv) diagnostic services. Genzyme's key technologies are carbohydrate engineering, protein chemistry, enzymology, and molecular biology.

In October 1988, ALZA formed Bio-Electro Systems, Inc. (BES); in March 1989, Centocor formed Tocor, Inc.; in August 1989, Immunex formed Receptech Corporation; and in August 1990, Genzyme formed Neozyme, with each new firm financed by subsequent units offerings. BES was financed by current ALZA shareholders through a rights offering, while Tocor and Receptech were offered to the general investing public. In December 1988, 3.25 million units of BES were sold at $11 per unit for net proceeds of approximately $33.5 million. Dr. Alejandro Zaffaroni (founder of ALZA), Ciba-Geigy Corporation (sole holder of ALZA Class B common stock), and ALZA's ESOP subscribed to approximately 22% of the BES units. In July 1989, 2.875 million units of Tocor were sold at $12 per unit for net proceeds of approximately $31 million. In November 1989, the sale of 2.29 million units of Receptech generated net proceeds of $24.5 million. In October 1990, Genzyme offered 2.06 million units of Neozyme for over $47 million. Exhibit 1 in the body of the paper describes these offerings.

BES is to develop drug delivery systems incorporating ALZAMER bioerodible polymers, which are patented by ALZA, and ALZA's Electrotransport. Tocor is to develop products for treatment of rheumatoid arthritis, multiple sclerosis and systemic lupus erythematosus. Receptech is to accelerate the development of certain soluble cytokine receptor products for the treatment of autoimmune and inflammatory conditions. Neozyme is to develop products for treating thyroid cancer and cystic fibrosis as well as products relating to coronary artery disease, severe burns, and fetal cell separation.

Competition is the driving force behind the R&D efforts. ALZA notes that its current and future products face competition from more traditional forms of drug delivery systems and from competitors' advanced systems. Potential competitors are all other pharmaceutical firms, including current ALZA clients, and they possess greater financial resources, technical staffs, and manufacturing and marketing capabilities than ALZA. Centocor echoes these considerations, noting that the products to be developed will compete in a field of rapid technological progress, and cites well-known firms such as SmithKline Beckman Corporation, Pfizer Inc., Merck & Co., Inc., and Johnson and Johnson as potential competitors.

Michael E. Solt is an Associate Professor of Finance at the College of Business, San Jose State University, San Jose, California.
COPYRIGHT 1993 Financial Management Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Corporate Investments Special Issue; stock warrant off-balance-sheet research and development; includes appendix
Author:Solt, Michael E.
Publication:Financial Management
Date:Jun 22, 1993
Previous Article:Financing multiple investment projects.
Next Article:Recognizing retiree health benefits: the effect of SFAS 106.

Related Articles
Pumping up your research arm.
Final Jenkins Committee report debuts at symposium.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters