Hide Notes 

Slide 1: Dear Friend, The objective of the Monthly Read is to provide our members and friends a quick one-hour overview of the most interesting recent news items in private equity investing, technology and healthcare. All volumes of the Monthly Read are archived at our web site (http://www.wnyventure.com/). December 2003’s Monthly Read Trends in Legal Terms in Venture Financings In the San Francisco Bay Area (Fenwick & West) We analyzed the terms of venture financings for 85 technology companies headquartered in the San Francisco Bay Area that reported raising money in the third quarter of 2003. Patenting Bioinformatics Tools Patents may be the dominant form of IP protection for bioinformatics tools, but they aren’t always the best choice. Nano-transistor self-assembles using biology A functional electronic nano-device has been manufactured using biological self-assembly for the first time. What to Look Out For in a Term Sheet: An Attorney's View Angel Investing Survey 3rd Quarter 2003 See separate Adobe PDF file in email. HAPPY HOLIDAYS, The WNY Venture Association Team
Slide 2: Trends in Legal Terms in Venture Financings In the San Francisco Bay Area (Fenwick & West) by Barry Krame r and Michael Patrick (Fenwi ck & West), 11/24 /200 3 (Third Quarter 2003) Background We analyzed the terms of venture financings for 85 technology companies headquartered in the San Francisco Bay Area that reported raising money in the third quarter of 2003. Financing Round The financings broke down according to the following rounds: Series A B C D Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 20% 30% 19% 18% 20% 24% 29% 12% 15% 15% 28% 26% 13% 18% 13% 20% 30% 22% 15% 15% 19% 31% 24% 11% 11% 32% 27% 15% 15% 7% 36% 18% 29% 10% E and higher 13% Price Change The direction of price changes for companies receiving financing this quarter, compared to their previous round, were as follows: Price Change Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 Down Flat Up 53% 12% 35% 56% 4% 40% 73% 7% 20% 68% 6% 26% 67% 8% 25% 52% 10% 38% 57% 10% 33% The percentage of down rounds by series were as follows: Series B C D Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 46% 44% 53% 26% 59% 64% 56% 68% 50% 67% 33% 71% 77% 20% 60% 73% 25% 62% 84% 100% 78%
Slide 3: E and higher 82% 93% 87% 83% 90% 82% 75% The direction of price changes in Q3 was generally similar to Q2, with downrounds continuing to outpace up-rounds, but by a less substantial margin than previous quarters. Liquidation Preference Senior liquidation preferences were used in the following percentages of financings: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 57% 55% 68% 64% 60% 56% 62% The percentage of senior liquidation preference by series was as follows: Series B C D Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 50% 56% 60% 43% 59% 55% 64% 65% 73% 73% 60% 51% 67% 78% 66% 33% 55% 83% 70% 38% 55% 69% 91% 44% 62% 69% 100% E and higher 73% Multiple Liquidation Preferences The percentage of senior liquidation preferences that were multiple preferences were as follows: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 21% 44% 40% 37% 32% 41% 58% Of the senior liquidation preferences, the ranges of the multiples broke down as follows: Range of multiples Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 >1x- 2x >2x - 3x > 3x 88% 12% 0% 78% 11% 11% 74% 16% 10% 79% 14% 7% 80% 13% 7% 87% 13% 0% 66% 27% 7% The reduced use of multiple liquidation preferences is notable. Participation in Liquidation The percentages of financings that provided for participation were as follows: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02
Slide 4: 68% 81% 77% 74% 73% 67% 56% Of the financings that had participation, the percentages that were not capped were as follows: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 66% 59% 63% 50% 51% 56% 32% Cumulative Dividends Cumulative dividends were provided for in the following percentages of financings: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 12% 4% 8% 9% 13% (There are no figures for Q2 or Q1 of 2002 as we initiated our analysis of these numbers beginning in Q3 of 2002.) Antidilution Provisions The uses of antidilution provisions in the financings were as follows: Type of Provision Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 Ratchet None 8% 5% 10% 88% 2% 16% 83% 1% 18% 77% 5% 15% 80% 5% 20% 78% 2% 29% 69% 2% Weighted Average 87% The trend toward reduced use of ratchet antidilution protection continued in Q3. Pay-to-Play Provisions The use of pay-to-play provisions in the financings was as follows: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 12% 20% 14% 25% 18% 18% 20% The pay-to-play provisions provided for conversion of non-participating investors' preferred stock into common stock or shadow preferred stock, in the percentages set forth below: Common Stock Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 80% 84% 50% 75% 71% 50% 56%
Slide 5: Shadow preferred sto ck Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 20% 16% 50% 25% 29% 33% 22% Redemption The percentages of financings providing for mandatory redemption or redemption at the option of the venture capitalist were as follows: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 29% 40% 32% 33% 37% 44% 36% Corporate Reorganizations The percentages of post-Series A financings involving a corporate reorganization were as follows: Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 21% 19% 19% 21% 19% 11% 20% Percentages of these corporate reorganizations involving a: Reverse split of outstanding stock Q3 '03 Q2 '03 Q1 '03 Q4 '02 Q3 '02 Q2 '02 Q1 '02 79% 87% 50% 55% 67% 70% 100% Conversion of senior securi tie s into more junior securities. Q3 '03 Q2 '03 52% 53% Q1 '03 53% Q4 '02 59% Q3 '02 47% Q2 '02 30% Q1 '02 67% Conclusion The overall terms of venture financings in 3Q03 improved mildly. The direction of price changes was similar to 2Q03, but the use of some of the tougher terms such as multiple liquidation preference, ratchet anti-dilution and pay-to-play decreased. Similar to this trend, we note that various industry publications have reported venture investment in 3Q03 to be generally flat with 2Q03.
Slide 6: Lastly, we note that Nasdaq increased approximately 9% in 3Q03, and that both the number of companies closing IPOs, and the number of companies in registration for IPOs, increased in 3Q03. To the extent these trends continue, we would expect venture terms to continue to improve.
Slide 7: Patenting Bioinformatics Tools Patents may be the dominant form of IP protection for bioinformatics tools, but they aren’t always the best choice. Bioinformatics uses computers to identify and find meaning in the patterns of biological data expression patterns, which may provide the answer to how cells respond to disease. Thus, the cornerstone of many bioinformatics companies is often a relational database connecting expression data or molecular structure data to the biological context in which the data arise. Additionally, many companies use proprietary algorithms to analyze the data: for example, to search for biomarkers, identify molecular pathways, or simulate molecular processes. Commercialization of bioinformatics toolsfor example, databases, algorithms, software, and hardwareinvolves a number of competing interests. Such tools may be important for facilitating new discoveries; yet new, proprietary algorithms or databases could take advantage of existing algorithms, databases, or methods. In any case, the institutions and companies that create bioinformatics tools are expending considerable resources to provide ways to decrease the time and money required to develop valuable drugs. Incentives for these efforts are currently provided by the overlapping intellectual property protection created from patent, copyright, and trade-secret law. Since patents are generally regarded as the dominant form of intellectual property protection, this article will focus on seeking patent protection for bioinformatics tools. What’s Patentable in Bioinformatics In 1990, the U.S. Supreme Court in Diamond v. Chakrabarty announced that “[a]nything under the sun that is made by man” could be the subject of a patent. However, judicial interpretation of the patent laws has foreclosed patent protection for abstract ideas, scientific laws, naturally occurring phenomena, products of nature, mental steps, and printed matter. For example, a claim to an algorithm underlying a computer program would be rejected if it were considered nothing more than an expression of ideas. Still, patent protection for an algorithm or a data structure may be obtained if the invention described by the claim can be pigeonholed into one of four categories: • • • • A process A machine An article of manufacture A composition of matter The courts have required that computer-related inventions additionally must provide “tangible results” or have a “practical application.” Initially, this was interpreted as requiring a physical output or transformation as an end result. The use of an algorithm merely to manipulate data was not sufficient, nor could the data themselves be patentable. The practice of the U.S. Patent and Trademark Office (PTO) has been to interpret recent cases such as State Street Bank and Trust v. Signature Financial Group and ATT v. Excel Communications to mean that for claims to software to be patentable, the data residing in the software must “interact” with a computer-readable mediumthat is, they must be able to direct a computer to accomplish a particular result. Despite the limits imposed by statute and policy, patents covering various aspects of bioinformatics provide important tools for commercializing this technology. For example, methods of using a bioinformatics database or algorithms implementing a search for biomarkers can be claimed, and in many ways these claims resemble those found in business-method patents, with little or no reference to software or hardware elements. Details, Details Just as with any other type of technology, claims in a bioinformatics patent must be novel and unobvious, as determined by a patent examiner. An examination of novelty and non-obviousness also may turn on whether the novelty of the invention resides in nonfunctional, descriptive data (e.g., such as the data in the database) or in functional data (e.g., a program for manipulating the data). PTO guidelines instruct patent examiners that the novelty of an invention must reside in functional data for a claim to be patentable. Under this standard, a novel compilation of data claimed in combination with a known means for searching the data may nonetheless be considered obvious and not patentable. Other issues created by the nature of bioinformatics must be dealt with during the examination process. As with software patents in general, there is concern in the PTO that critical publications may not be readily identifiable
Slide 8: and that examiners may not make appropriate comparisons between information already known in the industry (sometimes called “the prior art”) and a claimed invention. This is further complicated by the fact that the examiners looking at bioinformatics applications typically are not skilled in both biotechnology and information technology. In an effort to address this, the PTO has strongly recommended that applicants provide more detailed statements concerning the relevance of cited art. The PTO also encourages interviews, so that applicants can “explain” the invention to examiners and help focus the examination process. IP Protection vs. Business Goals While obtaining a patent has an obvious upside, it has a downside, too (see “Patent Pros and Cons”). These disadvantages may encourage companies to seek other forms of intellectual property protection for bioinformatics tools, such as copyright or trade-secret protection. But a bioinformatics company should also consider IP protection alternatives in case it needs to change its business model in the future. On one hand, a bioinformatics company shares with any biotechnology company the ultimate goal of developing and commercializing drugs based on cellular products. On the other hand, output for a bioinformatics company is, by definition, information. Creating a business plan around such output poses numerous challenges. For instance, the company may choose to focus on one or more of several core competencies: • • • Service providerusing a bioinformatics database to analyze the data of others Content providerproviding access to a database and permitting third parties to analyze their own data Software providermarketing a data management platform so that others can develop their own databases Traditional product companyusing its database to develop drugs in-house The company also must choose whether or not to disclose the information contained in the bioinformatics tool. These decisions will influence the types of intellectual property a bioinformatics company creates and seeks to protect, and the type of intellectual property protection it should choose. Organizations looking to commercialize bioinformatics algorithms and other tools should recognize the unique requirements of their business model, and work to match the best intellectual property regime to that model (see “Types of Intellectual Property Protection for Bioinformatics Tools”), while always being prepared to re-evaluate the protection as business conditions change. Since the law in these areas may be a moving target, too, it’s important to explore all avenues for intellectual property protection. Dianne Rees, Ph.D., is an associate in the Intellectual Property Practice Group of Edwards & Angell LLP, in Boston, Mass. She may be reached at drees@edwardsangell.com. John Ottaviani is a partner in the Intellectual Property Practice Group of Edwards & Angell LLP, in Providence, RI. He may be reached at .
Slide 9: Nano-tran sisto r self- assemble s using biology A functional electronic nano-device has been manufactured using biological selfassembly for the first time. Israeli scientists harnessed the construction capabilities of DNA and the electronic properties of carbon nanotubes to create the self-assembling nano-transistor. The work has been greeted as "outstanding" and "spectacular" by nanotechnology experts. The push to shrink electronic circuits to ever smaller dimensions is relentless. Carbon nanotubes, which have remarkable electronic properties and only about one nanometre in diameter, have been touted as a highly promising material to help drive miniaturisation. But manufacturing nano-scale transistors has proved both timeconsuming and labour-intensive. The team, at the Technion-Israel Institute of Technology, overcame these problems with a two step process. First they used proteins to allow carbon nanotubes to bind to specific sites on strands of DNA. They then turned the remainder of the DNA molecule into a conducting wire. Proo f of principle "DNA is very good at building things in molecular biology, but unfortunately, it does not conduct electricity. We had to get a metal conductor on the DNA," explains physicist Erez Braun, who led the research. "This is spectacular work," says Cees Dekker, a nanoscience expert at Delft University in the Netherlands. "It demonstrates that it's possible to use biology to build an inorganic device that works."
Slide 10: "But while it is a first step towards molecular computing based on this type of DNA configuration, we are still many years way from large scale self-assembly electronic devices, such as computers," Dekker cautions. Bac teria l protein Braun's team began their manufacturing process by coating a central part of a long DNA molecule with proteins from an E. coli bacterium. Next, graphite nanotubes coated with antibodies were added, which bound onto the protein. After this, a solution of silver ions was added. The ions chemically attach to the phosphate backbone of the DNA, but only where no protein has attached. Aldehyde then reduces the ions to silver metal, forming the foundation of a conducting wire. To complete the device, gold was added. This nucleates on the silver and creates a fully conducting wire. The end result is a carbon nanotube device connected a both ends by a gold and silver wire. The device operates as a transistor when a voltage applied across the substrate is varied. This causes the nanotubes to either bridge the gap between the wires completing the circuit - or not. Out of 45 nanoscale devices created in three batches, almost a third emerged as self-assembled transistors. They work at room temperature and the only restriction for future devices is that the components must be compatible with the biological reactions and the metal-plating process. The team have already connected two of the devices together, using the biological technique. "The same process could allow us to create elaborate selfassembling DNA sculptures and circuitry," says Braun. Journal reference: Science (vol 302, p 1380)
Slide 11: What to Look Out For in a Term Sheet: An Attorney's View by Alex Wilmerding, Aspatore Books Staff, 10/14/2003 VC Experts has teamed up with Aspatore Books to reprint some of the more salient sections of Alex Wilmerding's Deal Terms, The Finer Points of Venture Capital Deal Structures, Valuations, Term Sheets, Stock Options and Getting Deals Done. This excerpt features an interview with a leading legal expert in the private equity arena, James M. Crane, who has served as counsel at the Boston law firm of Testa, Hurwitz & Thibeault, LLP, a seasoned lawyer who tells us what to look out for in a term sheet. Jim's perspective is representative of the candid, forthright demeanor that entrepreneurs and venture professionals look for in counsel. Alex Wilmerding: What exactly would you consider typical when looking for common ground in the form of a term sheet? Is there a neutral form of term sheet we can use as a starting or reference point for our discussion? Jim Crane: While it is certainly relevant to include a balanced form of term sheet as a reference point, and I understand you have included one in the appendices of this book, it is frankly unlikely that any two lawyers would ever agree on the neutral form of any term. Each term sheet is affected by the subjective dealings of two parties; you can always come up with an argument for its favoring one party or another. AW: Are the terms that are generally incorporated into a Series A Preferred by VCs created specifically for the current financing, or do they reflect some standard a lawyer or venture firm prefers? JC: When you're talking about an early-stage preferred financing, a series A or perhaps a series B that occurs a year or two after the series A, very often the things that are coincidentally found in the term sheet and sometimes end up in the deal documents are terms that VCs have used in other recent financings. AW: It may be most interesting for entrepreneurs to consider how lawyers look at a term sheet when representing a VC's interests. As companies progress through successive rounds of financing, it is standard that new VCs will bring new capital to a company. In these instances, it is important to consider how VCs interpret the terms presented to a company. Considering how VCs look at term sheets is important if entrepreneurs are looking for a perspective on the priorities VCs have when they draft a term sheet to be presented to a company. Let's consider the core terms to look out for in a term sheet. Which core terms do you first examine and consider when examining a term sheet that has been presented to you for review? JC: The things that are important to me may be different if I'm representing a VC. A VC may draft a term sheet and have me review the draft before it is submitted to a company. A VC who is already invested in a company may pass me a term sheet that has been presented to the company by a prospective new investor.
Slide 12: If I am representing a VC who has a working first draft of a term sheet, the first thing I do is to take a quick look at the valuation of the company. The valuation section gives you a framework from which to work. The first questions I ask when examining valuation are the amount of capital going into the company and the price per share. I also look closely at the key terms of the preferred stock. They may be a little different or have different protections, depending on whether it's a first round of preferred stock, a Series A round, or a later round, a Series E round, for example. If you're looking at a Series E round, very often just by way of precedent, the other rounds and their attendant leverage predetermine much of the nature of the framework and context for the terms of a Series E round. The extent to which the new money will have leverage over the old money will be spelled out here. A Series A, by being the first preferred to be issued, has the opportunity to define terms for the preferred. A Series E sometimes will need to incorporate provisions from prior classes of preferred that you have to live with. In these instances, you have to pick and choose your battles for things you want changed in the structure coming in. In this respect, I always glance at the dividend provision, as well. I don't think whether there will be cumulative dividends is something a lawyer will dwell on, but the dividend provision is important to pay attention to, particularly when considering the implications for the preferred and the implications for a company with multiple classes of preferred, should they also hold dividend provisions. Next, I look at vesting. What you'll usually want to do if you're the VC is to put right up front what you want the vesting of the founders' shares in an early-stage financing and the options pool to be. You don't want to create a situation where your founders can sneak out and leave the company high and dry when they are more often than not - and certainly in the early stages of the company's life span - vital to a successful venture. Beyond the economic parameters that some of the other terms mentioned will dictate, the restrictions on the founders and key employees will be critical to an early-stage venture's success. After a company has been around for a while, shares will have vested; and if the company is starting to make it, then maybe the importance of keeping any one individual starts to decline. Vesting schedules are probably the biggest thing to look at when I am looking at an early-stage term sheet. These schedules could be more important to consider than the terms for the preferred; if there are few restrictions in place on the founders' and key employees' stock, there is the potential need to issue incremental stock options. While not considered part of the options pool, investors may also ask founders in an early-stage round to have their shares subject to vesting. This is sometimes generically referred to as "reverse vesting" - agreeing to have existing common stock, typically founders' shares, subject to vesting - to give founders incentive to perform. People don't often consider the importance of vesting schedules because people are trained to start digging into the terms of the preferred. And usually those provisions on the restrictions on the founder are stuck toward the back of the term sheet. But the implications for a company's future are significant. AW: Options, and particularly the subject of vesting, are of great interest to entrepreneurs. Clearly, the terms defining vesting and valuation are, to some extent,
Slide 13: interrelated. Is this a good example of the inherent flaw in dwelling too much on any one single term without considering implications of others? JC: It is a good example of how many terms are intricately linked and cannot be considered entirely independently. The value you put on a company, assuming you'll add key people in the future, will be affected if additional shares need to be added to an option pool after a closing to provide appropriate incentive to management. I don't think this is necessarily obvious in the math of the valuation of the company unless, when you do your initial analysis, you consider vesting schedules for existing staff and whether the number of options included in the proposed capital structure at the time of closing anticipates options needs for the foreseeable future. For a relatively early-stage company, one that has been in existence for one or two years, language describing vesting schedules and treatment of options vesting should give a pretty good picture of the profile of a company's options pool. Vesting and the subject of options are sensitive for founders because many have not been involved in this process before. With good reason, founders and key employees need to be prepared to accept the shock that their shares will be restricted right off the bat. Usually key employees and founders form the company. They set it up; they issue themselves common stock; and they think they own that stock. In a sense they do, but if they want to be financed, they'll have to put the restrictions on their ownership positions that the VCs are looking for, and that's often an emotional battle. I've had clients who were virtually throwing up at the thought that they had to turn over their shares and earn them back again. It is hard to educate highly educated people, but it is understandably new territory for them. Entrepreneurs can have trouble getting around it. The company's lawyers, if they have not done so already, have to get their clients to realize this is how venture financings work. AW: Assuming entrepreneurs and companies are well advised, are there ways to anticipate some of this heartburn? What then are the preemptive strikes you would advise a company to make to mitigate terms that may appear on a term sheet? JC: Accepting that a stock restriction agreement will be part of what will be expected by VCs is the first step. Very often what I do if I am representing a company in an earlystage investment is to try to preempt the VCs on this issue and set up the stock restriction agreement for the company and their founders before opening the negotiations with the VC. This way a company is prepared to say it already has an agreement in place, and it makes the issue a little tougher because the VCs then have to impose harsher vesting terms than the ones that already exist in the stock restriction agreement. If no agreement exists, it is easy for a venture firm to ask for certain thresholds straight off the bat. Many times, the VCs don't want to make enemies with the founders because doing so is in nobody's best interest, so changing an existing agreement may be something less likely to be initiated, once established. You can't really preempt VCs on any of the charter provisions and preferred stock terms. Usually a new company will have a very plain vanilla common stock charter. The VCs typically will come in and know what they're looking for, and you may have to redo your whole charter if they want to make changes. Probably the only meaningful battle you can preempt the VCs on is the type of vesting. You have to take an educated guess based on where the market is in any particular month. A common VC approach is to have
Slide 14: founders' shares put on a four-year vesting term. Entrepreneurs may try to preempt the vesting issue for their existing shares by adopting an agreement to have one-third vested immediately and the rest on a three-year term. Often the VCs will come in, and, depending on the relationships and how much they want to offend the parties, they'll say they'll let that stand. This is a good way for VCs to express their acknowledgement of the founders' value before the game even starts. AW: Let's move on to liquidation preference, which is often a sticking point and an important term on which to focus. What do you look for when examining a liquidation preference clause? JC: Liquidation preference is really what the deal is all about. The liquidation preference section defines how much each class of preferred stock will receive in preference to other stockholders if a company is sold or generates liquidity as a result of a transaction before any of the other stockholders receive proceeds. The definition of a liquidation event is therefore very important; it will define when liquidation preference terms take effect and how individual classes of preferred shareholders will be treated as compared to common. Liquidation will be defined as an actual dissolution of the company, and the definition could also include a merger, an acquisition, a sale of the company, or a change of control of anywhere from 50 percent to 80 percent to 100 percent of the voting power of the company - often referred to as a deemed liquidation. In the event of what is defined as a deemed liquidation of the company, liquidation preference terms would take effect. AW: What other term sheet text may reveal preferential treatment to the preferred? JC: If protective provisions are incorporated into a term sheet, the preferred are looking for a form of blocking power to block management from making certain decisions without their approval. Protective provisions typically require that a certain percentage of the preferred be required to vote to approve certain actions the company chooses to take. Without the vote and approval of a certain percentage of preferred stock, the company is blocked from taking certain actions. For example, protective provisions can stipulate limits on the amount of debt a company can take on. AW: How might protective provisions restrict a company from increasing its option pool or pursuing a subsequent financing without prior approval? JC: Protective provisions could restrict certain actions, including amending the charter of the company, pursuing a merger or acquisition, and increasing the size of the equity pool, for example. On the subject of options, protective provisions that require approval of some majority of the preferred to make any amendment to the charter of the company in effect limit a company's ability to increase its options pool beyond the authorized shares stipulated in the charter. The charter will detail the total authorized capital stock of the company, which has a limiting effect on the number of shares in the company's option pool. If the company wanted to increase the option pool and protective provisions required that approval from some majority of the preferred be secured to do so, the company would in effect be restricted from increasing the options pool unless some majority of the preferred were to approve an amendment to the charter. Assuming a company has an options pool of sufficient size to serve its needs, this power will give the preferred a degree of leverage, should a board and management look to increase the size of the options pool at
Slide 15: some future point. If, subsequent to a financing, additional options were required to create an additional incentive for additional staff, a further increase to the size of the options pool would be dilutive to the preferred. The preferred may exercise their blocking power to, in effect, require that the existing option pool be reapportioned, for example. Such a provision does not have to be an obstacle to future expansion to an option pool; companies that meet and exceed the business plans they commit to and that require an expanded option pool to accommodate growth will most probably receive the support of their investors if such an initiative is in line with investors' expectations. Protective provisions may also forbid the issuance of additional stock and hence a subsequent equity financing without approval of some majority of the preferred. This is an additional source of leverage for the preferred and presumably a deterrent to management who might otherwise assume that serial rounds of equity financing could be a panacea to their capital needs. AW: How creative can investors and lawyers get with protective provisions? JC: You can incorporate just about anything your heart desires. There are certain terms you almost always see, such as in the case of a merger or sale of the company. Similarly, provisions preventing changes in the nature of the company's business are not uncommon; an investor does not want to be buying stock in a company that's doing X, and the next thing they know the company's doing Y instead. AW: How are protective provisions of some early form of preferred - a series A, for example - incorporated into future rounds of preferred? JC: One thing that's important for a late-stage company financing is for the interests and therefore the protective provisions associated with early rounds of preferred to be balanced with the interests of later-stage investors. A company does not want investors from early rounds who may not have the capital to participate in later rounds of financing to hold the power to prevent the company from pursuing a good deal later on in the life of the company. It would be very unlikely that a Series B would agree to let an existing Series A vote independently on this issue unless, of course, the investors in the Series A and the Series B were the same. Instead, often a later series of preferred will require the protective provisions of earlier preferred be revised so that all classes of preferred vote collectively. AW: We've talked about valuation, vesting, liquidation preference, and protective provisions. What's next on your list? JC: One of the most important things is the impact of dilution protection provisions. That's always a key thing. In the term sheet stage, negotiating the terms for dilution protection provisions is not always easy. Often investors won't actually spell out the dilution protection provision formula in the term sheet; they'll wait to do so in the final definitive closing documents. They just say a weighted-average formula. AW: But there are several types of weighted-average formulas. JC: That can be the tricky part. It probably would be a better practice to actually put the formula into the term sheet, or at least clearly define it. It's not as much of an issue if antidilution is a full ratchet because that's pretty cut and dry. If it's a later-stage company, maybe it's not so much of an issue because they've had a weighted-average formula in
Slide 16: their charter, and the VCs have already seen it, and they'll just go with it. In an earlystage company, there might be a little bit of an argument about which shares will be included in the anti-dilution formula when you actually get to the drafting stage and have to put the language in. At that stage, there may be disagreement over what shares are included in the calculation. Another thing is that you'd be amazed at how often someone can later go back to the formula, or the language if there is no formula, and it doesn't read right. It's very tricky to use the English language to write out a mathematical formula. The language is awkward and weird, and it gets screwed up more often than you think it should. This is the nuts and bolts of the economics of the deal, and I think it's critical that any time you're going through anything involving numbers in a term sheet, you actually do the math and make sure the numbers work. I never read an anti-dilution provision and say, "Oh, that looks right. Those are the words I'm familiar with." Even if I am 99.999 percent sure that's the way I have seen them every other time, I still take out my calculator and a pen and paper. I take some numbers from the term sheet somewhere else and assume a scenario and plug in numbers to see how it comes out. Sometimes a very subtle mistake in how the words are transcribed from whatever previous deal you used will screw up a formula. AW: Let's consider the IPO process and how it is reflected in term sheets. Could you comment on registration rights and how they are treated? JC: Registration rights are almost more of a leverage tool that plays later on when a company goes public. This subject does not necessarily get a lot of attention from entrepreneurs when they are focused on getting enough capital in the door to get to cash flow break-even. Registration rights are one area of language in a term sheet that levels out the most in a down market - IPOs are more distant in everyone's minds. The things to look for there are the number and types of registrations and the time frames involved - how quickly a registration can be forced and when the registration rights expire. Very often registration rights agreements will have some built-in expiration, usually based on a period of time or when the shares become otherwise salable, for instance under SEC Rule 144. The expenses of registration are another thing to focus on. Typically the company bears the cost of registration, but if someone requests a registration and withdraws the request, it is not uncommon for the company to retain the right to pass the expense on to the investor who requested the registration. This is one important way for a company to put in place disciplines that encourage investors to make sure their requests for registration are well thought out. On the other hand, if you're a VC, you might agree to this approach but try to ensure that expenses will be covered by the company if the registration is withdrawn for reasons that were beyond the VC's knowledge or ability to foresee. A war or bombings that rattle the financial market are extreme examples. Investors will also want to include piggyback registration rights; they will want the right to participate not only in the first but also in subsequent public offerings. Investors will want to detail what happens if there's a cutback in how much stock can be sold on the public markets; they'll want to detail how each type of stock will be treated. Usually there will be a hierarchy: If there's a cutback, this one goes first, and we'll take it from there. I've seen some cutback provisions that get quite intricate about who gets to register their
Slide 17: shares in what order. But often that level of detail is worked out in the deal documents and not in the term sheet. "With appropriate cutback provisions" is about as specific as the term sheet language will get. One thing that's very important with respect to registration rights is that you don't want to have a lot of registration rights agreements out there. It's important that everyone is all tied up in one agreement. It's easier and less complicated because you don't want to have to go back through the old terms. If there are varying terms you want, at least try to have them all in one document, so you have only one place to look to find out where everyone stands. You want to have everyone who has registration rights at all to be party to the one agreement. Whether you're representing the company or the investor, I can't imagine a scenario where it would be in your interest to allow multiple registration rights agreements out there. The registration rights contained in the agreement are rarely used, but it's an extremely powerful tool that has very powerful consequences for everyone registering shares for public sale. You don't want those rights to just be floating out there - you want to know where they are, and you want everyone to be wrapped up in the same agreement. If you don't do that, or at least vigorously fight for it, then I don't think you're doing a service to anyone. Usually there's no objection at all - I've never had anyone say no to it. AW: Give some examples of how term sheet style and approach can be unique. JC: The terms commonly used by one investor for one type of deal can begin to creep into term sheets for companies facing very different situations for which such terms may not be appropriate. Take for example an investor who's been working in an environment where they go into struggling companies because they can pick leverage situations or turn companies around long enough to put some money in and then get out. It is not uncommon to see the terms used by these individuals popping up in term sheets in earlystage ventures where such terms may not be the common practice. Often when companies are struggling, the term sheets and deal documents might include milestones that spell out consequences for the company's failure to meet those milestones. Very often these consequences might include existing stockholders losing control of the board and investors taking control of the company if the company is not performing. In a different investing context, such as an early-stage financing, if you have a VC who's used to getting those terms, it's harder than you'd think to get them to drop them. The most dangerous thing about term sheets is for an investor to get into the habit of just grabbing the last one he had from a recent deal and revising it. Whenever I'm asked, I'll provide an original term sheet that covers all the bases. Term sheets are restrictive in nature because you're stuck with the terms spelled out on the page. As the corporate attorney sometimes you'll be given a draft of the term sheet you'll recognize from having worked on it in a previous deal. Just because you did it last time doesn't mean you want to do it this time. I try to start with a blank term sheet, not one I've used before. There's nothing wrong, however, with using a form as a tool, as long as you use it in the right way. I have a formal term sheet I like to look at, but it doesn't have everything that could possibly be incorporated into a term sheet. This one has suggestions for alternatives, bracketed items, and notes about things to consider. It includes alternative clauses. It does
Slide 18: not have everything. When you keep building off an older version of a term sheet, at some future point, you'll miss the chance to take another direction because the forms you always use say X, and you've forgotten the alternatives because they are not in front of you or you haven't had them in your deals lately. You've forgotten about the possibility of doing Y instead. AW: Is there a strategy you recommend people follow when negotiating a term sheet? JC: It depends on the facts for the particular deal and your sense of what's important to the company and to the VCs. As for the pricing and that kind of thing, that's better left out of the hands of the lawyers. Very often that's been settled before the lawyers even get the call from the client to look at the term sheet. I don't know that I have any advice on strategy that would apply to every situation. I think in early-stage investing, the key is the founders and their level of sophistication. If they're not sophisticated, investors may end up having problems getting past the hurdles I mentioned before, like the vesting on the founder ownership, which is hugely important. I don't know if there's a good way to soften them up before an investor starts hitting them with the other points. It really just depends on the people involved and the facts of the situation. AW: Have you seen really gross variances or differences in the evolution of the term sheet over the years? JC: Over the past five years, the staples of term sheets have remained relatively the same, and yet we have seen a big swing from company-favorable to investor-favorable financing climates, influencing which party the language favors. Consider protective provisions, for example. In a company-favorable investment climate, it is almost unheard of for a company to have a charter with a ratchet provision or very onerous provisions. In a more investor-favorable climate, the deals happen at a slower pace, and there's more time to negotiate in favorable terms. When deals are happening so fast that it may seem there are other VCs around the corner, in some cases, tripping over themselves to give a company money, investors are unlikely to incorporate extensive protective provisions. In both types of investment climates, the issues are the same. The biggest shift is in the negotiating leverage. In a very company-favorable climate, there are a lot of vanilla terms, and deals get papered very quickly. As the deals slow down, the terms get negotiated harder, and companies find themselves with more terms they have to live with. AW: Do larger law firms have certain views about how much work they'll do for earlyor expansion-stage companies? JC: Large firms will be selective; they often will not get involved in every angel round. That's not to say they won't do them if there's a relationship there or something they're interested in. Usually the angel rounds happen at very small amounts, and the terms aren't so complex; it's a friendlier negotiation, and everyone is just trying to get the company into the game. The company is not quite ready to venture into the real arena yet, so they don't need a large venture law firm to work on the documents. There are individuals or small firms that will do it, and there are groups of angels who support them. Either way, you want to engage a qualified professional, with experience, who won't create problems in documentation that will cost more money later to fix. Angel investors get in and pay attention to protecting their ability to stay in the game and going forward a little bit;
Slide 19: they're trying to make sure that down the road they're not just lumped in with common. Such terms are not necessarily complex enough to justify use of a large firm. AW: How would you recommend an entrepreneur go about selecting a lawyer or a particular firm to work with? JC: In the very early stages of a company, you at least want someone who's done this before who can hold your hand. The point is that there's almost nothing in the term sheet that isn't worth looking at. Depending on a company's bargaining position, it's important for a company to get good legal advice because you never know to what extent great thought has been put into the term sheet that may be presented to you, or if it's a remake of a previous form - some terms might not be entirely germane to the company in negotiation.