Secrets of Sand Hill Road

by Scott Kupor

  Now, in fairness, there are times later in a company’s financing life where a new investor may insist on having at least some of the corporate actions carved out of the capital “P” Preferred voting and decided by a separate series vote. And sometimes this may be appropriate to consider, but best not to set that precedent early in the life of the company.

  Where might this come up? When a later-stage investor decides to invest in a startup, she may be investing a significant amount of money, but still end up holding a relatively small ownership position in the company. This is because naturally the valuation for the company should have increased over its various rounds of funding, meaning that a dollar invested at today’s valuation buys an investor much less than did a dollar invested many years ago at the Series A round.

  As a result, our new investor will have a lot of money at stake in the investment, but likely will be a significant minority from a governance perspective relative to the other VCs who have been investors for a longer period of time and thus at much lower entry valuations. If all the protective provisions, therefore, rely on a majority vote of all the preferred voting together as a single class, the new investor may always find herself powerless to affect the vote; the earlier investors probably control more than 50 percent of the preferred and thus can carry a majority vote on their own.

  And their economic incentives could be very different than those of the new investor, given that the amount of capital they have invested is very different and their economic ownership could be very different. For example, there might be acquisition scenarios that produce a great return to our early VCs, but that only provide a return of invested capital to our new investor. It’s these types of situations that the new investor will seek to avoid.

  One way of doing that is to ask for a separate class vote for the new investor. That is, the new investor’s approval is separately required to approve any corporate actions. Obviously, this can be problematic for exactly the scenario I outlined above—the new investor’s economic interests may diverge from others, and you are giving her disproportionate voting control relative to her economic interests. An intermediate way of addressing this concern is to not provide a separate series vote for all corporate actions, but rather to enumerate a specific list of things that the new investor is most concerned about for which she can have a separate class vote.

  Another way to deal with this is to increase the voting threshold for required corporate actions to something greater than a simple majority. What the actual number should be will depend on the specifics of the situation, but the new investor might ask for some number that requires a greater diversity of existing investors to approve the transaction, with the hope that this lessens the likelihood of her always being outvoted by the very early VCs who might control a majority of the preferred shares.

  Now that we know who votes, the next question is, what do they get to vote on? You’ll see in the term sheet (this page) that there is a long list of things that the protective provisions give the Preferred shareholder a right to vote on.

  We won’t cover all the items, but let’s talk about a few:

  (iii) Authorization of new classes of stock—This one is pretty important because it is the mechanism that ensures that the Preferred get a chance to vote on future financings for the company. After all, to sell more stock as part of a new financing round, the company will likely need to issue a new class of stock to the new investor. Sometimes this section just gives a blanket vote to the Preferred to approve (or not) the creation of a new class of stock. In our term sheet, VCF1 and XYZ Company have agreed on a middle ground; the Preferred only get to vote on a new class issuance if it has rights equal to or greater than the rights that the current classes of Preferred have. For example, if XYZ Company wanted to issue a new class of preferred stock that had a liquidation preference that was junior to the preference of the VCF1 stock, arguably they could do that without consent of the Preferred.

  (v) Corporate actions—This one is also pretty important. It permits the Preferred to vote on an acquisition of the company and a sale of its intellectual property. This makes sense in that VCF1 likely invested in the company in part based on its intellectual property and thus would want to have a say in a sale of that and a sale of the company itself (which would likely include the intellectual property).

  (vii) Liquidation or recapitalization—If XYZ Company is going to get shut down (liquidated) or recapitalized (meaning that the current capitalization structure is going to get turned on its head), the Preferred will also get to have a say in these matters. We’ll talk more about a recapitalization later in the book, but briefly it is a transaction by which the ownership structure of the company is largely reset. For example, preferred shareholders might get forced to convert into common shareholders (to eliminate liquidation preference), or the ownership of existing shareholders might get reduced by what’s called a “reverse split.” This means that the number of shares that a stockholder has is reduced by some multiple. The purpose of this is to reduce the percentage ownership of these particular shareholders to enable the company to sell stock to new shareholders, such that the new shareholders can own a meaningful percentage of the company.

  (xi) Increases in the option plan—Recall that startups generally use stock options to incent employees. As the company grows, it may run out of options in the plan and thus need to increase the option pool in order to grant more equity to its employees. But doing so dilutes the existing shareholders. Hopefully that’s intuitive, but any time you have to add more shares to the company, the denominator of total shares outstanding increases, so if VCF1 owns 20 percent of the company before adding the new shares, it will own less than that amount after the addition of the new shares to the option pool. Understandably, then, the Preferred would like to be able to weigh in on the decision to increase the option pool.

  When you take a step back from these protective provisions, you’ll see that they really are designed to protect the economic interests of the Preferred shareholders. All the enumerated things we talked about above are essentially things that affect the economics of the investment—raising more money (by issuing new shares), selling the company or its intellectual property, liquidating or selling the company, and increasing the option pool. Thus, the protective provisions are really just as they sound—protection against an erosion of the economic value that VCF1 thought it was getting when it invested its $10 million.

  Registration Rights

  Our term sheet takes a bit of a shortcut by saying that VCF1 gets “customary registration rights.” Luckily for practitioners in the field, they understand what that means. This section is not worth spending much time on, not because it’s not important, but because it tends to be pretty noncontroversial when the lawyers are advising their clients on key elements of the term sheet. It’s also not heavily negotiated because it really doesn’t matter until the company is ready to go public and, at that time, the investment bankers will basically tell both the company and the investors what they think the appropriate market terms are for these provisions.

  At a high level, this section deals with what happens typically when a company goes public and VCF1 wants to sell its shares in the public market. In general, under US securities laws, shares need to be registered with the SEC in order to be fully liquid. Unregistered shares can be sold only if they comply with various exemptions from registration that the SEC has outlined, but in most cases the volume of shares and the timeline over which those shares can be sold will be restricted. The registration rights section defines under what circumstances VCF1 can either require XYZ Company to register its shares or “piggyback” on a registration of other classes of shares. Piggyback means exactly what it sounds like: if the company is otherwise registering some of its shares of stock, VCF1 gets the ability to piggyback on that registration and have its shares registered as well in the same process.

 
Pro Rata Investments

  We talked earlier about how VCs often reserve dollars for follow-on investments in a company after their initial investment round. The reason for this is that if VCF1 owns 20 percent of XYZ Company today and the company continues to do well, VCF1 might want to invest additional dollars in subsequent financings rounds to maintain its ownership. Otherwise, because a new round of financing entails the issuance of new shares of stock, VCF1 will be diluted by the financing round.

  This section gives VCF1 the right, but not the obligation, to purchase its pro rata amount of future rounds of financing to avoid dilution. You’ll also note that the right applies only to “major investors,” which was defined in the information rights section to be anyone who invests at least $2 million in the company.

  This is really just a matter of convenience. If you have a lot of small investors, sometimes it is just a pain to track them down and then wait for them to tell you whether they are going to invest their pro rata amount in the new round of financing. So the major investor definition reserves this right for those who are putting some material amount of money into the company. Of course, the threshold could be set at any level the company and investors agree to.

  Pro rata rights seem like a fair thing for existing investors to have, but oftentimes it creates challenges in a financing round. Specifically, when a company is doing really well and the financing round is oversubscribed, meaning there are more investors who want to put money in than the company cares to accommodate, pro rata rights can become an issue.

  Why is that? Because the new investor often wants to achieve some target percentage ownership. The reason for this is as follows: the critical scaling limiter for VC funds is the number of board seats that any particular GP can handle. While there are no rules here, it’s generally the case that GPs tap out around ten to twelve board seats. So every time a GP takes on a new board seat, that investment decision has an opportunity cost in that it consumes one of a limited number of investments she can make.

  Thus, in order to make more investments and scale the firm, VCs need to carefully watch the number of board seats they take and, when they make a decision to take on a board obligation, VCs want to own as much of any company as possible. A cardinal sin of venture investing is picking the right company in which to invest but failing to own enough of the company, such that the returns from that investment don’t meaningfully move the needle on the fund’s overall returns. Nobody wants to make that mistake, thus the desire to maximize ownership on each investment.

  Naturally, the other tension comes from the company itself. There is a finite amount of money the company likely wants to raise at a particular financing round and, correspondingly, a maximum amount of ownership dilution it is willing to accept. So if the company wants to raise a total of only $15 million and sell 10 percent of the company for that amount, the new investor may desire to take the full $15 million. But if the existing investors all have pro rata rights, they are entitled to their portion of the $15 million as well. At some point, either the company has to agree to raise more money (and thus potentially suffer greater dilution) or the new and existing investors have to come to some agreement between themselves.

  Stock Restriction

  This section packs a lot of punch, so we’ll need to break it down into its various sections. Notice that it applies to significant shareholders only; in this case, anyone who owns at least 2 percent of the company’s stock.

  Let’s start with the right of first refusal (ROFR)—this means that if I want to sell some of my stock, I can go to an outside party and get them to give me a price. But before I am allowed to sell my shares to that third party, I need to give the company first (and then the investors) the right to purchase the stock at the same price.

  You can think of this provision in two ways.

  In the most restrictive version, it really is intended to put a chilling effect on anyone’s ability to sell shares. After all, if I am a third-party potential buyer and I know that I have to set a bid price and then give the company and investors the right to match that and thus take the deal away from me, I am not likely to bid in the first place.

  The more innocuous interpretation of a ROFR is that it allows the company and its investors to control in whose hands the stock ends up. This is important for many startup companies because they generally don’t want potentially unknown investors holding significant shares of the company and thus being able to influence corporate decisions through the stock voting rights. The ROFR gives the company an opportunity to determine if they are comfortable with the third party before they approve the transaction, in which case they will waive the ROFR. If they are not comfortable selling to that third party, the company can pony up the money and effect the purchase itself.

  The co-sale agreement is kind of the inverse of a ROFR. A co-sale says that if I offer to sell my shares to a third party (or anyone else), all the other investors get a right to also sell their pro rata portion of shares to that buyer at the same price. In other words, I do all the work to generate a deal for my shares and then risk being cut back because other shareholders can pile onto my deal. For example, if I want to sell 10 percent of my shares, I need to identify a buyer who is willing to make that purchase. If the buyer agrees, then every other shareholder of the company who has the co-sale right gets to sell a corresponding portion of their shares in the transaction. Since the buyer likely doesn’t have endless amounts of money, nor an endless appetite to purchase shares, the number of shares that I am ultimately able to sell will get reduced by the number of shares that the other shareholders choose to sell.

  You might wonder why either of these provisions exists in the first place. Well, in general they are intended to make it harder (in this case) for common shareholders, in particular the founders, to sell their shares. And why would the VCs want this? Because the VCs are investing in the founders (who are usually the major holders of common stock) and they want to maintain maximum alignment with them to grow the equity value of the company. So, by restricting their ability to sell the shares, we are all in this together—we all either succeed or fail together, but nobody jumps ship too early.

  Now if you look to the last paragraph of this section, you’ll see that this alignment seems to be shattering. That section says that everyone (except for the Preferred stock) will have a blanket transfer restriction on stock, other than with approval from the disinterested board members. Cutting through the legal language, the provision says that the common shareholders can’t sell at all unless the noncommon board members say it’s okay. But the Preferred are exempted from this restriction, freeing them to sell without that approval. That seems a bit one-sided, and it is. Often you will see the more complete version of alignment where everyone is bound by the same stock transfer restrictions—we all hold hands and jump off the bridge together at the same time, or none of us does!

  Drag Along

  The drag-along provision is intended to prevent minority investors from holding out on a deal to try to get a better deal for themselves. So what it says is that if each member of the board of directors, the majority of common stock, and the majority of Preferred stock all vote in favor of an acquisition, then any of the other 2 percent shareholders (recall this was our major investor definition) gets dragged along in favor of the deal. The presumption here is that if all these other folks have decided that the acquisition is a good thing for the company, there is no reason that a small shareholder should be able to prevent the transaction from happening; tyranny of the minority is outlawed through the drag along.

  To provide some protection to these 2 percent holders, note that there are three separate votes required: (1) the board needs to approve; (2) the common voting as a separate class need to approve; and (3) the capital “P” Preferred voting as a separate class need to approve. Thus, there are some protections built in to make sure that the deal hopefully is in fact a good one for the company
to pursue.

  One other thing to note here: the drag along doesn’t apply to the shareholders with less than 2 percent. Why is that? Well, first as a practical matter, acquirers will sometimes close an acquisition with fewer than 100 percent of the shareholders agreeing, as long they get probably more than 90 percent, and all the big players, to agree. Second, you worry less about tyranny of the minority with very small (and often disparate) shareholders—that is, the likelihood of their being able to scuttle the deal or hold out for something better is just much less. So, in a way, we tolerate democracy as long as a democratic process can’t in fact affect the outcome!

  D&O Insurance

  This is a minor (and hopefully) noncontroversial section of the term sheet, but worth a quick mention as we will spend more time on liability later in the book. Recall that when we talked about the GP equity partners agreement, we mentioned indemnification—the idea that the fund can protect individual GPs from having to pay out of pocket for potential lawsuits that arise from their role. That is all true—and VC firms purchase their own D&O (directors and officers) insurance to help cover these potential expenses.

  But there’s never any harm in being extra careful—wearing both a belt and suspenders is a good way to make sure your pants stay up—so the portfolio company itself also purchases its own D&O insurance to protect its board members and officers against legal liability.

  The VCF1 GP who sits on the board of XYZ Company thus has several lines of defense: XYZ’s D&O policy first and then VCF1’s policy as a backup. And just as the venture firm indemnifies its GPs, XYZ Company will also indemnify its board members and officers. This enables them to be the beneficiaries of the D&O insurance policies.