Anti-dilution: the unloved child
There are few provisions in venture financing agreements that are more complicated than anti-dilution clauses. In contrast the rationale is rather simple: Compensate investors for some of the dilution they suffer in a so-called “down round”, i.e. if in a subsequent funding round shares are issued at a lower price than the price paid by the investors. One can (and we think should) discuss the adequacy of such provisions (the investor gets the full upside without the full downside), but since they — just like liquidation preferences — are, and for a long time have been, a standard element of venture financing rounds, this blog post will focus on the “how” rather than the “why”.
One clarification of terminology upfront: At times people think of anti-dilution protection as the right of shareholders to invest their pro rata portion in a new capital raise in order to avoid any dilution by the newly issued shares. While this concept definitely is a prime example of dilution protection, it is generally referred to as pre-emption or pro rata right. The anti-dilution which we’ll look at in this post means the additional protection granted to investors in a down round by giving them some shares for free.
Broad-based vs. narrow-based vs. full ratchet
The main reason why anti-dilution clauses are complicated (and — judex non calculat — unloved by many lawyers) is the formula to calculate the number of free shares to be issued to the existing investors. There are three different approaches that result in very different numbers, so unfortunately it is not possible to ignore the math altogether.
The most common approach is referred to as “broad-based weighted average”, which, simply spoken, means that the previous investors get that number of free shares that brings their overall average purchase price per share (including the new shares they get for free) down to the average price between the earlier round and the down round. This average price is calculated on a “broad basis”, meaning the earlier price is weighted by the entire fully diluted capital (= fully diluted shares) prior to the down round, whereas the down round price is weighted (only) by the number of shares issued in the down round.
“Narrow based” in contrast means the earlier price is weighted only by the number of shares issued in the prior funding. In this case, the weighted average comes out much closer to the down round price, meaning more shares have to be issued to the existing investors for free.
And full ratchet is not a weighted average at all. It simply means the existing investors get that number of shares that is needed to push their overall price down to the down round’s price.
As you can see, broad-based weighted average is the most founder-friendly option because it dilutes existing shareholders the least*. It’s also by far the most common approach, even though there still are some shareholders’ agreements with a narrow-based formula. Full ratchet is, fortunately, unheard of these days. Given the great prevalence of the broad-based approach, we’ll focus on this one from hereon.
A sneak peak at the math
Anti-dilution formulas are intimidating. That’s not because the math is particularly demanding, it’s more because the formulas consist of a number of letters or strange acronyms which are defined in complex legal prose in a separate list. There are principally two different approaches to calculate the number of anti-dilution shares, which appear quite different from the face of it, but upon closer inspection follow the same logic and come out at the same result:
The first approach, mostly used in Europe, calculates the weighted average share price (which, as a reminder, determines the number of free shares to be given to investors in order to lower their overall price per share) by adding the proceeds raised in the down round to the fully diluted value of the company after the prior round, and then dividing the result by the fully diluted number of shares (including the down round shares). The ratio between the weighted average price and the (higher) price paid by the existing investors is then used to calculate the number of anti-dilution shares:
The alternative approach, which is standard in the US, works with the so-called conversion ratio, which is the number of common shares that a preferred share can be converted into (esp. in case of an exit). The conversion ratio is initially 1 and recalculated in case of a down round. The investor does not immediately get additional shares when the down round is completed, but the relative weight of his or her preferred shares on an “as converted” basis goes up as the conversion ratio increases.
The new conversion ratio is derived by dividing the prior conversion price (which initially is the price at which the shares were purchased) by a new conversion price, which in the standard US templates is calculated as follows:
The fraction in the above formula comes out below 1 in a down round, so the new conversion price is lower than the prior conversion price. If this sounds familiar, it’s because it is: The new conversion price is the same as the WSP in the European approach. Likewise, the new conversion ratio is the same as the fraction used in the “European” formula above:
So the European and the US approach are just different expressions of the same concept with the main difference that in the European approach the WSP is used to calculate a certain number of additional shares that the investor actually subscribes to in connection with the down round, whereas in the US that same number of (common) shares is included in the number of shares that the investor shares convert into if and when they are converted.
If you’re not as excited by this as some math geeks are, don’t despair. To get behind anti-dilution it’s not necessary to understand the mathematical connections in detail. It’s entirely sufficient to know that the formulas are based on a ratio which compares the price of the earlier issued shares with a weighted average price of the prior round and the down round. Once you actually apply the formula (or check that your lawyer got it right), working the math is much easier than may appear at first sight. Except in one case…
The circular problem
If you have been through a funding round you may have seen that the investor(s) did their calculations based on a (generally: pre-money) valuation “on a fully diluted basis”. Background is that an investor who, say, invests EUR 2 million at a pre-money valuation of EUR 8 million wants to be certain that he or she ends up with a stake of 20%, even after all “equity-like” instruments are accounted for. This means all issued or authorised options, warrants, convertibles and similar securities are to be included in the fully diluted capital that is used to calculate the price at which the investor invests (purchase price).
Now, if the funding round is a down round and anti-dilution shares are being issued to existing investors as part of the round (or, in the US approach, included in the fully diluted capital because it’s calculated on an “as converted” basis), the number of anti-dilution shares will depend on the purchase price. The purchase price in turn will be affected (decrease) in parallel to the number of anti-dilution shares issued. Hence there is a mutual dependency that leads to an iteration where two numbers (shares purchased by the investor(s) on one side and anti-dilution shares on the other side) push each other down in an endless spiral. If you try to model this in Excel you end up with a circular reference and share numbers and prices that are approaching the infinite and infinitesimal, respectively.
Obviously, this is not the idea of anti-dilution. To make the numbers work the iteration has to stop at some point. Because conceptually anti-dilution is a one-time adjustment, the cut-off should be after the first iteration. This means in an initial step the purchase price is calculated on a fully diluted basis, but without anti-dilution shares. Based on this purchase price the number of anti-dilution is calculated. This increases the fully-diluted capital and hence the investor(s) will — rightfully — demand an adjustment of the purchase price, but that’s it, the number of anti-dilution will not again be adjusted. If you’re a founder (or early investor not benefitting from the anti-dilution), you should not accept any dilution above and beyond this first iteration.
Pay to play
While (broad-based) anti-dilution clauses as such are a standard element of early-stage funding rounds, not every detail is cast in stone. In particular, introducing certain requirements under which anti-dilution rights may be exercised are far from unusual, notably a pay-to-play requirement. This means that in order to receive anti-dilution shares the existing investor(s) must participate in the relevant funding round by purchasing their pro rata portion of the down round shares for cash like the other (new) investor(s).
Pay-to-play clauses, although definitely not unreasonable, are still rather rare, but they are certainly not unheard of. Ultimately, however, if an investor pushes back against a pay-to-play qualifier, founders might want to consider being generous on that one and keep their powder dry for more important issues (like governance). This is because, when push comes to shove and there indeed is a (material) down round, investors will generally want the people running the business to continue being sufficiently “invested”. It’s therefore far from unusual that the ESOP is expanded significantly to compensate founders and employees at least for some of the dilution they would otherwise face. In this case the brunt of dilution is borne by the very early investors with entry prices below the down round price who may well be washed out if they don’t invest in the round, but this is an inherent and generally accepted risk of early-stage venture investing.
What it boils down to
As a founder, if you’re faced with an intricate anti-dilution clause, keep calm and carry on. It’s fully sufficient if you ensure (= get your lawyer to confirm) that the formula is broad-based and in line with standard templates. You may also want to double-check that a few standard exceptions are included in the draft which do not trigger anti-dilution, notably the issuance of employee options (this is so obvious and uncontroversial that options would probably be considered exempt even without express wording). Beyond that you can test the waters regarding pay-to-play, although if there are more important issues still to be agreed, focus on those and use pay-to-play as an easy give-away (if at all). You may also be inclined to discount the relevance of anti-dilution clauses in the first place given the seemingly never-ending bull-market, which in past years has turned down rounds into an outright rarity. Yet as every seasoned investor knows, the past is not always a reliable indicator of the future… Therefore: A minimum amount of care for the seemingly odd contractual term makes sense, even if it feels like a waste of time.
Note:
* The data in the chart assumes there were initially 200 (common) shares and that in the “prior round” 100 preferred shares were issued, followed by another 100 shares in the down round.
