Full ratchet vs weighted average

What it means

Both are anti-dilution provisions, which repay an earlier investor when the company later issues shares more cheaply than they paid. They work by lowering the price at which that investor's preferred shares convert into common, which gives them more common shares for the same money. Full ratchet reprices their entire holding to the new lower price, however few shares were issued at it. Weighted average reprices in proportion to how much cheap stock was actually sold, so a small down round produces a small adjustment. Weighted average comes in a broad-based and a narrow-based form, which differ in how many shares go into the denominator of the formula.

Why it matters

The gap between the two is not a matter of degree, it is a matter of kind. Under full ratchet a single share issued at a low price can reprice an entire earlier round, so a small bridge financing on poor terms can transfer a large block of the company. Under weighted average the adjustment tracks the size of the event that caused it. Everybody who does not have the protection pays for it, which in practice means the founders and the employees, so this is one of the few terms where the interests of the team and of the earlier investors genuinely diverge.

Worked example

An investor put in EUR 2m at EUR 1.00 a share, holding 2,000,000 preferred shares against 10,000,000 outstanding. The company then raises EUR 2m at EUR 0.50 a share. Under full ratchet their conversion price drops to EUR 0.50 and their 2,000,000 shares become 4,000,000, doubling their position out of everyone else's. Under broad-based weighted average the new conversion price lands at about EUR 0.7333, giving them roughly 2,727,000 shares. Narrow-based, counting a smaller denominator, gives about EUR 0.7286 and slightly more shares. The two weighted average figures differ by less than half a cent, which is why they have to be compared at four decimal places rather than two.

The common mistake

Treating narrow-based weighted average as a reasonable compromise because it is not full ratchet. It is much closer to broad-based than to full ratchet in effect, so conceding it costs little, but that also means winning it gains the investor little, and the negotiation is often spent there while full ratchet elsewhere in the document goes unread. The other mistake is assuming anti-dilution protects against dilution generally. It does not. It is triggered only by a lower price, and an investor with full protection is diluted quite normally by a round priced above what they paid.

In practice

Broad-based weighted average is the market standard in Europe, the UK and the US, and a term sheet asking for full ratchet is asking for something outside the norm. It is most likely to appear in a bridge round, in a first institutional round with an inexperienced lead, or where an investor is pricing in real doubt. It is worth pushing back on hard and it is usually winnable, because most investors know the standard and are asking rather than insisting.

What to ask

Ask which of the three formulas the document uses, and if the answer is full ratchet, ask why, because the answer tells you whether it is a standard ask or a signal about how the investor sees the risk. Ask what the carve-outs are: most anti-dilution clauses exempt shares issued to employees, on conversion of existing instruments, and in certain acquisitions, and a clause with narrow carve-outs will fire on events nobody thinks of as a down round. Ask your lawyer to run the formula on a hypothetical fifty percent down round and tell you the resulting founder percentage. That single number turns an argument about a formula into an argument about a figure, which is a much shorter conversation. Then ask whether the protection expires after a set period or on a qualified financing, because many do and it is rarely mentioned unless raised.

Related terms

• Fully diluted shares The share count that assumes every option and every convertible has already converted. The only number worth quoting.
• Liquidation preference stack The order in which several rounds of preferred shares get paid when there is not enough to pay them all.
• Participating preferred stock Preferred shares that take their liquidation preference and then also share in what is left over.