Co-underwriting wins allocation and access. The bill arrives only when the sell-down fails, which is exactly when the deal is worth least.
A co-underwriter should be paid at least the expected loss on the part of the stake it fails to sell down, priced on the assumption that sell-downs fail when the deal has gone wrong. On an illustrative 100 million co-underwrite with 60 million to syndicate, a 25 per cent chance of the sell-down falling short and a 14.5 per cent loss on whatever is left over, the expected cost is 1.52 million: a break-even fee of 2.54 per cent of the amount offered for syndication, or 1.52 per cent of the whole cheque.
Worked in full in The Co-Investment Practitioner by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Co-underwriting is the step up from ordinary co-investment. Instead of waiting for the sponsor to syndicate after closing, the LP commits a larger cheque at signing, helps the sponsor give the seller certainty, and sells the excess to other investors in the months after closing. The reward is access and allocation. The risk is that the sell-down does not happen, and the LP is left holding a position it never meant to own, at exactly the moment it is worth least.
| Input | Value |
|---|---|
| Equity cheque for the deal | 400 million |
| Taken by the sponsor's fund | 250 million |
| Co-investment, underwritten at signing by one LP | 100 million |
| LP's intended final hold | 40 million |
| To be syndicated within 6 months | 60 million |
| Syndication price | Cost plus 6.0% a year |
| Probability: all 60 sold / half sold / none sold | 75% / 15% / 10% |
| Markdown of the deal when the sell-down fails | 5.0% |
| Secondary discount to that mark on the excess | 10.0% |
| LP's co-investment programme | 600 million |
The syndication price matters only for carry. Sold at cost plus 6.0 per cent a year, the 60 million that is placed after six months returns 1.80 million of carry, which by assumption is what the money cost the LP. On a successful sell-down the co-underwriter therefore makes nothing and loses nothing. Everything below is about the unsuccessful ones.
First, what happens to the excess. When buyers do not appear it is rarely random: the first quarter's numbers have disappointed, the sector has re-rated or the debt is trading down. So the failure case is priced as a 5.0 per cent markdown of the deal, followed by a sale of the unwanted stake on the secondary market at a further 10.0 per cent discount to that mark.
Loss rate on unsold stake = 1 − (1 − markdown) × (1 − secondary discount) = 1 − 0.95 × 0.90 = 14.5%
Expected loss = Σ probability × unsold amount × loss rate
In Excel, with probabilities in B2:B4, unsold amounts in C2:C4 and the loss rate in B6: =SUMPRODUCT(B2:B4,C2:C4)*B6
| Outcome | Probability | Unsold | Loss if it happens | Weighted loss |
|---|---|---|---|---|
| Full sell-down | 75% | 0.0 | 0.00 | 0.00 |
| Half sold | 15% | 30.0 | 4.35 | 0.65 |
| Nothing sold | 10% | 60.0 | 8.70 | 0.87 |
| Expected | 10.5 | 1.52 |
The expected cost of 1.52 million can be recovered in different currencies, and the percentage depends on the base:
A sponsor offering "1 per cent on what you syndicate" is offering about a third of the risk on these assumptions. The co-underwriter may still accept, because the access is worth something and because a co-underwriter that is reliable gets shown the next deal first, but it should know how much of the price is being paid in relationship rather than in cash.
Holding the excess instead of selling it does not avoid the cost; it changes its form. A 100 million position is 16.7 per cent of a 600 million programme against a planned 6.7 per cent, which is a concentration most programmes cannot carry within their own limits, a point worked in how many co-investments an LP needs to diversify.
| Failure odds (half / none) | No markdown, 10% discount | 5% markdown, 10% discount | 10% markdown, 15% discount |
|---|---|---|---|
| 7.5% / 5.0% | 0.87% | 1.27% | 2.06% |
| 15.0% / 10.0% | 1.75% | 2.54% | 4.11% |
| 30.0% / 20.0% | 3.50% | 5.08% | 8.22% |
The fee scales linearly with the failure odds and rises with the loss rate, and the two tend to move together: the deals that are hardest to syndicate are the ones where the mark falls before buyers commit. A large, cyclical or highly levered deal belongs in the bottom-right corner, and the fee that makes sense there is several times the one that makes sense for a defensive business with a crowded co-investor list.
Pricing the failure with the secondary discount alone. Assume the deal holds its value and only a 10.0 per cent discount is needed to clear the excess, and the expected loss is 1.05 million, a break-even fee of 1.75 per cent. That understates the cost by 31 per cent, because it treats the sell-down as failing for reasons unrelated to the deal. In practice failure and markdown are the same event, and the price has to be set on the joint case.
The opposite error is to count the carry. Once the excess is sold at cost plus interest, as in what buying a co-investment at cost plus interest costs, the co-underwriter has been reimbursed for funding it; the carry is not a cost of underwriting and should not be in the fee.
Price a co-underwrite as a short option on the syndication: probability of failure, times the amount left over, times the loss on that amount in the state of the world where it is left over. On the illustrative terms that is 1.52 million, 2.54 per cent of what is offered for sale. Below that, the co-underwriter is paying for access, which may be a good trade, but should be a deliberate one. The programme-level model behind this book's co-investment numbers is in the free workbook for this case.
An LP commits a larger co-investment at signing than it intends to keep, so the sponsor can give the seller certainty, then sells the excess to other investors after closing. On an illustrative deal the LP signs for 100 million, plans to keep 40 million and syndicates 60 million within six months, usually at cost plus interest so that its carry is reimbursed.
Because sell-downs fail when the deal has disappointed. If the excess is sold at a 10.0 per cent discount with no markdown, the illustrative break-even fee is 1.75 per cent of the amount offered. Add the 5.0 per cent markdown that typically accompanies a failed syndication and the loss rate becomes 14.5 per cent and the fee 2.54 per cent, 45 per cent higher. Put the other way, the discount-only fee understates the cost by 31 per cent.
On the amount offered, if possible. A fee paid only on what is sold vanishes in the cases that cause the loss, so it has to be higher to compensate: 3.08 per cent of an expected 49.5 million sold, against 2.54 per cent of the 60 million offered. Across the full 100 million cheque the same cost is 1.52 per cent.
This article is one calculation from The Co-Investment Practitioner. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: private equity and private markets → · All 1501 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.