A compounding preferred return makes carry a function of time: the same exits, shifted by one and two years, through a whole-fund waterfall.
Often all of it. On an illustrative 300 million real estate fund returning 1.60x with an 8 per cent compounding preferred return and a full catch-up, carry on plan is 36.0 million; one year late it is 21.04 million, and two years late it is zero. The profit is the same 180 million in every case. Only the calendar moves, and the hurdle grows with it.
Worked in full in Raising a Real Estate Fund by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
The fund draws 300 million in three equal calls and sells four assets in years 4 to 7. Cash flows are to the limited partners, net of management fees and expenses, before carry. The terms are a common package for a first-time real estate manager: whole-fund (European) waterfall, 8 per cent preferred return compounded annually, 20 per cent carry, 100 per cent catch-up.
| Item | Year | Amount |
|---|---|---|
| Capital call | 0 | 100 |
| Capital call | 1 | 100 |
| Capital call | 2 | 100 |
| Exit, asset 1 | 4 | 90 |
| Exit, asset 2 | 5 | 130 |
| Exit, asset 3 | 6 | 140 |
| Exit, asset 4 | 7 | 120 |
| Proceeds 480 on 300 drawn | 1.60x |
On plan the fund's return before carry is 10.80 per cent, 2.80 points above the hurdle. That margin is what the delay has to eat through, and it is thinner than a 1.60x multiple makes it look.
Hurdle balance = last year's balance × 1.08 + new calls − distributions to LPs
Each distribution goes first to the LPs until the hurdle balance is zero. The excess then goes to the GP until the GP holds 20 per cent of all profit distributed (the catch-up), and anything left is split 80/20.
Catch-up needed = (20% × profit distributed so far − carry already paid) / (1 − 20%)
On plan the first three exits all go to the LPs. Before the year 7 exit the hurdle balance is 60.80. The LPs take that, the GP takes a catch-up of the next 30.2, and the last 29.0 is split, giving the GP 36.00 in all: exactly 20 per cent of the 180 profit.
| Exit | On plan | One year late | Two years late |
|---|---|---|---|
| Asset 1 (90) | 378.66 | 408.95 | 441.67 |
| Asset 2 (130) | 311.75 | 344.47 | 379.80 |
| Asset 3 (140) | 196.29 | 231.63 | 269.79 |
| Asset 4 (120) | 60.80 | 98.96 | 140.17 |
| Carry to the GP | 36.00 | 21.04 | 0.00 |
One year late, the last exit of 120 meets a hurdle balance of 98.96. The LPs take 98.96 and the remaining 21.04 all goes to the GP as catch-up, which never completes. Carry falls by 41.5 per cent. Two years late, the hurdle balance before the last exit is 140.17, larger than the exit itself: the LPs take all 120 and the GP gets nothing. The 180 million of profit is unchanged, but at 8 per cent compounding the investors are owed more than the fund makes.
| Delay, years | Return before carry | Carry | Change vs plan | PV of carry at 10% |
|---|---|---|---|---|
| 0 | 10.80% | 36.00 | 0.0% | 18.47 |
| 1 | 8.78% | 21.04 | −41.5% | 9.82 |
| 2 | 7.39% | 0.00 | −100.0% | 0.00 |
| 3 | 6.39% | 0.00 | −100.0% | 0.00 |
In present value the damage is worse than in the nominal column: a one-year slip roughly halves the value of the carry to the house, from 18.47 to 9.82, because what survives also arrives later.
Delay does not erode carry in proportion. It pulls the fund's return down towards the hurdle, and once the return is at or below 8 per cent the promote disappears in one step. A fund 2.80 points above its hurdle on plan has roughly one year of slack.
Scale every exit up and rerun the waterfall with the two-year delay. Carry only starts once the fund returns more than 1.66x, and the catch-up completes, restoring a full 20 per cent of profit, only at 1.80x. At that multiple the carry is 48.25. Put the other way, a manager who expects its business plans to slip by two years needs to underwrite roughly 0.20x more multiple than its plan shows simply to stand still on carry.
A softer catch-up does not help. With an 80 per cent catch-up the one-year-late carry is 16.83 rather than 21.04: the catch-up tier is where a late fund's carry lives, and a slower catch-up pays the GP even less of it.
Managers model carry as 20 per cent of profit and then sensitise the profit. That treats carry as linear in performance and silent about time. In a compounding-hurdle waterfall the timing of distributions is a first-order input. The carry projections in a fundraising deck, and the house's budget built on them, should show the delayed case beside the base case. On this fund a projection of 36.0 million is, on a one-year slip, 21.04, and on a two-year slip nothing at all, with the management company's costs unchanged.
Investors will run this case in diligence whether or not the manager does. A track record of business plans that ran late is read as a carry projection that will not arrive, and as an incentive problem in the later years of the fund.
Chapter 10 of Raising a Real Estate Fund tells the manager to model the case in which everything is realised two years later than planned, and the companion's first case runs it on the book's own four-asset fund. Both are in the free workbooks for this book. For why the multiple at which carry switches on depends on timing, see at what multiple carry switches on; for how the GP's economics are built up from fees and carry, see how much a real estate fund manager earns from fees and carry.
Because the preferred return compounds while the profit does not. Each year of delay adds 8 per cent to what the LPs are owed before the GP sees anything. On the illustrative fund the hurdle balance before the last exit rises from 60.80 million on plan to 140.17 million two years late, more than the 120 million exit, so the carry falls from 36.0 million to zero.
Measure it in return, not multiple. A fund returning 10.80 per cent before carry against an 8 per cent hurdle has 2.80 points of slack, which absorbs roughly one year of delay: carry falls 41.5 per cent at one year and disappears at two. Two years late, the same fund needs 1.80x rather than 1.60x to earn a full 20 per cent of profit.
The GP's final entitlement falls broadly the same way under both, because a whole-fund clawback brings deal-by-deal carry back towards the whole-fund answer. The difference is who carries the risk in the meantime: deal by deal, early exits pay carry that a later delay turns into a clawback, so the investors bear it. In a whole-fund waterfall the GP simply waits and, here, receives 21.04 million one year late.
This article is one calculation from Raising a Real Estate Fund. 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: real estate investing, finance and fund management → · All 1614 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.