A redemption gate is only a liquidity management tool if the fund can actually pay the gated amount every dealing date of a run, so the calibration is a multi-quarter cash projection, not a one-quarter liquidity profile.
A redemption gate should be calibrated to the largest share of NAV the fund can pay on every dealing date of a run of gated quarters, without selling the loans it means to hold. For an illustrative 400 million open-ended credit fund that can pay 15.20 per cent of NAV in a single quarter, the gate that survives four consecutive quarters is 7.23 per cent of NAV, so the policy gate is set at 7 per cent. A 10 per cent gate looks prudent and fails after one quarter.
Worked in full in The AIFMD II Handbook by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
AIFMD II requires the manager of an open-ended AIF to select at least two liquidity management tools from the directive's list, and a redemption gate is one of them. The question the board and the supervisor will ask is not whether a gate exists but whether its level can be honoured. A gate set at the first quarter's liquidity cannot: once the stock of liquid assets is spent, the next dealing date arrives with the same gate and no cash behind it.
| Item | Amount | % of NAV | Stress assumption |
|---|---|---|---|
| Cash | 24.0 | 6% | 2% of NAV kept as an operating buffer |
| Liquid syndicated loans | 32.0 | 8% | Sold at a 3% haircut when needed |
| Originated loans held to maturity | 344.0 | 86% | Never sold; repay 4% of the book a quarter |
| NAV | 400.0 | 100% | |
| Undrawn borrower commitments | 30.0 | 20% of what remains drawn each quarter | |
| Net interest income | 1.5% | Of NAV a quarter, retained as cash |
The test is simple to state: every quarter the gate is hit in full, the fund pays redeeming investors the gate times the NAV of that dealing date, and it fails the quarter it would have to sell an originated loan to keep its 2 per cent cash buffer.
Step 1: the stock. Cash above the buffer is 24.0 minus 8.0, so 16.00. The liquid loans raise 32.0 × 0.97 = 31.04. The stock is 47.04 million, 11.76 per cent of NAV, and it can be spent only once.
Step 2: the recurring flow. In the first quarter the loan book repays 4 per cent of 344.0, that is 13.76; borrowers draw 20 per cent of the 30.0 undrawn, 6.00; income adds 6.00. Net flow is 13.76 million, 3.44 per cent of NAV. It repeats every quarter, shrinking slightly as the book runs down.
Step 3: the one-quarter answer, and why it is wrong. Stock plus flow is 60.80 million, 15.20 per cent of NAV. That is what the fund can pay once. A rough four-quarter answer spreads the stock: 3.44 + 11.76 ÷ 4 = 6.38 per cent. It is close but conservative, because the gate is applied to a NAV that falls every quarter and the borrower draws taper off as the undrawn book is used.
Step 4: project the run. Solve for the highest gate that clears four gated quarters.
Each quarter: casht = casht−1 + repayments − draws + income − gate × NAVt; if cash < 2% × NAV, sell liquid loans at (1 − haircut); if none are left, the gate has failed.
In Excel, lay the quarters across columns with the gate in one input cell, flag the first column where liquid loans would go negative, and use Goal Seek (or Data Table over gate levels) to find the highest gate whose flag stays clear through column four. The answer here is 7.23 per cent.
| Quarter | NAV start | Gated redemptions | Repayments | Draws | Income | Liquid loans sold | Liquid left | NAV end |
|---|---|---|---|---|---|---|---|---|
| Q1 | 400.0 | 28.0 | 13.8 | 6.0 | 6.0 | 0.0 | 32.0 | 378.0 |
| Q2 | 378.0 | 26.5 | 13.4 | 4.8 | 5.7 | 9.8 | 22.2 | 356.9 |
| Q3 | 356.9 | 25.0 | 13.1 | 3.8 | 5.4 | 10.3 | 11.9 | 337.0 |
| Q4 | 337.0 | 23.6 | 12.7 | 3.1 | 5.1 | 8.8 | 3.1 | 318.2 |
| Four quarters | 103.0 | 3.1 | 318.2 |
In the first quarter the fund needs no sale: 24 + 13.76 − 6.00 + 6.00 − 28.00 leaves 9.76 of cash against a buffer of 7.56. From the second quarter it sells liquid loans, losing 0.87 million to haircuts over the year. After four full gated quarters investors have received 103.0 million, 25.8 per cent of the starting NAV, and 3.1 million of liquid loans remain. A fifth quarter at 7 per cent would force the sale of an originated loan.
| Quarterly gate, % of NAV | Quarters paid |
|---|---|
| 5.0% | 12+ |
| 6.0% | 8 |
| 7.0% (policy) | 4 |
| 7.5% | 3 |
| 10.0% | 1 |
| 15.0% | 1 |
| Change | Gate, % of NAV |
|---|---|
| Base case | 7.23 |
| Repayments 2% of the book a quarter | 5.33 |
| Repayments 6% of the book a quarter | 9.13 |
| No undrawn commitments | 8.51 |
| Liquid loan haircut 10% | 7.08 |
| No liquid loans at all | 4.95 |
The gate is mostly a function of the loan book's repayment speed in stress. The haircut barely matters because the liquid sleeve is small; the undrawn commitments cost 1.28 points of gate, more than half the 2.28 points the whole liquid sleeve contributes. The duration of the run is the other choice: the same fund supports 15.48 per cent for one quarter (slightly above the 15.20 of Step 3, because the projection holds the 2 per cent buffer on the smaller post-redemption NAV), 7.23 per cent for four and 6.07 per cent for eight. The same projection run the other way, fixing the gate and solving for the cash to hold, is how big a liquidity sleeve should be. A monthly-dealing fund would express the 7 per cent as roughly 2.33 per cent a month, but should rerun the projection monthly rather than divide.
Calibrating the gate to the Annex IV liquidity profile or a one-quarter redemption coverage ratio. Both measure the stock of liquidity at one date, and both say this fund can pay about 15 per cent. Investors who are refused at one dealing date resubmit at the next, so a gate that binds is hit again. A gate at 10 per cent here is honoured once and then requires the very asset sales it was meant to prevent; the multi-quarter queue that results for investors is worked in how long it takes to exit a gated fund.
All figures are illustrative. The repayment rate, draw rate and haircuts are stress assumptions the manager must calibrate and back-test on its own book, and the run length is a policy choice to document.
The book's semi-liquid credit fund, with its facility, undrawn book and gate computed side by side, is in the free workbook for this case.
Many semi-liquid funds set gates of around 5 per cent of NAV a quarter, but the right level is fund specific. In this illustrative credit fund, 5 per cent can be paid for more than twelve consecutive quarters, 7 per cent for four, and 10 per cent for only one before originated loans would have to be sold. The calibration should show that arithmetic, not cite a market convention.
No. First-quarter capacity counts the whole stock of cash and liquid loans once: here 47.04 million of stock plus 13.76 million of net quarterly inflow, 15.20 per cent of NAV. A gate at that level empties the stock in one dealing date, and the second gated quarter forces sales of loans the fund intended to hold. A gate exists for runs that last several quarters.
Yes, and it is the input that moves the answer most. Halving stressed repayments from 4 to 2 per cent of the loan book a quarter cuts the four-quarter gate from 7.23 to 5.33 per cent of NAV; at 6 per cent it rises to 9.13. Repayment speed in a stress should be back-tested against the fund's own loan book, because it is the recurring source of cash.
This article is one calculation from The AIFMD II Handbook. 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: regulation, compliance and banking → · All 1340 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.