The carve-out for borrowing covered by uncalled commitments is generous at launch and can quietly stop applying as capital is called.
Under AIFMD II, borrowing that is fully covered by investors' uncalled capital commitments does not count as exposure when a loan-originating AIF tests its leverage limit, so a covered subscription line drops out of the ratio. On an illustrative closed-ended credit fund, that takes year-2 leverage from 310.0 per cent, over the 300 per cent cap, to 235.0 per cent. The carve-out fails late in the investment period: once the line drawn exceeds uncalled commitments, the fund is no longer fully covered, and in year 3 the ratio is 312.5 or 287.5 per cent depending on how the uncovered part is read.
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 →
Directive (EU) 2024/927 caps the leverage of a loan-originating AIF at 175 per cent of net asset value if it is open-ended and 300 per cent if it is closed-ended, measured under the commitment method as exposure divided by NAV. A closed-ended direct lending fund typically uses two kinds of borrowing at once: a subscription line, bridging capital calls, and a back-leverage facility secured on the loan book. The carve-out treats them very differently, and the difference moves over the life of the fund.
| Snapshot | Called capital = NAV | Uncalled commitments | Subscription line drawn | Back-leverage drawn | Loans held |
|---|---|---|---|---|---|
| Year 1 | 100 | 400 | 120 | 0 | 220 |
| Year 2 | 200 | 300 | 150 | 270 | 620 |
| Year 3 | 400 | 100 | 120 | 730 | 1,250 |
Total commitments are 500. The subscription line is used the way most funds use it: drawn ahead of calls and repaid in batches, so it stays large while uncalled commitments shrink.
Leverage = exposure / NAV, with exposure = loans held = NAV + subscription line + back-leverage
Year 1: 220 / 100 = 220.0%. Year 2: 620 / 200 = 310.0%. Year 3: 1,250 / 400 = 312.5%.
Covered if subscription line drawn ≤ uncalled commitments
Year 1: 120 against 400, cover 3.33 times. Year 2: 150 against 300, cover 2.00 times. Year 3: 120 against 100, cover 0.83 times. Not covered.
Leverage = (loans held − covered subscription line) / NAV
Year 1: (220 − 120) / 100 = 100.0%. Year 2: (620 − 150) / 200 = 235.0%.
Excel, with loans in B, NAV in C, line drawn in D and uncalled commitments in E: =IF(D2<=E2,(B2-D2)/C2,B2/C2)
| Snapshot | Every borrowing counted | With the carve-out | Back-leverage capacity, line counted | Back-leverage capacity, carve-out |
|---|---|---|---|---|
| Year 1 | 220.0% | 100.0% | 80 | 200 |
| Year 2 | 310.0% | 235.0% | 250 | 400 |
| Year 3 | 312.5% | 312.5% or 287.5% | 680 | 680 or 780 |
In years 1 and 2 the carve-out does exactly what it was written to do. Capacity under the cap is 3 × NAV in exposure, so back-leverage capacity is twice NAV less any borrowing that counts. With the covered line excluded, the year-2 fund can draw up to 400 of back-leverage instead of 250: the carve-out frees one euro of capacity for every euro of covered line, 150 here. Counted in full, the line would have put the fund 10.0 points over the cap.
Year 3 is the trap. Uncalled commitments have fallen to 100 but the line is still drawn at 120, so the line is not fully covered and the wording no longer applies on its face. Two readings are possible:
The directive does not spell out the partial case, and a fund should not build its headroom on the generous reading without confirming it with its regulator and counsel. The prudent model uses the first.
Calling capital does not restore the cover. If the fund calls 20 and uses it to repay 20 of the line, uncalled commitments fall to 80 and the line to 100: the gap is still 20. What changes is NAV, which rises to 420 while the loans stay at 1,250, so the whole-line ratio comes down to 297.6 per cent, inside the cap but with only 2.4 points to spare. To restore the carve-out itself, the line has to be repaid from another source, for instance by moving 20 onto the back-leverage facility, which gives 287.5 per cent on any reading.
| Capital called and applied to the line | Uncalled commitments | Line drawn | Coverage gap | NAV | Leverage |
|---|---|---|---|---|---|
| 0 | 100 | 120 | 20 | 400 | 312.5% |
| 20 | 80 | 100 | 20 | 420 | 297.6% |
| 60 | 40 | 60 | 20 | 460 | 271.7% |
| 100 | 0 | 20 | 20 | 500 | 250.0% |
The coverage gap is fixed by the drawn line exceeding remaining commitments, so it survives any amount of calling. A covenant in the line that caps drawings at uncalled commitments, which a borrowing base set as an advance rate on uncalled commitments already does, prevents the gap opening at all. The mechanics are in how to calculate a subscription line borrowing base.
Excluding the subscription line permanently because it was excluded at launch. The test is a snapshot: cover that is 3.33 times in year 1 can be 0.83 times by year 3 without any new borrowing, simply because capital has been called. Run the coverage test at every leverage calculation date, next to the commitment-method arithmetic set out in gross vs commitment method leverage. An open-ended fund usually has no uncalled commitments, so for it the carve-out typically offers nothing: a fund with 200 of NAV, 30 of line and 100 of other borrowing is at 165.0 per cent against its 175 per cent cap.
A covered subscription line is outside the AIFMD II leverage ratio and frees a euro of back-leverage capacity for each euro drawn. Check the cover every time: here it moves from 3.33 times to 0.83 times over the investment period, and the year-3 ratio is a breach on the strict reading. The liquidity and leverage arithmetic of a loan-originating fund is worked in the free workbook for this case.
Yes. The carve-out covers only borrowing fully covered by investors' contractual capital commitments, which describes a subscription line, not a facility secured on the loan book. In the illustration, 730 of back-leverage on 400 of NAV is fully in the exposure. Under the 300 per cent closed-ended cap the fund's total exposure can be three times NAV, so back-leverage capacity is twice NAV less any borrowing that counts.
The line is no longer fully covered, so the carve-out does not apply on its face. In the illustrative year 3, a 120 line against 100 of uncalled commitments puts leverage at 312.5 per cent if the whole line counts, or 287.5 per cent if only the 20 of excess counts. The directive does not settle the partial case, so the strict reading is the prudent one to model.
Rarely. An open-ended fund usually has no uncalled commitments for a lender to rely on, so any line it draws is normally counted. With an illustrative 200 of NAV, 30 of line and 100 of other borrowing, its ratio is 165.0 per cent against the 175 per cent open-ended cap, leaving 10.0 points of headroom.
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 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.