A lock-up does not take cash away from equity, it parks it. What it costs is time, and the rate the parked cash earns while it waits for the coverage ratio to recover.
A distribution lock-up does not take cash from equity, it delays it, so its cost is the time value of the trapped cash. In an illustrative project whose cash flow runs 16 per cent below the base case, a lock-up at 1.20x DSCR traps distributions for eight years and cuts the equity IRR from 6.08% to 5.83%, a cost of 26 basis points. If the trapped cash is instead swept to prepay debt after two failed tests, the cost falls to 1 basis point.
Worked in full in The Infrastructure Investment Analyst by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Lenders set the lock-up between the base-case coverage ratio and the default ratio: high enough that shareholders stop taking cash before the project gets into real trouble, low enough that it does not bite in the base case. Equity analysts usually model it as a switch, run the downside, see that cash comes back eventually, and move on. The number worth knowing is how long it stays shut, and what the cash earns while it waits.
| Input | Value |
|---|---|
| Enterprise value | 100.0 |
| Equity invested | 30.0 |
| Senior debt, annuity repayment | 70.0 at 6% over 15 years |
| Annual debt service | 7.21 |
| CFADS, year 1 | 9.5, growing 1% a year |
| Downside: CFADS from year 2 | 16% below base case |
| Distribution lock-up, historic DSCR | 1.20x |
| Default covenant | 1.05x |
| Interest on the blocked account | 2% |
| Equity life | 20 years |
DSCR = CFADS ÷ debt service. Year 2 in the downside: 8.06 ÷ 7.21 = 1.118x, below 1.20x, so the 0.85 left after debt service goes to the blocked account.
Trapped balance carried forward: previous balance × 1.02 + this year's free cash. In Excel: =IF(DSCR<Lockup, Trapped_prev*(1+Deposit)+(CFADS-DS), 0), and distribution =IF(DSCR>=Lockup, (CFADS-DS)+Trapped_prev*(1+Deposit), 0).
The test fails from year 2 to year 9 (year 9 at 1.199x, a miss by a rounding error). It passes in year 10, when 9.91 of trapped cash is released alongside that year's 1.52.
| Year | CFADS | DSCR | Distribution, no lock-up | Distribution, lock-up | Trapped balance |
|---|---|---|---|---|---|
| 1 | 9.50 | 1.318x | 2.29 | 2.29 | 0.00 |
| 2 | 8.06 | 1.118x | 0.85 | 0.00 | 0.85 |
| 5 | 8.30 | 1.152x | 1.10 | 0.00 | 4.01 |
| 8 | 8.56 | 1.187x | 1.35 | 0.00 | 8.12 |
| 9 | 8.64 | 1.199x | 1.43 | 0.00 | 9.71 |
| 10 | 8.73 | 1.211x | 1.52 | 11.43 | 0.00 |
| Downside case | Equity IRR | Total distributions | Multiple | Cost vs no lock-up |
|---|---|---|---|---|
| No lock-up | 6.08% | 69.1 | 2.30x | nil |
| Lock-up, cash held until the test is passed | 5.83% | 69.9 | 2.33x | 26 bp |
| Lock-up, swept to prepay debt after two failed tests | 6.07% | 70.4 | 2.35x | 1 bp |
The lock-up raises total distributions, because 9.12 of trapped cash earns 0.78 of interest, and still lowers the IRR, because that cash arrives up to eight years late at 2 per cent when the equity is priced at more. The downside itself costs 447 basis points against the base case; the lock-up is a small addition to a large problem, which is the right way to present it to a committee.
With a sweep, the 1.80 trapped by the end of year 3 prepays debt, and so does year 4's cash. Debt service is re-amortised over the remaining tenor and falls in two steps, from 7.21 to 6.99 and then to 6.84. Two things follow. The swept cash earns the 6 per cent the debt costs rather than 2 per cent on deposit. And because debt service is lower, the DSCR clears 1.20x from year 5, at 1.215x, instead of year 10. The 3.0 swept comes back to equity as lower debt service over the rest of the tenor. It looks harsher in a term sheet and is almost free in the model.
| CFADS below base | Lowest DSCR | Years locked | IRR, no lock-up | IRR, held | IRR, swept |
|---|---|---|---|---|---|
| 8% | 1.22x | 0 | 8.36% | 8.36% | 8.36% |
| 12% | 1.17x | 3 | 7.24% | 7.15% | 7.18% |
| 16% | 1.12x | 8 | 6.08% | 5.83% | 6.07% |
| 20% | 1.07x | 12 | 4.91% | 4.66% | 4.94% |
| Lock-up threshold | Years locked | Cost if held | Cost if swept |
|---|---|---|---|
| 1.10x | 0 | 0 bp | 0 bp |
| 1.15x | 3 | 5 bp | 1 bp |
| 1.20x | 8 | 26 bp | 1 bp |
| 1.25x | 12 | 51 bp | 1 bp |
The cost of a held lock-up is driven by duration, and duration is driven by how far the threshold sits above the downside DSCR and how slowly CFADS grows back. Moving the threshold from 1.20x to 1.25x roughly doubles it here. At a 20 per cent downside the sweep even beats no lock-up, because prepaying 6 per cent debt is a better use of cash than the equity's own 4.91% return.
The usual error is modelling the blocked account at zero interest and with no release, which turns a timing cost into a permanent loss and makes the downside look far worse than it is. The opposite error is subtler: releasing trapped cash on the first pass when the loan agreement requires two consecutive passes, or a projected test as well as a historic one. With CFADS creeping up at 1 per cent a year, one extra condition adds a year of lock-up. Read the release mechanics as carefully as the threshold. The cash flow that feeds the test is built in how to calculate CFADS.
For the investment memo: report, in the downside case, the first year the lock-up bites, the year it releases, the peak trapped balance (9.71 here) and the IRR cost. Four numbers tell a committee more than a switch.
A lock-up is a deferral priced at the difference between what equity wants and what the blocked account pays. It costs little for a short dip and tens of basis points for a long one, and a sweep into expensive debt can cost almost nothing. The book's own debt sculpting case, with its coverage floor and leverage solved rather than assumed, is in the free workbook for this case.
A condition in the loan agreement that blocks payments to shareholders while a coverage ratio, usually the historic or projected DSCR, is below a threshold set above the default level. Illustrative levels are 1.20x for the lock-up against 1.05x for default. The cash stays in the project, often in a blocked account, and is released when the test is passed again, or swept to repay debt if it is failed for too long.
Not directly. The trapped cash is paid out later with the interest it earned, so in the illustrative case equity receives 69.9 in total under the lock-up against 69.1 without it. What falls is the IRR, from 6.08% to 5.83%, because 9.12 of distributions arrive up to eight years late while earning 2 per cent on deposit against a 6 per cent cost of debt.
Often not. Swept cash prepays debt that costs 6 per cent, so debt service falls and the coverage ratio recovers sooner. In the illustrative case a sweep after two failed tests moves 3.0 into prepayment, the DSCR clears 1.20x from year 5 rather than year 10, and the equity IRR is 6.07% against 5.83% for a long lock-up. The sweep looks harsher in the term sheet than in the model.
This article is one calculation from The Infrastructure Investment Analyst. 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: infrastructure, data centres and energy → · 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.