Cash flow available for debt service starts at EBITDA and ends after every cash call that ranks ahead of the lenders.
CFADS, cash flow available for debt service, is EBITDA less cash taxes, the increase in working capital, maintenance capex and transfers into the maintenance reserve, plus any release from that reserve. On an illustrative toll road year, EBITDA of 33.6 becomes CFADS of 28.0, a DSCR of 1.30 times against debt service of 21.5. Dividing EBITDA by debt service instead would report 1.56 and overstate coverage by a fifth.
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 →
The asset is an illustrative toll road in a normal operating year, financed with a senior term loan. Figures are in millions. Routine maintenance is spent every year; a resurfacing of 12.0 falls due every few years and is funded through a maintenance reserve account that receives 1.5 a year.
| Input | Value |
|---|---|
| Toll revenue | 48.0 |
| Operating costs | 14.4 |
| Tax depreciation | 12.0 |
| Interest on senior debt | 10.0 |
| Scheduled principal | 11.5 |
| Tax rate | 25% |
| Increase in working capital | 0.4 |
| Routine maintenance capex | 0.8 |
| Transfer to maintenance reserve | 1.5 |
EBITDA = revenue - operating costs = 48.0 - 14.4 = 33.6
Cash tax = (EBITDA - depreciation - interest) x tax rate = (33.6 - 12.0 - 10.0) x 25% = 11.6 x 25% = 2.9
CFADS = EBITDA - cash tax - increase in working capital - maintenance capex - reserve transfer + reserve release
= 33.6 - 2.9 - 0.4 - 0.8 - 1.5 + 0 = 28.0
DSCR = CFADS / (interest + principal) = 28.0 / 21.5 = 1.30
In Excel, with the lines in a column: =B4-B5-B6-B7-B8+B9 for CFADS and =B10/(B11+B12) for the DSCR.
| Line | Amount |
|---|---|
| Revenue | 48.0 |
| Less operating costs | -14.4 |
| EBITDA | 33.6 |
| Less cash tax | -2.9 |
| Less increase in working capital | -0.4 |
| Less routine maintenance capex | -0.8 |
| Less transfer to maintenance reserve | -1.5 |
| CFADS | 28.0 |
| Debt service: interest 10.0 and principal 11.5 | 21.5 |
| DSCR | 1.30x |
Three conventions matter. Tax is the cash tax actually paid in the period, computed after interest, so in a sculpted model it creates a circularity between debt size and tax. Interest itself is not deducted, because it is part of the debt service CFADS is meant to cover. And the definition that counts is the one in the credit agreement, which is where items such as reserve transfers, insurance proceeds and equity cures are expressly put in or left out.
The maintenance reserve is what keeps the resurfacing year from becoming the binding year. Compare the same road with and without it.
| Case | CFADS | DSCR |
|---|---|---|
| Normal year, reserve funded at 1.5 | 28.0 | 1.30x |
| Resurfacing year, 12.0 spent and 12.0 released from reserve | 28.0 | 1.30x |
| Normal year, no reserve | 29.5 | 1.37x |
| Resurfacing year, no reserve | 17.5 | 0.81x |
Without the reserve, normal years look better at 1.37 times, and the resurfacing year falls to 0.81, a payment default unless the debt service reserve covers it. Debt on a fixed repayment profile is sized on the tightest year, so a lumpy capex profile shrinks the loan across every year, not just one; even sculpted debt loses the capacity of the resurfacing year. Smoothing the cost through the reserve transfer is what lets the lender size on 28.0 throughout.
Releases from the maintenance reserve belong in CFADS because they match a deduction for the capex they fund. Releases from the debt service reserve do not: they are the lenders' own security being used to pay them.
CFADS falls faster than revenue in percentage terms, because operating costs are fixed, and more slowly in absolute terms, because tax absorbs a quarter of the fall in profit.
| Measure | Base | Stress |
|---|---|---|
| Revenue | 48.0 | 43.2 |
| EBITDA | 33.6 | 28.8 |
| Cash tax | 2.9 | 1.70 |
| CFADS | 28.0 | 24.4 |
| EBITDA / debt service | 1.56x | 1.34x |
| DSCR on CFADS | 1.30x | 1.13x |
A revenue fall of 4.8 cuts CFADS by 3.6, or 12.9 per cent. With an illustrative lock-up at 1.20 times, CFADS needs to stay above 25.8, so revenue can fall only 2.93, or 6.1 per cent, before distributions stop. On EBITDA the same stress still shows a comfortable 1.34.
The same logic runs backwards into sizing. Lenders size sculpted debt so that CFADS divided by debt service equals the target ratio in every period, so every item deducted between EBITDA and CFADS reduces the loan directly. An adviser who forgets a recurring cash item in the base case is not making a presentational error: the loan is larger than the cash flow can carry, and the first test date reveals it. The diligence question for each line is simply whether the cash leaves the project before the lenders are paid.
The most expensive error is using EBITDA as a proxy for CFADS, typically in an early screening model. It overstates coverage here by 0.26 times and hides a lock-up under modest stress. The second is adding back the debt service reserve release in a stressed year: a 3.0 release would lift the stressed DSCR from 1.13 to 1.27 and report a breach as compliance. The third is deducting both the reserve transfers and the resurfacing in the year it is spent without adding back the release that funds it, which counts the resurfacing twice.
Start at EBITDA, deduct every cash item that ranks ahead of the lenders in the waterfall, and stop before interest and principal: 28.0 of CFADS and 1.30 times here. How the reserve shapes the debt quantum on a real concession is worked in the free workbooks for this book, and the sizing that follows from CFADS is shown in sculpted debt against an annuity.
EBITDA is an operating profit measure; CFADS is cash. CFADS deducts cash taxes, the change in working capital, maintenance capex and transfers to the maintenance reserve, and adds reserve releases. On the illustrative toll road the deductions total 5.6, so CFADS is 83.3 per cent of EBITDA, and the DSCR is 1.30 rather than 1.56.
Normally not for the ratio test. A DSRA release is the lenders' own collateral being used to pay them, so most credit agreements exclude it when calculating the DSCR. Including a 3.0 release in the stressed year would lift the illustrative DSCR from 1.13 to 1.27 and hide a lock-up breach.
Cash tax depends on taxable profit, which deducts interest, and interest depends on the debt that CFADS supports. In a sculpted model this creates a circular reference, usually broken with a copy-paste macro or an iterative calculation. In the illustrative year tax is 2.9 on taxable profit of 11.6 after 10.0 of interest.
Debt sculpting and the DSCR are worked on the Calder Toll Road in chapter 4 of 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 453 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.