Northgate’s net operating income would carry 538.1 million of senior debt at a cover ratio of 1.45 times. The loan signed is 427.3 million, because a second test caps it at 65 per cent of cost. The cap refuses 110.7 million, 20.6 per cent of what the cover test would lend, and the 1.45 in the term sheet plays no part in the size of the loan until the cap is raised past 80.08 per cent. What the 110.7 buys is the part worth pricing.
Two tests, one loan
Northgate is a 96 MW data centre that costs 614.4 million to build and is let for fifteen years on a triple-net lease at 56.00 dollars a kilowatt a month, indexed at 2.5 per cent. Year-one net operating income is 61.61 million. The loan costs 6.10 per cent all in, a swap of 4.10 and a margin of 2.00, and pays a level instalment on a twenty-five year profile over a fifteen-year term, so a balloon falls due in the month the only lease expires.
The lender sizes it twice. The cover test divides year-one income by 1.45, which permits 42.49 million a year of debt service, and multiplies by the annuity factor on the profile, 12.6628: 538.1 million. The cost test takes 65 per cent of the funding requirement. That requirement is itself a function of the loan, since interest during construction, the arrangement fee and the reserve all scale with it, and the model solves the loop on its own sheet: 657.4 million, of which 65 per cent is 427.3. The loan is the lower of the two, so the cost test binds, and by a wide margin.
The level instalment on 427.3 million is 33.75 million, so year-one cover is 1.826 times, not the 1.45 the covenant describes. Cover is lowest in year one, because the rent indexes and the instalment does not, and it only rises from there.
Where the cover test starts to bind
Raise the cap and the loan follows it, while the cover test stays exactly where it was. The table runs the cap up with nothing else changed.
Loan to cost
Loan
Funding requirement
Equity
Year-one cover
Balloon
Equity IRR
65 per cent
427.3
657.4
230.1
1.826
247.2
16.60%
70 per cent
463.3
661.9
198.6
1.684
268.0
17.75%
75 per cent
500.1
666.8
166.7
1.560
289.3
19.20%
80 per cent
537.4
671.8
134.4
1.452
310.9
21.16%
80.08 and above
538.1
671.9
133.8
1.450
311.3
21.19%
Northgate, million dollars. Only the loan-to-cost input moves; the funding loop re-solves each time. From 80.08 per cent the cover test binds and the loan stops growing.
Two things in that table are easy to miss. The funding requirement rises with the loan, by 14.4 million between the first row and the last, because a bigger loan carries more interest during construction, a bigger fee and a bigger reserve; so the crossover is not 538.1 divided by 657.4 but 538.1 divided by 671.9, which is 80.08 per cent. And the equity return climbs steadily all the way up, from 16.60 to 21.19 per cent, while the cheque falls from 230.1 million to 133.8. On the sponsor’s own spreadsheet every step up the cap looks like a gain.
On this deal the 1.45 times in the term sheet is not the sizing constraint anywhere between 65 and 80 per cent of cost. The number that decides how much the lender advances is the cost cap, and a sponsor negotiating the cover ratio is negotiating a term that does not bind.
What the refused 110.7 million buys
The cover test is a forecast: it believes year-one income and lends against it. The cost test distrusts the forecast and lends against what was spent. The difference between the two loans is the price of that distrust, and it can be read off the model in four places.
Measure
Loan at 65% of cost
Loan at the cover test
Senior debt (million)
427.3
538.1
Rent at which cover reaches 1.00, debt held ($/kW-month)
31.81
39.40
Fall in rent that takes cover to 1.00
43.2%
29.6%
Loan life cover ratio at close
1.619
1.286
Balloon at the end of the term (million)
247.2
311.3
Rent at which a 65% lender refinances the balloon ($/kW-month)
27.58
33.78
as a share of the expiring rent of 79.13
34.9%
42.7%
Debt held as signed, as the stress switch holds it. Discounted contracted income is 691.7 million in both columns; the refinancing rent capitalises the year-sixteen income at the 7.25 per cent exit cap and adds 4.20 million of non-recoverable cost.
The second and third rows are the operating cushion. With the loan at 427.3 the rent can fall 43.2 per cent before income only just meets the instalment; at 538.1 it can fall 29.6 per cent. The cap buys 13.6 points of rent. On a signed lease that cushion is not for a weaker market; it is for a tenant that pays less than it signed for, or stops paying, which is the one event a single-tenant building cannot diversify away.
The loan life cover ratio says the same thing over the whole term. Fifteen years of contracted income discounted at the loan rate is 691.7 million. Against 427.3 that is 1.619 times; against the loan the cover test would have allowed, 1.286. The model does not print a covenant floor, because none follows from the arithmetic, but any floor between the two is cleared by one loan and failed by the other.
The last three rows matter most on a single-tenant building, because they are about the day the lease ends. A twenty-five year profile over a fifteen-year term leaves a balance outstanding at maturity, and a larger loan leaves a larger one: 311.3 million instead of 247.2. For a lender advancing 65 per cent on the same exit cap to take that balloon out, the building has to be worth 478.9 million rather than 380.3, and a new tenant has to pay 33.78 dollars a kilowatt a month rather than 27.58. The expiring rent is 79.13. The cap lets the rent fall 65.1 per cent at renewal and still refinance; the cover-sized loan, 57.3 per cent.
What it means in practice
Read the sizing section of a project term sheet from the bottom up. The cover ratio is the covenant everyone discusses, because it is the one that tests the forecast, and on a highly contracted asset with a single strong tenant it is often not the constraint at all. When the cost test binds by 20 per cent, the useful questions are about the cap, the cost it is measured against, and the balloon it leaves:
Compute both tests and state which binds, and by how much. A sheet that shows only the loan hides whether the cover ratio is a live term.
Find the cap at which the other test takes over. Here it is 80.08 per cent, and every point below it is a point of leverage the cover test would have allowed.
Price each step up the cap in rent cushion and in refinancing rent, not only in equity return. The return rises smoothly with the cap; the cushion falls just as smoothly, and only one of the two is on the sponsor’s summary page.
Test the balloon against a re-letting rent, not the expiring one. A loan that matures with the lease is refinanced on whatever the next tenant pays.
None of this argues that 65 per cent is right. It argues that on Northgate the cost cap is the covenant that sizes the loan, that it is worth 4.60 points of equity return to the sponsor, from 16.60 to 21.19 per cent, and that the lender is paid for it in 13.6 points of rent cushion and 7.8 points of refinancing headroom. That is a trade to be made deliberately, and it is only visible when both tests are on the page.
Reproducing it in the workbook
Everything above is in Northgate_Model.xlsx. On Debt sizing, B9 is the loan the cover test supports (538.1), B13 the loan the cost test supports (427.3) and B17 names the one that binds; B20 is the level instalment, B21 the balloon, B26 the loan life cover ratio at close and B27 the same ratio had the loan been sized to the cover test (1.286). B28 is the rent that refinances the balloon (27.58).
For the first table, change the maximum loan to cost on Assumptions, B21, and read the loan, the funding requirement on Sources and uses, minimum cover in Returns B5 and the equity return in B22. The Solve sheet re-runs the funding loop and should still read CONVERGED in B26. The book’s figures on the Checks sheet stop matching, as they should, while the identities that hold for any inputs stay at OK. Row 28 on Debt sizing takes the refinancing lender’s advance from the same B21, so to keep that lender at 65 per cent while the construction loan moves, apply its formula by hand: the balloon over 0.65, times the 7.25 per cent exit cap, plus 4.20 million of year-sixteen cost, divided by 1.152.
For the cushion, set the stress switch on Assumptions B34 to YES, put the loan in B35 and cut the rent in B9: cover on Cash flow C13 reaches 1.00 at 31.81 with the signed 427.3 and at 39.40 with 538.1. With the switch at NO the loan re-sizes to every rent and the cover ratio never falls below its target, which is why a sensitivity cannot show this and a stress case can.
The workbook behind this article
Every figure above is a live formula in the free companion files for
The Project Finance Handbook: Northgate and Aurora corrected, the blank set, thirteen working documents, forty questions and three cases. Each workbook ends with a Checks sheet setting the
printed figure beside the computed one. No account and no email address.
At what wealth does a family office pay?A five-person office costs the same $2,442,000 at $100m as at $2bn. Against a tiered multi-family schedule the two curves cross exactly once.
Does an undrawn revolver count as liquidity?67.4567 per cent of the headroom, and it leaves at a fall in annual revenue of 8.4127 per cent — not the 13.3333 per cent the board deck implies.
Does supply chain finance count as debt?66,049,315 released, and leverage of 2.7622× or 3.1362× depending on which balance the test is asked of — the second is a breach at today's EBITDA.
Does the fulcrum move with enterprise value?A worked capital structure at three enterprise values. The break migrates from the first lien to the subordinated notes without a single document changing.
How many companies does a reserve pool defend?Holding pro rata to the Series D costs $14.76m per company. A $40m reserve defends 2.71 of forty — and 61 per cent of the cost is the last round.
How much room a 1.15x downside actually leaves79 basis points of margin. The leverage covenant fails first, five basis points before the coverage covenant the chapter reports on, and 4.6 per cent of EBITDA is the whole cushion.
How the GP catch-up is actually solvedMost write-ups state the catch-up formula then hardcode the answer. Set it as a solve and it moves on its own when the carry rate does.
How do you set a minimum cash balance?Three components, 124,596,281 in total — and the double count that turns a margin of 27,403,719 into an apparent shortfall of -3,988,281.
Is a 2/10 net 60 discount worth taking?An annualised 14.8980 per cent against a revolver at 4.50 per cent — worth 5,471,890 a year, and 0.3132 turns of reported leverage.
Should corporate debt be fixed or floating?78.1818 per cent of gross debt fixed, and a net floating exposure after cash of -608,000.00 — the proportion, not the instrument.
Six ways to state the same fund's return7.12 per cent, 18.04 per cent, or 1.323 times the public market — one fund, one set of flows. The eleven-point gap is the valuation, and it is not cash.
The IRR at which a closed-ended fund merely tiesA drawdown fund advertising 13.25 per cent ties an evergreen netting 8.64. On committed rather than called capital, 78.7 per cent of the edge vanishes.
The month a remediation trigger has to be armed byMonth 8, on a one-year window and a 14-week hiring lead time. The date is 12W − L/4.33 and holds for any portfolio size — the book only changes the euros.
Two routes to the same NAVFive findings in one quarter-end close, and not one of them was a modelling error. Every one was a control that did not exist.
What a CBAM certificate buffer costs to holdCarry and protection are both linear in the buffer, so the break-even probability is identical at 5 per cent and at 30 — and 2027 costs exactly twice 2026.
What a clawback actually collateralises24 million recovers 13.20 net of tax and 7.20 is escrowed. The rule is e ≥ 1 − t, and every euro won on the tax clause lands uncollateralised.
What a co-investment programme actually saves37 basis points, and a 22.8 per cent total-loss rate erases it. At the usual per-deal cap a single failure is 37 per cent of the annual cohort.
What a diligence exercise actually recoversEighteen basis points is only the part that reaches the yield. Three of the four instruments never do, and the exercise is worth thirty-one.
What a follow-on reserve is actually worth0.019 turns under plain dilution, 0.285 under pay-to-play. The optimum is a property of the shareholders’ agreement, not of the portfolio.
What a lease mark-to-market is actually worth5,788,000 becomes 807,365 once the over-market stream, the expiry dates and the cost of re-letting are priced — and nothing at all below a 43.7 per cent renewal rate.
What a minimum payment actually does£81.53 leaves the account and £20.46 reaches the debt. On minimums alone the set clears in 258 months and repays 1.78 times what was borrowed.
What a month of delay costs in an enforcement463,000 euros of present value a month. No single error about enforcement, and no pair of errors, moves the indifference point from 58.7 cents to the 72 on the table.
What a pro rata cheque actually costsDefending costs 1.25 times the initial cheque across two rounds, not more in each — and below a 521 exit the follow-on dollars come back worth less than they cost.
What a subscription line does to the IRR20 per cent becomes 38.41 on the same asset while the multiple falls. Two dollars of interest buys 18.4 points, and the longer the line runs the cheaper each point gets.
What one point of fees actually costsOne point on the annual charge turns 738,846 into 567,961 over forty years — 23 per cent of the pot, and 2.43 of wealth lost per 1 of fee.
When the promote stops clearingCarried interest is not proportional to performance near the hurdle. It is a step function, and $2m of headroom is what stands between a promote and none.
Why a secondaries bid-ask gap has no zoneThe seller’s floor and the buyer’s ceiling are one formula at two rates. At equal required returns the zone is exactly zero — here it is 11.02 points apart.
Why refining a due diligence rubric makes it weakerThree of twelve managers carry a terminal flaw and the average approves all three. Splitting eight domains into sixteen makes it easier to hide, not harder.
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.