Merchant Tail Debt Sizing: How Much Uncontracted Revenue Does the Lender Actually Credit?
A project has a 15-year PPA and a 25-year operating life.
Does the lender give any credit to years 16–25?
There is no single market rule.
One filed financing gives up to five post-PPA years credit at a 2.00x P50 DSCR.
Another separates contracted cash at 1.25x P50 / 1.00x P99 from merchant cash at 1.80x P50 / 1.40x P99 and caps the amount of debt supported by merchant cash at 25%.
Another uses 1.35x on contracted P50 cash and 1.85x on non-contracted P50 cash, with separate P99 divisors.
Another financing includes at least ten years of merchant energy cash flow in the base forecast while still capping merchant-supported debt at 25% of total term commitments.
Those documents are more useful than a sentence saying “lenders take a 30% merchant haircut.”
They show the real principle:
merchant revenue can enter debt sizing, but the lender decides separately how much of it to believe, for how long, under which production case, at what coverage ratio, and how much of the total loan may depend on it.
Merchant tail and debt tail are different things
These terms are easy to confuse.
A merchant tail is the period after contracted or hedged revenue ends.
A debt tail is the period between scheduled debt maturity and the end of the revenue contract or other bankable cash-flow period.
Example:
PPA expiry Year 15
Debt maturity Year 13
Asset life Year 30
The project has:
2-year contracted debt tail
15-year merchant asset tail
Now move debt maturity to Year 18:
PPA expiry Year 15
Debt maturity Year 18
The lender is now relying on three merchant years to repay part of the debt.
That is a different credit structure.
The important question is not whether the asset has merchant value.
Equity clearly cares about years 16–30.
The lender asks a narrower question:
How much debt should be outstanding when contracted revenue ends, and how much uncontracted cash am I willing to rely on after that date?
There is no universal merchant haircut
The old version of this article described merchant treatment as a stack of approximate market haircuts.
That is too crude.
A lender can reduce merchant-supported debt in several different ways.
1. Use a higher DSCR
The 2026 Big 5 financing is explicit.
Contracted fixed-price cash is tested at:
P50 1.25x
P99 1.00x
Merchant cash is tested at:
P50 1.80x
P99 1.40x
The same dollar of lender CFADS therefore supports materially less debt service once it is merchant.
2. Apply a harsher production case at the same time
The facility does not simply replace P50 with P99.
It calculates both and takes the lower debt result.
That matters because price risk and production risk can hurt simultaneously.
3. Limit how many merchant years enter the sizing period
The DESRI filed agreement allows a defined post-PPA merchant period of up to five years, subject to its term limit.
Cash after that period may be very valuable to equity and irrelevant to initial debt sizing.
4. Cap merchant-supported debt as a percentage of the loan
A 2026 filed senior facility limits debt supported by merchant electricity cash flows to 25% of the term loans.
Another project financing with a long hypothetical amortization period also caps the amount of term debt supported by merchant cash at 25% of total term commitments.
That is a portfolio / structural limit, not a price haircut.
5. Use a conservative merchant price case or severe repayment case
The Big 5 refinancing also requires full repayment under a severe downside case that combines P99 production with a low historical power-price assumption and escalation.
This tests the whole repayment path rather than merely changing one annual DSCR.
6. Give merchant revenue zero credit in a particular covenant or test
A facility can distinguish total DSCR from a Forward Contracted DSCR that only recognizes contracted revenue.
That does not mean the asset has no merchant value.
It means one particular lender test refuses to rely on it.
These mechanisms can be combined.
That is why “the lender haircut is 30%” is usually the wrong abstraction.
One filed agreement can contain several merchant protections at once
The 2026 Big 5 senior facility is worth reducing to one page:
CONTRACTED CASH
P50 DSCR 1.25x
P99 DSCR 1.00x
MERCHANT CASH
P50 DSCR 1.80x
P99 DSCR 1.40x
ADDITIONAL TESTS
Severe-downside repayment required
Merchant-supported debt ≤ 25% of term loans
20-year sculpted amortization
Absolute / structural limits also apply
The lender did not choose one merchant haircut.
It used at least four protections:
- higher coverage;
- severe production case;
- severe price / repayment case;
- merchant-support cap.
That is the architecture a model should reproduce.
Separate the contracted and merchant debt-service streams
Do not use one blended DSCR across the full tenor.
Suppose:
Years 1–15 Contracted CFADS
Years 16–18 Merchant CFADS
Then calculate:
Contracted Debt Service(t)
=
Contracted CFADS(t)
/ Contracted Target DSCR(t)
Merchant Debt Service(t)
=
Merchant Lender CFADS(t)
/ Merchant Target DSCR(t)
The total opening debt is the present value of both blocks, subject to all other sizing constraints.
This lets the model report the merchant contribution directly:
Debt with contracted period only A
Debt with contracted + merchant tail B
Merchant Tail Contribution
=
B - A
That number is much more useful than debating whether the merchant forecast is “conservative.”
A worked tail using one explicit set of assumptions
Assumptions only:
PPA term 15 years
Debt rate 6.5%
Contracted CFADS $14.52m/year
Contracted DSCR 1.30x
Merchant lender CFADS $9.14m/year
Merchant DSCR 2.00x
Merchant years credited 3
The 1.30x / 2.00x split is used here as an illustrative structure similar to a filed DESRI agreement, not as a market recommendation.
Contracted annual debt service:
14.52 / 1.30
=
$11.17m
Merchant annual debt service:
9.14 / 2.00
=
$4.57m
Discounting the first 15 years of contracted debt service at 6.5% produces approximately:
Contracted-only debt
≈ $105.0m
Add three merchant years at the stricter 2.00x ratio:
Contracted + merchant-tail debt
≈ $109.7m
Merchant tail contribution:
$109.7m - $105.0m
≈ $4.7m
The project has three additional operating years in the debt period.
They add only about $4.7 million of initial debt under these assumptions.
That is the decision number.
Why the sponsor's merchant forecast is not the lender's CFADS
The sponsor might forecast merchant revenue using:
- a consultant or internal price curve;
- P50 production;
- expected basis;
- expected capture rate;
- expected curtailment;
- expected availability.
The lender can change any of those assumptions before the merchant DSCR is even applied.
So merchant sizing should have a bridge:
Sponsor merchant revenue
- price / capture adjustment
- basis adjustment
- curtailment adjustment
- production-case adjustment
- availability / operating adjustment
= Lender Merchant Revenue
- lender operating costs
- cash tax / other CFADS items
= Lender Merchant CFADS
÷ Merchant DSCR
= Merchant-supported debt service
Do not hide all of that inside a single “merchant haircut” cell.
The bridge tells you which assumption actually destroyed the debt capacity.
Merchant production and contracted production can be treated differently
It is tempting to assume the same P50 / P99 methodology should apply on both sides of PPA expiry.
Filed agreements show otherwise.
The DESRI agreement uses P50 and P99 during the PPA period, then applies a high P50 DSCR for a limited post-PPA period.
The Big 5 structure uses P50 and P99 tests for merchant cash as well, but at much higher ratios.
A 2026 financing agreement for another project explicitly separates contracted and non-contracted cash flows under both P50 and P99 cases, applying different divisors to each component.
So the correct model is not:
Merchant = P90 × one merchant DSCR
It is:
Merchant treatment = what this credit agreement actually says
The same applies to the contracted period.
Merchant years can be valuable and still fail the debt cap
Suppose the merchant tail supports $30 million of debt on coverage.
The credit agreement caps merchant-supported debt at 25% of a $100 million facility.
Allowed merchant-supported debt:
25% × $100m
=
$25m
The remaining $5 million of coverage capacity is irrelevant to final sizing.
That is why a model needs both:
Merchant coverage capacity
Merchant contractual cap
and an explicit minimum between them.
The same issue can arise with an absolute commitment cap or a leverage cap.
Merchant cash can be strong and still not move final debt because another constraint binds.
A hedge does not magically turn merchant cash into PPA cash
A hedge can remove or reduce one risk while leaving others.
A hub-settled swap can reduce outright-price risk and leave basis.
A fixed-volume hedge can leave shape and volume mismatch.
A floor can protect downside while preserving upside but cost a premium.
A proxy-generation structure can change which volume risk sits with which party.
The debt-sizing question is therefore:
Which risks are actually transferred by the hedge?
Which merchant sizing treatment does the credit agreement apply afterward?
Do not automatically switch the DSCR from “merchant” to “contracted” because a hedge exists.
The lender may still use a distinct hedge / non-contracted category.
Longer tenor is only valuable to the extent the tail adds debt
A 20-year asset life does not justify 20 years of debt.
The incremental merchant years have to earn their place.
For every proposed extension, calculate:
Debt with shorter amortization
Debt with longer amortization
Incremental debt proceeds
Incremental interest / hedge / structuring cost
Incremental refinancing or covenant risk
If three merchant years add $4.7 million of proceeds and cause the entire facility to reprice materially wider, the extension can destroy value.
If the incremental proceeds are strategically important and pricing barely moves, the same three years can be worthwhile.
There is no universal answer.
The alternative is to leave the merchant tail to refinancing
The sponsor can decide not to finance far into the uncontracted period.
A mini-perm or shorter legal maturity can leave a balance to be refinanced before or near PPA expiry.
That does not remove merchant risk.
It changes its form:
Long initial tenor
→ more initial leverage
→ more direct merchant repayment exposure
Shorter / mini-perm structure
→ less merchant credit today
→ refinancing risk later
The right comparison is initial debt proceeds versus expected refinance exposure and the effect of cash sweeps / margin step-ups on equity distributions.
That belongs with the separate mini-perm article, but merchant-tail sizing is what creates the choice.
Equity and lenders value the tail for different reasons
Equity owns the residual asset value.
The lender cares about repayment before maturity.
A ten-year merchant period after the loan is fully repaid can be a large component of sponsor NPV and zero component of debt capacity.
That is not a disagreement about the power forecast.
It is a disagreement about which cash flows belong to which capital provider.
This is why arguing endlessly over a Year-25 merchant price with a lender whose loan matures in Year 15 is usually unproductive.
The lender may care about that year only through residual-value or refinance considerations.
Equity cares directly.
What the debt-sizing page should publish
At minimum:
Contracted-period debt capacity
Merchant-period coverage capacity
Merchant contractual cap
Severe-downside capacity
Debt without merchant tail
Debt with merchant tail
Incremental debt from merchant tail
% of final debt supported by merchant cash
Binding merchant constraint
Merchant years credited
Then the team can answer the actual commercial question:
What did the lender give us for accepting merchant exposure inside the debt term?
If the answer is $4.7 million, say $4.7 million.
Do not describe it as “three years of extra tenor” and leave the economic value implicit.
A useful comparison of filed structures
| Filed financing | Contracted treatment | Merchant / non-contracted treatment | Additional constraint |
|---|---|---|---|
| Big 5 refinancing | P50 1.25x / P99 1.00x | P50 1.80x / P99 1.40x | Severe downside + merchant-supported debt cap |
| DESRI financing | P50 1.30x / P99 1.00x during PPA | P50 2.00x for limited post-PPA period | Term / period limit |
| 2026 credit agreement | Contracted cash has separate P50/P99 divisors | Non-contracted cash has separate higher divisors | Maturity-balance constraint |
| Eiffel financing | P50 1.25x / P99 1.00x in long notional structure | Merchant energy included under defined assumptions | Merchant-supported debt ≤25% |
These are examples of transaction design.
They are not a market survey.
That distinction is exactly the point of the article.
The series ends with one discipline
Part 1 showed that debt is the minimum of a stack of constraints.
Part 2 showed how an approved debt-service stream becomes an amortization schedule.
The merchant tail is where both ideas meet.
Uncontracted cash may be included in the sculpt, but it is often given:
- a different coverage ratio;
- a different production case;
- a different price basis;
- a shorter eligible period;
- a separate cap;
- a severe-downside test.
The model should not blend those differences away.
It should publish exactly how many dollars of initial debt the merchant tail adds, and exactly which lender assumption limits that number.
That is the number a sponsor can negotiate.
Sources: SEC — 2026 Big 5 Refinancing Credit Agreement · SEC — DESRI Financing Agreement · SEC — 2026 Contracted / Non-Contracted Debt Sizing Parameters · SEC — Eiffel Financing Agreement