Cash flow versus collateral: how structured credit is actually underwritten

A structured credit facility is decided by two questions that are routinely confused. The first is whether the business can pay. The second is what happens if it does not. These are not versions of one another. They are answered with different evidence, they produce different documents, and a proposal built to answer one will not survive a committee asking the other.

The confusion is expensive because it stays invisible until late. A borrower with substantial assets assumes the assets are the argument. A borrower with strong EBITDA assumes the numbers speak for themselves. Both are declined by lenders who rarely articulate which question the proposal failed, and both conclude that the market is closed to them, when what actually happened is that they answered the wrong one.

The two questions are not interchangeable

Cash-flow underwriting asks whether the business generates enough, reliably enough, and early enough to service the facility on its schedule. Collateral underwriting asks what the lender recovers if it does not, how long recovery takes, and what it costs to get there.

A facility can pass the first and fail the second. A profitable company whose premises are leased and whose only other asset is a receivable book of uncertain enforceability generates cash and secures very little. A facility can equally pass the second and fail the first: an asset-rich borrower with no serviceable cash flow is not offering a loan, it is offering a recovery process with interest attached.

The instinct in the Indian mid-market is to lead with collateral, because that is the language bank lending has used for thirty years. In structured credit it is usually the weaker of the two arguments, and understanding why is the whole of this piece.

What collateral underwriting actually tests

The headline number is cover: the ratio of security value to exposure. It is also the least informative figure in the analysis, because it is assembled from three judgements that are each far more contestable than a single ratio suggests.

• Valuation basis. Book value, replacement cost and realisable value in a forced sale are three different numbers, and only the third is the one being lent against. The gap between them widens precisely when it is tested, because a distressed asset reaches the market at the moment the market for it is worst.
• Enforceability. Whether the charge is perfected, whether third-party consents are required, whether the asset is capable of transfer at all. A charge registered over an asset the borrower cannot lawfully transfer is not security; it is paperwork.
• Time. Recovery delayed is recovery reduced. The statutory routes in India are real, and secured creditors have enforcement powers that operate outside the courts for certain classes of asset. But practice routinely runs longer than the statute contemplates, and a lender underwriting to the statutory timetable is underwriting the optimistic case.

Put those together and a facility at two times cover on book value may be at a little over one times on realisable value, and below par once eighteen months of carry, legal cost and enforcement expense are applied. The ratio did not move. What it was measuring did.

What cash-flow underwriting actually tests

The equivalent headline is coverage: debt service cover, interest cover, some ratio of cash generated to cash owed in the period. It carries the same weakness. The ratio is only ever as good as the definition of its numerator, and the numerator is where the argument is won or lost.

• Which cash. EBITDA is not cash. The distance between them is working capital movement, maintenance capital expenditure and tax, and in a growing mid-market business that distance frequently consumes the entire margin the lender believed it had.
• Whose cash. Money arriving in the business is not money available to this facility. It may be committed to trade creditors, to a working capital lender holding a prior charge over current assets, or to obligations that never present as debt on the balance sheet at all.
• When. A business that generates in nine months what the facility demands in twelve does not have a credit problem. It has a structuring problem, and this is the single most common reason an otherwise sound proposal is declined.

The third of these is where structured credit earns its name. Where the cash is real but mistimed, the answer is not a higher margin to compensate for a risk that is not actually present. The answer is a repayment schedule dated to the cycle that produces the cash.

Why the distinction decides the structure

This is the part that matters commercially, and it is why the distinction is not academic. The two tests do not merely produce different analysis. They produce different facilities.

A facility underwritten primarily on collateral is structured to protect the lender's recovery route. The covenants sit on the asset: restrictions on disposal, on further encumbrance, on anything that dilutes the security or complicates a sale. The lender is comparatively indifferent to how the business is run day to day, because the business is not the repayment source.

A facility underwritten primarily on cash flow is structured to protect the lender's access to the cash. That means escrow, a trust and retention arrangement, or at minimum a designated account through which collections must pass. It means a waterfall fixing the order in which money is applied. And it means covenants that trip on coverage rather than on asset value.

Those are different documents, different negotiations and different costs to the borrower. A borrower who has prepared a collateral argument and is then asked for cash-flow control mechanics discovers late that the operational concession being sought, routing collections through an account the lender can see and control, is a larger concession than anything that was ever discussed on price.

Most facilities are hybrid, and the ranking is what matters

Very few structured credit facilities are purely one or the other. The useful question is therefore not which test applies, but which of the two is primary, because that is what governs the outcome when the two disagree.

A lender wants two ways out. What decides the structure is which one it is relying on, and which one it is merely keeping.

Where cash flow is primary and collateral secondary, the security exists to establish a floor rather than a plan. The lender expects to be repaid from operations and holds the charge so that the downside is bounded. Pricing follows the cash-flow analysis, and the advance rate can be higher because the asset is not doing the work.

Where collateral is primary, the reverse applies. The lender expects the asset to do the work and treats operating cash as useful but secondary comfort. Pricing follows the asset, the advance rate falls, and the covenant package tightens around disposal and encumbrance.

Confusing the ranking is how facilities are mispriced in both directions. A borrower with strong, controllable cash flow who accepts collateral-led terms pays for protection it did not need. A collateral-led proposal presented as a cash-flow story reaches term sheet and then dies in diligence, having consumed the time the borrower did not have.

How to present, if you are the borrower

State which question you are answering, and answer it properly, before the lender is left to work it out from the file.

• Lead with the repayment source, explicitly. Name what services the facility: a contracted receipt stream, a defined collection pool, a specific event with a date. That the business is performing well is context, not a repayment source.
• Show the whole obligation stack, including what does not appear as debt. A lender who uncovers an undisclosed prior claim in diligence reprices the entire facility, and reprices for the discovery rather than for the claim itself.
• Give realisable value rather than book value, and show how you arrived at it. A borrower presenting a conservative number they can defend is taken more seriously than one presenting a favourable number they cannot.
• Be explicit about any timing mismatch instead of smoothing it out of the model. A schedule that follows the cycle is a structure a lender can write. A schedule that assumes the cycle away is one they decline.

The practical conclusion

Proposals rarely fail because the underlying risk was unacceptable. They fail because the proposal was built to answer a question the lender was not asking, and because the mismatch surfaced only after enough time had passed that the borrower had run out of it.

Establishing which test governs, before anything is circulated and well before pricing is discussed, is not preparatory work. It is the decision that determines what the facility looks like, what it costs, and whether it closes at all.