NBFCs and housing finance: no deposits, and the asset-liability table that decides survival

On its profit-and-loss account, a non-banking finance company looks like a smaller, often more profitable bank. It earns interest on loans, pays interest on its funding, and keeps the spread - a wider spread than a bank's, in fact, because it lends to customers a bank will not touch. Return on assets can be excellent, capital adequacy well above the requirement, bad loans low. And it can still be dead within a fortnight, for a reason that appears nowhere on the P&L and nowhere in the capital ratios. illustrative

The reason is that an NBFC has no deposits. A bank funds itself with sticky retail deposits that stay put in a crisis; an NBFC funds itself by borrowing - from banks, from the bond market, from short-term commercial paper (short-dated IOUs it sells to investors to raise cash for a few months) - and that borrowing has to be repaid and replaced continuously. It borrows for months and lends for years, and it survives only as long as lenders keep refinancing the maturing borrowings. The day they stop - a market freeze, a rating downgrade, a scare in the sector - it must repay money it does not yet have back from its own borrowers, and no amount of capital or profit fixes a problem of timing.

So the most important table in an NBFC's annual report is not the P&L and not the capital statement. It is the asset-liability maturity table - the ALM - which lines up when liabilities fall due against when assets repay. The profit tells you how well the lender earns; the stage-3 assets tell you how good its loans are; but the ALM tells you whether it will still be here next quarter, and for an NBFC that is the question that comes first. Remove the deposit franchise and two numbers move to the centre of the read: the cost of funds, which becomes a competitive variable rather than a given, and the maturity gap, which becomes the survival question because wholesale borrowing must be actively refinanced in a way deposits never are. An NBFC is a bank with the deposits removed and the funding risk turned up. Housing finance is the same shape stretched further still - home loans that repay over fifteen to twenty years against funding measured in months - so everything here reads harder, not differently.

The ALM table sorts everything the NBFC will pay and receive into time buckets - up to a month, one to three months, three to twelve, one to three years, beyond three - and asks, in each bucket, whether more is coming in than going out.

The near-term gap is negative by design. An NBFC lends for years and borrows for months, so in the next few buckets more borrowing falls due than loans repay. The cumulative gap, running from the shortest bucket outward, is negative at the front and turns positive only further out. That is not a flaw - it is maturity transformation, the same thing every lender does. It becomes lethal only when the borrowings falling due cannot be refinanced.

In normal times the gap is filled by rolling over. Each month, maturing commercial paper and bank lines are simply replaced with new ones; the near-term deficit is never paid in cash because the lenders keep lending. An NBFC can run for years with a large negative near-term gap and look perfectly healthy - which is exactly why the gap lulls people.

In a freeze the gap must be met from cash - and that is the test. When refinancing dries up, the maturing borrowings have to be repaid from what the NBFC actually has: the loan repayments arriving over those months, plus whatever liquid buffer it holds. Enough, and it survives the window and buys time. Short, and it defaults - not because its loans went bad, but because it cannot pay what is due today from assets that repay tomorrow.

That is the distinction the whole module turns on. Solvency is whether the assets exceed the liabilities eventually; liquidity is whether the cash is there when the liability falls due. An NBFC can be solvent and illiquid at once, and illiquidity is what kills it. A liquidity default is terminal: the lender that cannot pay this month does not get to collect the loans that would have repaid next year. The cost of funds sits beside the ALM - it sets how wide a spread the lender needs and how exposed it is to a downgrade - and stage-3 with its ECL coverage tells you the quality of the book, read exactly as gross and net NPA were for a bank. But the ALM is read first, because it answers whether the company survives to worry about the others.

Where to find it: the maturity table is not on the face of the balance sheet. It sits in the notes to the accounts, usually under a heading like "Maturity analysis of assets and liabilities" required by the RBI's disclosure norms, alongside a separate liquidity-risk note and, for larger NBFCs, a liquidity coverage ratio. The company's own stress commentary - how much of the near-term gap it says is covered by liquid investments and undrawn bank lines - lives in that note and in the rating agency's press release, not in the P&L most readers stop at. Learning to turn to that note first is half the skill.

The maturity gap means completely different things depending on what funds it. The identical "borrow short, lend long" that is the safe, normal heart of banking for a deposit-funded bank is the single most common cause of death for a wholesale-funded NBFC - and wider still for a housing-financier. The structure on the balance sheet is the same; only the stickiness of the funding differs, and that is what decides everything.

This is why the regulator lets a bank lever ten times but holds an NBFC to a much higher capital ratio and scrutinises its ALM far harder: the bank's deposits are the cushion an NBFC does not have. Read an NBFC's maturity gap with a bank's tolerance and you miss the risk that actually matters. And read a housing-financier as merely a slightly-longer NBFC and you understate it - its assets are the longest of any lender, so its mismatch is the widest, and survival there depends more than anywhere on long-tenor funding and a deep liquidity buffer.

Read the composite vehicle-finance NBFC. On the P&L it looks strong: interest income of ₹7,100 crore against finance costs of ₹3,830 crore gives net interest income of ₹3,270 crore, a spread of about 6.9% - wider than any bank, because it lends to riskier borrowers. Return on assets is 2.7%, capital adequacy a comfortable 20.1%, gross stage-3 assets down to 3.9%. By every measure a bank reader reaches for, this is a good lender. illustrative

Now read the table that decides its life. Its assets - vehicle loans - repay over three to five years, but a large share of its ₹39,500 crore of borrowings falls due within a year. In the near buckets the ladder runs a deficit: in the up-to-one-month and one-to-three-month buckets more borrowing comes due than loans repay, and the cumulative gap is negative - around minus ₹2,300 crore by the three-month mark. In normal times this is invisible, because the maturing borrowings are rolled over each month. The company has run this way profitably for years.

Put it under stress and the picture changes instantly. Suppose a scare in the sector freezes wholesale funding and only part of the maturing borrowings can be refinanced. The un-rolled portion must now be repaid in cash - from the loan repayments arriving in those months plus whatever liquid buffer the NBFC holds. If the buffer is thin, the cash available falls short of what is due, and the company defaults on borrowings it fully intends and is able to repay over time. Its loans are good. Its capital is ample. And it is finished, because ₹2,300 crore was due before the assets backing it came home. That is not a hypothetical; it is precisely how large Indian finance companies have failed.

The habit to build: for any NBFC or housing-finance company, read the ALM before the profit. Look at the cumulative gap in the near buckets and ask one question - if refinancing stopped tomorrow, could this company meet its maturing liabilities from loan inflows and its buffer alone? Then read the cost of funds and its trend, because a downgrade that raises the cost of funds and cuts access to it is the trigger that turns the latent gap into a crisis - and access matters more than price, since a dearer spread is survivable but a withdrawn line is not. Only after those do you read the spread, the stage-3 assets and the return ratios. A profitable NBFC with a bad ALM is not a good business with a footnote; it is a fragile one wearing a good P&L.

The ladder shows the gap; this tool shows what happens to it under stress. Drop the rollover rate of short-term borrowings - that is a wholesale-funding freeze - and watch the near-term maturities that could always be refinanced suddenly demand cash. Set the liquidity buffer held against them.

At 100% rollover the company is fine: every maturing borrowing is replaced and the negative gap never has to be paid. Slide the rollover down and a shortfall opens - the cash that must be repaid exceeds loan inflows plus the buffer - and the verdict flips to liquidity failure, even though the loan book is solvent throughout. Raise the buffer and watch how much cushion it takes to survive a given freeze. The point becomes unavoidable: a well-capitalised NBFC with good loans can still be killed by the timing of its funding, and the only defences are matching maturities, diversifying funding, and holding real liquidity.

After any funding scare, the ALM is the first thing a serious analyst probes, and the question stress-tests the ladder past the company's own assumptions: "In a stress case where wholesale rollovers halt for three months, walk us through your positive liquidity - what is the cumulative gap in the up-to-three-month bucket, and how much of it is covered by liquid investments and undrawn bank lines?"

A good answer.

"We run exactly that scenario. In the up-to-three-month bucket our contractual cumulative gap is about ₹2,300 crore. Against it we hold ₹2,600 crore of liquid investments and ₹1,800 crore of undrawn, committed bank lines, so positive liquidity in a full three-month rollover halt is roughly ₹2,100 crore even before any incremental collections. We've deliberately termed out our borrowing this year - commercial paper is down to 8% of the mix from 19% - to reduce exactly this risk. It's all in the ALM disclosure and the liquidity note."

It gives the gap, the buffer against it, the effect of a genuine stress halt, and the structural action taken to reduce the risk, with the source. It lets you judge survival, not just profitability.

An evasive answer.

"We maintain a very comfortable liquidity position and are in full compliance with all RBI liquidity guidelines. We have strong, long-standing relationships with our lenders and access to diversified funding sources. Asset quality is robust and our capital adequacy is well above regulatory requirements. We don't foresee any liquidity concerns."

Reassuring and unusable. It cites compliance and relationships rather than the gap and the buffer, and it answers a liquidity question with capital adequacy - a solvency measure - which is the exact confusion that kills NBFCs. "Strong lender relationships" is precisely what evaporates in a freeze. No number for the three-month gap, no buffer figure, no stress result.

The follow-up nobody asks. "In a three-month rollover halt with zero fresh wholesale funding, what is your positive liquidity after meeting all maturing liabilities, and how does it change if your rating is cut one notch?" That forces the answer onto the survival number and its most likely trigger. If "comfortable, compliant, strong relationships" is allowed to stand, investors are trusting the funding to behave well in exactly the moment it is most likely to run. The silence is the tell - either the stressed liquidity is thin, or a downgrade would open a hole the company would rather not quantify on the record.

The ALM table shows the gap the company chose to disclose, on the assumptions it chose to make, and those assumptions can flatter it. Behavioural maturities - the guess that some borrowers prepay and some lenders stay - are management estimates, and an optimistic set can make a dangerous ladder look manageable. The honest read stress-tests the disclosed gap with harsher rollover and prepayment assumptions than the company used, which is exactly what the tool above lets you do. Nor does the on-balance-sheet ALM capture funding risk moved off it: NBFCs assign and securitise loan pools and enter co-lending arrangements, and the liquidity commitments attached can sit outside the headline table while still being the company's problem in a crisis. Those live in the notes on securitisation and co-lending, and they have to be read alongside the ALM, not instead of it. Finally, the table cannot tell you how the funding will behave when frightened, only how it is contracted - two NBFCs with identical ladders fare completely differently depending on who their lenders are, how concentrated the funding is, and whether a downgrade trips covenants that accelerate repayment. A gap funded by a handful of nervous wholesale lenders is far more dangerous than the same gap funded by diversified, staggered, long-tenor sources.

The traps follow directly from all this. The first is judging an NBFC by its profit and capital and concluding it is safe - a high return on assets and a fat capital ratio measure how well it earns and whether its assets exceed its liabilities, both genuinely good and both beside the point of survival. An NBFC does not fail because it is unprofitable or insolvent; it fails because it cannot refinance its borrowings when they fall due. The second is confusing solvency with liquidity - it is genuinely counterintuitive that a company whose loans are all good and whose capital is ample can still go under, but it can, if the assets repay later than the liabilities fall due and no one will bridge the gap. The third is reading a wide spread as pure strength - an NBFC earns more than a bank because it lends to riskier borrowers and funds itself more expensively, so a fat spread comes bundled with higher credit and funding risk, not instead of it. The spread tells you how much the lender makes when everything works; it says nothing about what happens when the funding stops, which is the scenario that decides whether the business exists.