Drawing Power vs Ratio Covenants in Cash Credit Facilities

Drawing power caps what a borrower may draw today, recalculated monthly from stock and book debts. Ratio covenants test whether the borrower is still the company you underwrote, measured quarterly or annually on financial statements. They run on different clocks and different data, so a borrower can pass one and fail the other in the same week.

Key facts

• YuSight compresses decision turnaround by 3x, with stock and book-debt statements auto-classified and mapped to the right borrower entity — which matters when drawing power is recomputed twelve times a year against covenants tested four times.
• A cash credit account goes NPA if it stays out of order for a continuous 90 days, where 'out of order' means the outstanding remains continuously above the sanctioned limit or drawing power, whichever is lower (RBI Master Circular RBI/2025-26/13, *Prudential norms on Income Recognition, Asset Classification and Provisioning*, 1 April 2025, para 2.2.1).
• Stock statements used to compute drawing power must not be older than three months. Drawing power calculated from an older statement is "deemed as irregular", and 90 continuous days of such drawings makes the account NPA "even though the unit may be working or the borrower's financial position is satisfactory" (same Master Circular, para 4.2.4(1)).
• SMA-1 begins at 31 days out of order and SMA-2 at 61 days for revolving facilities (RBI, *Prudential Framework for Resolution of Stressed Assets*, DBR.No.BP.BC.45/21.04.048/2018-19, 7 June 2019).
• For borrowers with aggregate fund-based working capital limits of ₹150 crore and above, at least 60% must sit in a loan component, not cash credit, effective 1 July 2019 (RBI/2018-19/87, DBR.BP.BC.No.12/21.04.048/2018-19, 5 December 2018).

What is each control actually protecting?

They protect different things, which is why they disagree.

Drawing power protects the security cover. It is an arithmetic statement: at today's stock and receivable levels, after the bank's margin, this is the value of current assets the charge actually covers. It is a collateral control, computed from a self-declared monthly statement, and it moves every month whether the business is doing well or badly.

Ratio covenants protect the credit thesis. TOL/TNW, current ratio, DSCR and interest cover test whether the borrower still has the leverage, liquidity and earnings profile the sanction assumed. They are financial condition controls, computed from audited or provisional financials, and they move once a quarter at best.

Neither is a substitute for the other. A fully secured borrower can be insolvent. A comfortably solvent borrower can have no eligible current assets on 31 October.

How is drawing power calculated in a cash credit facility?

Paid stock = Total stock − creditors for goods Stock component = Paid stock × (1 − stock margin) Book-debt component = Book debts within the eligible ageing × (1 − book-debt margin) Drawing power = Stock component + Book-debt component (capped at the sanctioned limit)

Three rules decide most of the outcome:

1. Only paid stock counts. Stock bought on credit is already financed by the supplier. Deducting sundry creditors for goods prevents the bank from funding it twice — and it is the single most common error in a DP working.
2. Only eligible book debts count. Most sanctions cap eligibility at 90 days from invoice date, exclude related-party receivables entirely, and exclude debts from parties who are also creditors to the extent of the set-off.
3. Margins are policy, not regulation. 25% on paid stock and 40% on book debts are widespread conventions, not RBI requirements — check your own sanction letter.

For the mechanics in more detail, including how margins move by commodity and how to treat consignment stock, see the drawing power calculator walkthrough, and for how the limit itself was sized in the first place, the MPBF calculation.

Why do the two controls run on different clocks?

The lag column is where the trouble lives. A borrower whose September quarter collapsed will show it in a DP working filed in October and in a covenant test filed in November or December — if the provisionals arrive on time. When they do not, the covenant test happens after the account has already been out of order for two months.

A 12-month example where the two diverge

Sundaram Fabrics Private Limited, a Tirupur textile processor. Cash credit limit ₹600 lakh. Margins: 25% on paid stock, 40% on book debts up to 90 days. Covenants tested as follows:

• Interest coverage ≥ 2.50x, quarterly on provisionals
• Current ratio ≥ 1.25x, annually on audited financials
• TOL/TNW ≤ 3.00x, annually
• DSCR ≥ 1.25x, annually

The monthly drawing power working (₹ lakh)

July working, line by line:

Paid stock = 610 − 230 = 380 Stock component = 380 × 0.75 = 285.00 Book-debt component = 580 × 0.60 = 348.00 Raw DP = 285.00 + 348.00 = 633.00 DP applied = min(633.00, 600) = 600.00 ← capped at limit

October working, the month it turns:

Paid stock = 590 − 225 = 365 Stock component = 365 × 0.75 = 273.75 Book debts ≤ 90 days fell from 600 to 520 — ₹80 lakh of the export receivable crossed the 90-day ageing cut-off and became ineligible. Book-debt component = 520 × 0.60 = 312.00 DP applied = 273.75 + 312.00 = 585.75 Outstanding = 594.00 Excess over DP = 8.25 ← account is 'out of order'

Nothing about the business changed in October. The customer did not default. Ninety days simply elapsed.

What the covenants said, on their own clock

Q2 (Jul–Sep 2025) provisional, filed 14 November 2025. EBITDA ₹48 lakh, interest ₹21 lakh.

Interest coverage = 48 ÷ 21 = 2.29x against a covenant of 2.50x → BREACH

At the end of September, when this quarter closed, the account had ₹8 lakh of headroom and had never been out of order. DP-compliant, covenant-breaching. The compression came from margin, not from security value, and drawing power has no way to see margin.

FY2026 audited financials, at 31 March 2026 (₹ lakh): revenue 2,850, EBITDA 245, D&A 62, interest 78, PBT 105, tax 26.2, PAT 78.8. Balance sheet: stock 470, debtors 560 (all ageings), cash 25, other current assets 40, net fixed assets 385, intangibles 20. Cash credit 560, creditors 195, other current liabilities 60, current portion of long-term debt 55, long-term debt 180, other non-current liabilities 30, net worth 420.

Current assets = 470 + 560 + 25 + 40 = 1,095 Current liabilities = 560 + 195 + 60 + 55 = 870 Current ratio = 1,095 ÷ 870 = 1.26x vs 1.25x → PASS TOL = 870 + 180 + 30 = 1,080 TNW = 420 − 20 = 400 TOL/TNW = 1,080 ÷ 400 = 2.70x vs 3.00x → PASS DSCR = (78.8 + 62 + 78) ÷ (78 + 55) = 218.8 ÷ 133 = 1.65x vs 1.25x → PASS Annual interest coverage = 245 ÷ 78 = 3.14x vs 2.50x → PASS

Four covenants tested on the audited balance sheet. All four pass. On the same date, the account had been continuously out of order since 8 October 2025, was classified SMA-1 in November, SMA-2 in December, and became an NPA in early January 2026 — roughly 90 days after the excess first appeared, in line with para 2.2.1 of the Master Circular.

Why the covenants could not see it

Look at the current ratio calculation again. Debtors of ₹560 lakh sit in current assets at full value. Those are the same receivables that aged past 90 days and destroyed drawing power. The covenant treats them as liquid; the DP working treats ₹140 lakh of them as worthless.

That is not a flaw in either control. It is the difference between an accounting classification and a collateral eligibility rule. The receivables that killed the drawing power are the receivables that saved the current ratio.

And the inversion runs the other way too. In September, drawing power was at the cap of ₹600 lakh — the security cover was so strong it was being wasted — while the Q2 interest coverage was already through the floor. Full security, failing economics.

What should the credit team actually do?

When DP is breached but covenants pass:

1. Get the debtor ageing, not the summary. The DP statement gives a single "≤90 days" figure. The ageing schedule tells you whether ₹140 lakh went stale in one customer or across forty. One customer is a concentration event; forty is a demand problem.
2. Reconcile the stock figure to something external. GST returns give an independent read on sales and purchases — the technique is set out in GST return analysis for lending. If the stock statement says ₹470 lakh and the GSTR-3B run rate implies half that, the statement is the problem, not the business.
3. Count the days precisely and in writing. The 90-day clock in para 2.2.1 runs on continuous excess. A single day back within DP resets it — which is why round-tripped credits appearing in the account on the 30th of each month deserve scrutiny rather than relief.
4. Do not fix it by revising the margin. Reducing the book-debt margin from 40% to 30% would have lifted October's DP by ₹52 lakh and cured the excess on paper. It changes the recorded asset classification without changing a rupee of recovery, and it is exactly the adjustment an inspection will find.

When covenants are breached but DP is comfortable:

1. Treat it as the earlier signal, because it is. The Q2 interest-coverage breach was visible in November. The DP breach was visible in October but was caused by ageing that began in July. Both point back to the same quarter.
2. Decide on the waiver before the borrower asks. A covenant breach is usually a contractual right to act, not an obligation. Choosing not to act is a decision that belongs in a note on file with a reason, not in a silence.
3. Re-test the sanction assumptions. If interest coverage fell because the limit was fully drawn all year at a rate that repriced, the covenant did not fail — the assessment did.
4. Tighten the reporting frequency before tightening the covenant. Monthly provisional P&L for two quarters tells you more than a renegotiated ratio that will next be tested in six months.

In both cases: the stock statement date matters as much as its contents. Para 4.2.4(1) deems drawings irregular if the statement behind the DP is older than three months, regardless of how healthy the numbers are. An account running on a statement from four months ago is irregular by definition, and 90 continuous days of that is an NPA.

How should this be monitored in practice?

Twelve DP workings, four covenant tests, one debtor ageing a month and a GST reconciliation is roughly 200 data points per borrower per year. Done in spreadsheets, the failure mode is not arithmetic — it is that nobody notices October until December.

What makes it tractable is having the stock and book-debt statement, the provisionals and the audited financials read, mapped to the same borrower entity and computed against the same sanction terms, with every figure traceable to the page it came from. YuSight's Document Intelligence classifies and maps the monthly statements, Financial Spreading standardises the financials and computes the leverage, liquidity and coverage ratios the covenants test, and the audit trail records what was computed, from which document, on which date. Lenders running that stack report 3x faster decision turnaround on the credit side — and, more usefully here, the monitoring stops depending on who remembered to open the file.

FAQ

How do you calculate drawing power in a cash credit facility?

Take total stock, subtract creditors for goods to get paid stock, apply the stock margin, then add eligible book debts after applying the book-debt margin. Cap the result at the sanctioned limit. Only stock the borrower has actually paid for and receivables within the eligible ageing count.

What margin is applied to stock and debtors?

Commonly 25% on paid stock and 40% on book debts up to 90 days, but these are bank policy and vary by segment, commodity and borrower. RBI does not prescribe margin percentages — read the sanction letter for the facility in front of you.

How often is drawing power recalculated?

Monthly, from the borrower's stock and book-debt statement. RBI's Master Circular requires that the statement behind a DP working is not older than three months; drawings on an older statement are deemed irregular, and 90 continuous days of that makes the account an NPA.

Can a borrower breach drawing power and still pass every covenant?

Yes, and Sundaram Fabrics is the case. The receivables that aged past the 90-day eligibility cut-off and collapsed drawing power still sat in current assets at full value, so the current ratio held at 1.26x while the account was already an NPA.

Which breach is more serious?

The drawing power breach, because it carries an automatic asset-classification consequence at 90 continuous days out of order. A covenant breach is an event of default that the lender may choose to waive; NPA classification is not a choice.

Does a covenant breach make the account an NPA?

No. Asset classification follows the account's conduct — days out of order, days past due — not the financial covenants. A borrower can breach every ratio in the facility agreement and remain a standard asset if the account stays within limits and is serviced.

What happens between the DP breach and NPA classification?

The account moves through SMA sub-categories: SMA-1 from 31 to 60 days out of order, SMA-2 from 61 to 90, and NPA beyond 90. Exposures above the reporting threshold also flow into CRILC, so the stress is visible to other lenders before the classification changes.

Can reducing the margin cure a DP breach?

Arithmetically yes, and it is a bad idea. Lowering the book-debt margin from 40% to 30% would have added ₹52 lakh to October's drawing power and erased the excess without improving recovery by a rupee. Margin revisions belong at renewal, with a documented rationale.

Should covenants be tested more often than quarterly?

Only where the data is reliable. Monthly provisional P&L is worth more than a quarterly test on unaudited numbers that arrive six weeks late, but testing a leverage covenant monthly on a balance sheet nobody has closed produces noise.

What is the first thing to ask for after a DP breach?

The debtor ageing schedule, not a revised stock statement. The ageing tells you whether the ineligible receivable is one customer or forty, which is the difference between a concentration event and a demand problem.

Key takeaways

• Drawing power is a collateral control on a monthly clock; ratio covenants are a financial-condition control on a quarterly or annual one. They will diverge, and the divergence is information.
• Sundaram Fabrics breached interest coverage in a quarter when it had ₹8 lakh of DP headroom, and passed all four annual covenants in a year it became an NPA.
• The 90-day continuous-excess clock is not discretionary. Count it precisely, and treat a stock statement older than three months as irregular regardless of what it says.
• The debtor ageing schedule explains both breaches. The single "≤90 days" figure on the DP statement does not.

For the wider credit context these controls sit inside, see the credit appraisal process in Indian banks and the 24 ratios that actually drive a lending decision.

See covenant testing run automatically — [book a live demo](https://yuverse.ai/yusight).