Liabilities & Financing

Phase 1 · Accounting: The Language of Business · Week 12 · ~20 hours (one immersive week)

You arrive here able to read all three statements and close the linkage loop. You can handle the asset side too, from revenue and working capital through to long-term assets and impairment. This week you cross to the right-hand side of the balance sheet. It starts with debt at amortized cost and leases. Provisions and contingencies come next, then deferred taxes. Leases are the centerpiece, because they're the single biggest comparability trap in modern financial statements. After this comes the equity module (stock-based compensation, consolidation), then the Phase 1 capstone. One thread runs through all four topics. Each of them feeds Question 4 of the five-questions spine, what can kill it? Companies rarely die from a bad income statement. They die from the right-hand side of the balance sheet.

Learning objectives

By the end of the week you can:

1. Price a bond from its coupon and market yield, state whether it issues at par, discount, or premium, and build a complete effective-interest amortization schedule (including the effect of issuance fees) from a blank sheet.
2. Explain why a lease is borrowing, compute a lease liability and right-of-use asset, and produce year-by-year income-statement effects under IFRS 16 / Ind AS 116 and under both ASC 842 models.
3. Quantify the EBITDA, EBIT, margin, cash-flow, and leverage differences between an IFRS/Ind AS reporter and a US GAAP operating-lease reporter for the same lease, and apply the EBITDAR fix.
4. Decide, for a given uncertain obligation, whether IAS 37 / Ind AS 37 and ASC 450 require a provision, a disclosure, or nothing, and compute the measurement under each, including where they give different numbers from identical facts.
5. Read an Indian contingent-liabilities note the way an analyst does: size it against net worth and identify the disputed-tax pipeline.
6. Build a deferred tax liability and a deferred tax asset from first principles, explain the valuation-allowance judgment, and read an effective-tax-rate reconciliation note line by line.
7. Locate and interpret the borrowings note, current maturities, lease note, provisions roll-forward, and tax note in a real 10-K and a real Indian annual report.

Prerequisites & connections

Builds on. M1.01, since every journal entry here is drilled in debits and credits. M1.03 for current vs non-current classification. M1.04 for where interest, lease payments and taxes land in the cash-flow statement. M1.05 for the linkage, since new liabilities create new paths between statements. M1.07 for depreciation mechanics, which the ROU asset reuses, and for the discounting habit.

Feeds forward. M1.09 reuses the present-value machinery for pensions and financial instruments. M1.10's hand-spread requires every note taught here. M1.11 works the entry for debt retired early, and lessor accounting waits for M1.11 too, taught from the lessor's side with this week's aircraft. Phase 2 turns this into judgment: leverage and coverage ratios (M2.02), the provisions roll-forward as a shenanigan detector (M2.06), cash-tax vs book-tax divergence (M2.07). Phase 3 needs effective interest for cost of debt and lease-adjusted capital for ROIC and WACC. Phase 5's airline, hotel, retail, and NBFC playbooks stand on the lease and borrowing mechanics here.

Core teaching content

4.0 The financing story on one page

Flip any balance sheet. The left side says what the business owns; the right side says who has claims on it and in what order. Lenders and lessors hold contractual claims: fixed amounts on fixed dates, senior to shareholders. Tax authorities hold statutory claims. Courts may create claims out of past conduct. Equity gets what is left. An analyst reads liabilities with one organizing question: how big is the claim, when is it due, and how certain is it? Debt is dated and certain. Leases are dated and certain, but were hidden for decades. Provisions are estimates. Contingent liabilities are possibilities. Deferred taxes are timing. This week walks that spectrum from most certain to least, exactly the order in which claims kill companies.

4.5 The financing read: this module on a real filing

From now on, run this five-note sweep on every filing. It becomes part of your 45–60 minute statement read. (1) The borrowings note: total, secured vs unsecured, currency, maturity ladder against cash. (2) Current maturities against liquidity. (3) The lease note: liabilities, discount rate, maturity table, variable payments. Recompute EBITDAR whenever you compare across borders. (4) Contingent liabilities: size against net worth, the tax-dispute pipeline, guarantees. (5) The tax note: the ETR reconciliation, DTA/DTL components, any valuation allowance or unrecognized DTAs, and the gap between cash and book tax. That's fifteen minutes and five notes, and afterward you'll know more about survival risk than most readers of the income statement ever will.

Common mistakes & how experts think differently

1. "Interest expense = the coupon we paid." That's wrong whenever debt is issued off par or with fees. Expense = carrying amount × effective rate, and the coupon is just the cash slice. Experts pull the effective rate from the debt note and treat that as the cost of debt.

2. "IFRS 16 / Ind AS 116 made companies riskier: look how debt jumped in 2019-20." The reporting caught up with reality, and the business didn't change. Experts re-base history, either by capitalizing the pre-2019 years too or by analyzing post-2019 only. They also don't let a company blame the standard when its lease commitments really are rising. Check whether the liability grows faster than the fleet or the store count.

3. Comparing EBITDA margins or EV/EBITDA across the GAAP/IFRS border. The ten-point margin gap in the twin-airlines table is pure accounting. For airlines, hotels, retail and shipping, experts reach for EBITDAR or fully lease-adjusted metrics by reflex. Before they trust a screen, they check which convention each data vendor's "EBITDA" field uses.

4. "US GAAP still keeps operating leases off balance sheet." That's a decade out of date. ASC 842 puts the liability and ROU on the balance sheet. The differences from IFRS are the expense pattern, which is a single straight-line cost inside EBITDA, plus where the cash flows land.

5. Trusting CFO comparability across standards. Ind AS shovels the entire lease payment, and all interest paid, into financing. ASC 842-operating shovels it all into operating. Experts define FCF after all lease payments, so the number holds up whichever standard applies.

6. Skimming the contingent-liabilities note as boilerplate. In India that note is a pipeline of possible solvency events. AGR moved out of it and became ₹50,000+ crore of real liability. Experts ratio it to net worth, track its growth and read the largest single dispute.

7. "A provision means cash has been set aside." A provision is a bookkeeping credit. There's no bank account behind it. And a provision release isn't operating income. It's an admission that the earlier expense was too big. Experts read the roll-forward (additions vs utilizations vs reversals) before they believe a margin improvement.

8. Treating every DTL as debt due next year, or ignoring the tax note entirely. A growth company's depreciation DTL may never shrink in aggregate. Meanwhile a valuation-allowance release can double net income with zero cash. Experts model cash taxes for valuation and build the forecast rate from recurring reconciliation items only. They treat DTA-driven earnings jumps as accounting events. Nothing changed in the business.

Worked examples

Worked example 1: Zero-coupon debt: accretion does all the work

Meridian Finance issues a zero-coupon NCD. Investors pay ₹100 crore today, and Meridian repays ₹133.10 crore in 3 years. The effective rate comes from 100 × (1+r)³ = 133.10, which gives r = 10%.

In year 1 the entry is Dr Interest expense 10.00 · Cr NCD liability 10.00. At maturity it's Dr NCD liability 133.10 · Cr Cash 133.10. Over three years the P&L carries ₹33.10 crore of interest, while the cash-flow statement shows zero interest paid until maturity. The whole borrowing cost is non-cash, right up until it suddenly isn't. Interest coverage computed on cash interest would flatter this issuer. The effective-interest expense tells you the truth.

Worked example 2: One lease, both standards, cradle to grave

The lease is ₹10 crore/year for 3 years, paid at year-end, with a discount rate of 6%. Initial liability = ROU = 10 × [1 − 1.06⁻³]/0.06 = ₹26.73 crore.

Under IFRS 16 / Ind AS 116, depreciation = 26.73/3 = 8.91:

Under ASC 842 operating there's a single straight-line lease cost of 10.00, and ROU amortization = plug = 10.00 − interest accretion:

Take three things from these tables. (1) The liability schedule is identical under both, with the same balance at every year-end. (2) Lifetime totals are identical too, at ₹30 crore. Only the pattern differs, front-loaded at 10.51/10.01/9.48 against a flat 10.00. (3) Under ASC 842-operating the ROU amortizes more slowly early on (8.40 < 8.91), so plug + interest = level rent. With level year-end payments, the ROU equals the liability at each year-end. Under IFRS the straight-line ROU runs below the liability mid-life, which means slightly lower equity in the early years. Every EBITDA and EBIT difference you saw in the leases material falls out of these two tables.

Worked example 3: Provisions: expected value, most likely, and the transatlantic gap

(a) Warranty: a large population, so expected value. Deccan Appliances sells 100,000 units under a one-year warranty. Experience says 90% bring no claim, 7% need a minor repair costing ₹400, and 3% need a replacement costing ₹2,500. Expected cost per unit = 0.07 × 400 + 0.03 × 2,500 = 28 + 75 = ₹103. Provision = 100,000 × 103 = ₹1.03 crore, booked as Dr Warranty expense · Cr Provision for warranties. US GAAP agrees here, since warranties in aggregate are probable and estimable.

(b) Single lawsuit: IFRS most-likely against the ASC 450 threshold. Counsel puts it at a 60% chance of losing ₹50 crore and 40% of paying nothing. Under IAS 37/Ind AS 37 that's probable (60% > 50%), and for a single obligation you take the most likely outcome, so you provide ₹50 crore. Under ASC 450, 60% typically fails the "likely" bar. There's no accrual, and you disclose it as reasonably possible. Identical facts, and ₹50 crore of difference in pre-tax profit.

(c) Probable loss, flat range: the range rule. A US subsidiary faces a probable loss of $100–300m, with no point in the range better than another. ASC 450 accrues the minimum, $100m, and discloses up to 300. IAS 37 takes the best estimate of a flat range, which is the midpoint, $200m. Keep both in mind when you compare a dual-listed company's IFRS accounts with a US peer's.

Worked example 4: Deferred tax end to end, with the ETR story

Kaveri Components Ltd (synthetic) has elected Section 115BAA, so its statutory rate is 25.168%. In FY1 its book EBT is ₹240 crore. That includes ₹6 crore of CSR expense, which is never deductible. It also includes a ₹20 crore provision expensed in the books but deductible only when paid, so its tax base is zero. Tax depreciation exceeds book depreciation by ₹36 crore.

Step 1: taxable income. 240 + 6 (CSR, permanent) + 20 (provision, temporary) − 36 (extra tax depreciation, temporary) = ₹230 crore. Current tax = 230 × 25.168% = ₹57.9 crore.

Step 2: deferred tax. The depreciation difference is a taxable temporary difference of 36, which gives DTL 36 × 25.168% = ₹9.06 crore of deferred expense. The provision is a deductible difference of 20, which gives a DTA of ₹5.03 crore, a deferred credit. Net deferred expense = ₹4.03 crore. CSR creates no deferred tax, because it never reverses.

Step 3: total and ETR. Total tax = 57.9 + 4.03 = ₹61.9 crore, so ETR = 61.9/240 = 25.8%. To reconcile, statutory 25.168% × 240 = 60.4, and the CSR effect of 6 × 25.168% adds +1.5 to reach 61.9, which ties. The temporary differences moved cash tax and left the ETR alone. The permanent difference moved the ETR by ~0.6 points, and it will do that every year. So 25.8%, and not 25.168%, is your forecast base. The entry is Dr Tax expense 61.9 · Cr Current tax payable 57.9 · Cr DTL (net of DTA) 4.0.

Worked example 5: A five-year US bond at a discount, with issuance fees

Cascade Semiconductor Inc. (synthetic) issues a 5-year note with face $500m and an annual coupon of 5%, which is $25m of cash a year. The market yield for its risk is 6.5%. Underwriting, legal and rating fees total $6m.

Step 1: price and classify. The coupon (5%) is below the yield (6.5%), so the note issues at a discount. Gross price = PV of five $25m coupons + PV of $500m face, all at 6.5% = $468.83m.

Step 2: fold in the fees. Both frameworks net issuance costs against the liability, so the carrying amount starts at net proceeds = 468.83 − 6.00 = $462.83m. The true cost, fees included, is the rate that discounts the promised cash flows (25, 25, 25, 25, 525) back to 462.83. Solve it and you get ≈ 6.80%. That's above the 5% coupon and the 6.5% gross yield. Phase 3 uses this 6.80% as the cost of debt. It never uses the coupon.

Step 3: the schedule. Interest expense = opening carrying amount × 6.80%. The coupon is a flat $25m, and the difference accretes the carrying amount from $462.83m toward face:

The table is in $m. Work to more decimals in a spreadsheet and the closing ties to 500.00. Lifetime interest expense = $162.17m, which is total coupons of $125m plus the discount and fees absorbed ($500 − 462.83 = $37.17m). Every dollar of fee becomes interest inside the effective rate. None of it is a day-one expense. Now read it as an analyst. An interest-coverage ratio built on the $25m cash coupon understates the true annual cost by $6–8m every year, so pull the effective rate from the debt note.

Worked example 6: A decommissioning provision and its silent interest

Coastal Cement Ltd (synthetic) commissions a limestone quarry that it's legally obliged to restore in 10 years, at an estimated cost of ₹80 crore. The risk-adjusted discount rate is 8%. You've seen the other half of this already. It's the liability twin of the site-restoration asset (ARO) you capitalized when you studied long-term assets.

Step 1: recognize at present value. IAS 37 / Ind AS 37 record the provision at the PV of the expected outflow, 80 ÷ 1.08¹⁰ = ₹37.06 crore. The same amount is capitalized into the asset's cost (Dr Asset 37.06 · Cr Provision 37.06) and depreciated over ten years. So the restoration cost reaches the P&L by two routes, as depreciation of the ROU-style asset and as the unwinding interest below.

Step 2: unwind the discount through finance cost. Each year the provision accretes at 8% (Dr Finance cost · Cr Provision), growing to ₹80 crore at year 10:

The table is in ₹ crore. Total finance cost over the life = 80.00 − 37.06 = ₹42.94 crore. That's real interest expense, and it appears on no loan document and pays no cash for a decade.

Here's the analyst's read. Firms with large decommissioning, mine-closure or asset-retirement obligations carry a provision that grows every year on its own. Think cement, mining, oil & gas, nuclear and telecom towers. That growth loads the finance-cost line without any borrowing. When you strip finance costs to build operating metrics, decide on purpose whether this unwind goes above or below your EBIT. And interrogate the discount rate. A lower rate inflates the day-one provision (and the asset), while a higher one defers the charge. So "provisions are estimates," from the financing story's opening page, turns into a live valuation judgment.

Practice set

Work each problem fully on paper or a blank sheet before you open the solution (working agreement #3). P1–P4 are guided, P5–P9 independent and P10–P12 timed. Give the timed three 45 minutes in all, with P12 alone capped at 15.

P1 (guided). Pricing and classifying. Face ₹100 crore, 2-year bond, 6% annual coupon, market yield 8%. Issue price, and discount or premium?

Solution. Price = 6/1.08 + 106/1.08² = 5.556 + 90.878 = ₹96.43 crore. The coupon (6%) is below the yield (8%), so it's a discount of 3.57. Check the logic as well as the formula. A stingy coupon means investors pay less than face.

P2 (guided). One year of effective interest. For P1's bond: year-1 interest expense, discount amortized, closing carrying amount, journal entry.

Solution. Expense = 96.43 × 8% = 7.71. The coupon is 6.00, so amortization is 1.71 and closing = 98.15. Check it: 98.15 × 1.08 − 6 = 100.00 at maturity, which ties. Entry: Dr Interest expense 7.71 · Cr Cash 6.00 · Cr Bonds payable 1.71.

P3 (guided). Premium schedule. Face ₹100 crore, 2-year, 10% coupon, yield 8%. Build the full schedule.

Solution. Price = 10/1.08 + 110/1.08² = 9.259 + 94.307 = ₹103.57 crore. In year 1, expense = 103.57 × 8% = 8.29, premium amortized = 10.00 − 8.29 = 1.71, and the closing balance is 101.85. In year 2, expense = 101.85 × 8% = 8.15, 1.85 is amortized, and the closing balance is 100.00. On a premium bond interest expense falls each year, because the carrying amount shrinks.

P4 (guided). Fees raise the rate. A 3-year bond has face ₹100 crore and a 9% coupon. It's priced at par and carries ₹3 crore of issuance costs. (a) Find the net proceeds and the initial carrying amount. (b) Is the effective rate above or below 9%, and roughly what is it? (c) Say which way year-1 interest expense moves against the no-fee case.

Solution. (a) ₹97 crore for both. (b) Above, because you receive less money for the same promised payments. Solve 9·AF(r,3) + 100·PF(r,3) = 97. At 10% PV = 97.51 and at 10.5% PV = 96.30, so interpolating gives ≈ 10.2%. (c) Year-1 expense = 97 × 10.2% ≈ 9.90 against 9.00, so it's higher, and the carrying amount accretes from 97 to 100 over three years.

P5. IFRS 16 / Ind AS 116 lease. 4-year lease, ₹8 crore/year paid year-end, IBR 7%. Initial liability/ROU, year-1 interest, depreciation, total expense vs rent.

Solution. Liability = 8 × [1 − 1.07⁻⁴]/0.07 = 8 × 3.3872 = ₹27.10 crore = ROU. Year-1 interest = 27.10 × 7% = 1.90 and depreciation = 27.10/4 = 6.77, for a total of 8.67 against rent of 8.00. That's front-loaded by 0.67, which unwinds in later years. Lifetime totals equal ₹32 crore either way.

P6. Same lease under ASC 842 operating: the EBITDA gap. The lessee has revenue ₹40 crore and other operating costs ₹20 crore. Year-1 EBITDA, EBIT, pre-tax profit under (a) Ind AS 116 and (b) ASC 842 operating.

Solution. (a) EBITDA = 40 − 20 = 20.00, because the lease cost is below the line. EBIT = 20 − 6.77 = 13.23, and pre-tax = 13.23 − 1.90 = 11.33. (b) The single lease cost goes inside opex, so EBITDA = 40 − 20 − 8 = 12.00, EBIT = 12.00 and pre-tax = 12.00. For the Ind AS reporter the gaps are EBITDA +8.00 (the whole rent), EBIT +1.23 (the interest portion) and pre-tax −0.67 (front-loading). Same business, same cash.

P7. The fix and the leverage read. Take P5/P6's twins. They have no other debt, and a year-end lease liability = 27.10 × 1.07 − 8 = 21.00. Find EBITDAR for both, then compare lease-inclusive net debt/EBITDA with net debt/EBITDAR.

Solution. EBITDAR for the US twin = 12 + 8 = 20. For the Ind AS twin it's 20 + 0 = 20, since there's no rent line to add back. Now they're comparable. Lease-inclusive leverage on raw EBITDA gives Ind AS 21/20 = 1.05× and US 21/12 = 1.75×, which look misleadingly different. On EBITDAR both are 21/20 = 1.05×. Always pair a consistent numerator with a consistent denominator.

P8. Classification quickfire. For each case, state the treatment under IAS 37/Ind AS 37 and under ASC 450. (a) A lawsuit with a 40% chance of losing ₹30 crore. (b) A lawsuit with a 75% chance of losing ₹30 crore. (c) Expected future operating losses of a weak division. (d) An unavoidable contract, now loss-making by ₹12 crore, where the cheapest exit penalty is ₹9 crore. (e) The CEO tells a journalist a restructuring is "being studied," and no plan is approved. (f) A customs dispute judged remote.

Solution. (a) It's not probable under either, so disclose it (as a contingent liability, or as reasonably possible). (b) It's probable under both. IAS 37 says provide ₹30 crore, the most likely outcome. Under ASC 450, 75% meets most practitioners' "probable," so accrue ₹30 if it's estimable. (c) Never provisioned under either. There's no present obligation from a past event, and the losses belong to future periods. (d) Under IAS 37 the contract is onerous, so provide the lower of 12 and 9 = ₹9 crore. US GAAP has no general onerous-contract provision, so nothing is booked until losses are incurred. (e) There's no detailed formal plan and no valid expectation raised, so no provision under either. (f) Remote means nothing, and no disclosure is required.

P9. Warranty + the roll-forward read. Zenith Tools accrues warranties at 2% of sales. Sales ₹900 crore; opening provision ₹21 crore; claims paid ₹14 crore; year-end review reverses ₹5 crore as excess. (a) Expense charged and closing balance? (b) Seeing reversals of 5 against additions of 18, what two questions should an analyst ask?

Solution. (a) Addition = 2% × 900 = ₹18 crore, and closing = 21 + 18 − 14 − 5 = ₹20 crore. The net P&L charge is 13 if the reversal runs through the same line. (b) First, is the 2% rate still right? Actual claims of 14 plus an admitted 5-crore excess suggest over-providing, which would mean prior profits were understated and this year's are flattered. Second, is the release a one-off, or a pattern across 3–5 years of roll-forwards? A serial over-provide-and-release cycle is the cookie-jar mechanic the earnings-games module ahead dissects.

P10 (timed). DTL build. Machine ₹200 crore; book depreciation ₹100/year × 2; tax depreciation ₹140 then ₹60; rate 25%; book EBT ₹150 crore/year. Build both years: taxable income, current tax, carrying amount vs tax base, DTL, deferred and total expense, ETR.

Solution. In year 1, taxable = 150 + 100 − 140 = 110, so current tax is 27.5. CV 100 against TB 60 gives a difference of 40 and a DTL of 10.0. Deferred tax is 10.0, total is 37.5 and the ETR is 25%. In year 2, taxable = 150 + 100 − 60 = 190, so current tax is 47.5. CV = TB = 0, so the DTL is 0 and deferred tax is (10.0). Total is again 37.5, with an ETR of 25%. Cash tax swings from 27.5 to 47.5 while book expense holds steady. That's deferred tax doing its smoothing job.

P11 (timed). DTA and the valuation allowance. A US startup (21% federal rate only, for simplicity) has $200m of accumulated NOLs. (a) Compute the full DTA. (b) Management judges only 60% of the NOLs more-likely-than-not usable. Find the valuation allowance and the net DTA. (c) Two years later, sustained profits justify releasing the entire allowance. Give the P&L effect and the cash effect. Then say what adjustment an analyst should make.

Solution. (a) 200 × 21% = $42.0m. (b) Allowance = 40% × 42 = $16.8m, leaving a net DTA of $25.2m. (c) The entry is Dr Valuation allowance 16.8 · Cr Tax expense 16.8, so net income jumps $16.8m and the cash effect is zero. Exclude it from run-rate earnings. Do note the real signal in it, which is that management now expects sustained profits. Then check that next year's consensus isn't lapping an inflated base. Under IFRS/Ind AS you'd recognize $25.2m directly and disclose the rest as unrecognized. The substance is the same, with no allowance line.

P12 (timed: 15 minutes). Reading an ETR reconciliation. Nimbus Infotech (Indian, 115BAA) discloses for FY1 (EBT ₹500 crore): tax at statutory 25.168% = 125.8; lower-taxed overseas branches (9.0); CSR/non-deductible +3.5; prior-period true-up (6.5); recognition of previously unrecognized acquired-company losses (8.0); total tax expense 105.8 → ETR 21.2%. (a) Pick out the items that belong in next year's forecast ETR. (b) Estimate the sustainable ETR. (c) Which item tells you nothing about this year's cash taxes, and why? (d) If consensus models 21% tax next year, what's the risk?

Solution. (a) The recurring items are the statutory rate, the overseas-rate benefit (while the mix persists) and the CSR add-back. The one-offs are the true-up and the DTA recognition on acquired losses. (b) 125.8 − 9.0 + 3.5 = 120.3, which is ≈ 24.1%. (c) The DTA recognition (8.0). It's a purely book, deferred credit. Cash tax follows the current-tax computation. (d) Consensus has baked one-offs into perpetuity. Tax expense will likely jump ~3 points, cutting EPS ~4% against the model. That's a forecastable "miss," and you can see it coming in a footnote.

P13 (guided). A four-year discount bond, first two years (US $). Face $200m, 4-year note, 4% annual coupon, market yield 5.5%. (a) Issue price and classification. (b) Year-1 and year-2 interest expense and closing carrying amounts.

Solution. (a) The coupon is $8m, and price = 8·AF(5.5%,4) + 200·PF(5.5%,4) = $189.48m. That's a discount, since the coupon of 4% is below the yield of 5.5%. (b) In year 1, expense = 189.48 × 5.5% = 10.42, amortization is 2.42 and the closing balance is 191.91. In year 2, expense = 191.91 × 5.5% = 10.55, amortization is 2.55 and the closing balance is 194.46. Expense rises as the carrying amount accretes toward the $200m face, while the $8m coupon never moves.

P14 (guided). Where the lease cash goes (India ₹). An Indian retailer signs a 5-year store lease, ₹9 crore/year paid year-end, IBR 8%, reported under Ind AS 116. (a) Initial lease liability. (b) Year-1 interest and principal split. (c) Where do the ₹9 crore land in the cash-flow statement under Ind AS 7 versus a US peer on an ASC 842 operating lease, and the resulting CFO difference?

Solution. (a) Liability = 9 × [1 − 1.08⁻⁵]/0.08 = 9 × 3.9927 = ₹35.93 crore = ROU. (b) Interest = 35.93 × 8% = ₹2.87 crore, and principal = 9.00 − 2.87 = ₹6.13 crore. (c) Ind AS 7 forces interest paid into financing, so the entire ₹9 crore lands in CFF and CFO is untouched. The US peer's ASC 842 operating lease routes all ₹9 crore through CFO. The cash is the same, but the Indian retailer's CFO looks ₹9 crore higher every year. Subtract all lease payments before you compare free cash flow.

P15. Three provisions, two rulebooks (India ₹). For each case, state the recognition and the amount under IAS 37 / Ind AS 37 and under ASC 450. (a) A one-year warranty on 200,000 units sold, where experience is 92% no claim, 6% a ₹500 repair and 2% a ₹3,000 replacement. (b) A single lawsuit with a 65% chance of losing ₹40 crore, and nothing otherwise. (c) A probable loss with a flat range of ₹60–120 crore, with no outcome more likely than another.

Solution. (a) It's a large population, so use expected value. Cost/unit = 0.06 × 500 + 0.02 × 3,000 = ₹90, and provision = 200,000 × 90 = ₹1.8 crore under both frameworks, because warranties in aggregate are probable and estimable. (b) Under IAS 37 it's probable (65% > 50%) and a single obligation, so take the most-likely outcome and provide ₹40 crore. Under ASC 450, 65% typically fails the "likely" bar, so you disclose only, with no accrual. (c) ASC 450 has no point better than another, so it accrues the minimum, ₹60 crore. IAS 37 takes the best estimate of a flat range, the midpoint, ₹90 crore. Watch the trap here. Expected value is for populations and most-likely is for single events, so never probability-weight one lawsuit.

P16. A US DTL build with a permanent difference (US $). A US company (blended statutory rate 25%) reports book EBT $400m, which includes $12m of non-deductible regulatory fines. Tax depreciation exceeds book depreciation by $40m. (a) Taxable income and current tax. (b) The deferred tax and total tax expense. (c) The ETR, and the rate you would carry into a forecast.

Solution. (a) Taxable = 400 + 12 (fines, permanent add-back) − 40 (extra tax depreciation, temporary) = $372m, and current tax = 372 × 25% = $93m. (b) The depreciation difference gives DTL = 40 × 25% = $10m of deferred expense, so total tax = 93 + 10 = $103m. (c) ETR = 103/400 = 25.75%. To reconcile, 25% × 400 = 100, and fines of 12 × 25% = 3 bring it to 103, which ties. The fines recur, so 25.75%, and not 25%, is the forecast rate. The depreciation difference moved cash tax only.

P17. Reading the Indian contingency note (India ₹). Bharat Infra Ltd (synthetic) has net worth ₹4,000 crore. Its contingent-liabilities note lists: disputed income tax ₹1,800 crore, disputed GST ₹500 crore, bank guarantees ₹300 crore, and one arbitration claim ₹900 crore. The ratio was 22% two years ago and 30% last year. (a) Contingent liabilities ÷ net worth now. (b) The tax-dispute and single-largest-item concentrations. (c) Does this belong on your Question-4 list, and why?

Solution. (a) Total = 1,800 + 500 + 300 + 900 = ₹3,500 crore, so CL ÷ net worth = 3,500/4,000 = 87.5%. (b) Tax disputes (income tax + GST) = 2,300/3,500 = 65.7% of the note, and the single arbitration = 900/3,500 = 25.7%. (c) Yes. The ratio has jumped from 22% to 30% to 87.5%, and it's well past the ~50%-of-net-worth flag. It's also concentrated in the disputed-tax pipeline plus one large arbitration. That's the AGR-style pattern, where a contingency can crystallize into a solvency event. Read management's assessment and any auditor emphasis-of-matter next.

P18 (timed: 15 minutes, all five parts). Cross-border financing sprint ($ and ₹). (i) A US note is carried at $478m, with an effective rate of 6% and an annual coupon of $30m. Find the year-end carrying amount. (ii) Find the initial liability on an Indian lease of ₹7 crore/year for 4 years, paid at year-end, IBR 9%. (iii) A probable US loss has a flat range of $80–160m. Give the accrual under ASC 450 and under IAS 37. (iv) An Indian asset has carrying amount ₹180 crore, tax base ₹120 crore and rate 25.168%. Compute the DTL. (v) A shipping firm is locked into charters ₹15 crore/year above market for two more years, with an exit penalty of ₹22 crore. What provision does Ind AS 37 require, and how does US GAAP differ?

Solution. (i) 478 × 1.06 − 30 = $476.68m, accreting toward face. (ii) 7 × [1 − 1.09⁻⁴]/0.09 = 7 × 3.2397 = ₹22.68 crore. (iii) ASC 450 accrues the range minimum, $80m. IAS 37 takes the midpoint, $120m. (iv) The taxable temporary difference gives 60 × 25.168% = ₹15.10 crore of DTL. (v) It's an onerous contract, so provide the lower of fulfilment (2 × 15 = ₹30 crore) and exit (₹22 crore), which is ₹22 crore. US GAAP has no general onerous-contract provision, so the loss is recognized as incurred or through impairment of related assets. To score yourself: 5/5 in time means you're fluent across all four topics. At ≤3, revisit the worked examples before the mastery check.

Applied mini-project

The financing stack, live. You'll work on two companies. For InterGlobe Aviation (IndiGo), get the latest annual report from the investor-relations page or BSE/NSE. For Delta Air Lines, get the latest 10-K from SEC EDGAR (sec.gov, full-text search). They're two lease-heavy airlines on opposite sides of the standards border, which is where this week's material bites hardest. Allow 3–4 hours.

1. Debt: from each borrowings/debt note, list total borrowings (ex-lease), the next 12 months' maturities, and cash + equivalents. One line each: comfortable, tight, or refinancing-dependent?
2. Leases: pull lease liabilities, disclosed discount rate(s), and the undiscounted maturity table. Delta: note the operating-vs-finance split and where operating lease cost shows up in the P&L. IndiGo: note ROU depreciation and lease finance cost separately.
3. The trap, quantified: compute each airline's EBITDA margin as reported, then EBITDAR margin (for Delta add back aircraft and other rent). Show the before/after gap in a two-row table.
4. Contingencies: summarize each contingencies note in ≤5 lines. For IndiGo, compute contingent liabilities ÷ net worth.
5. Tax: from each ETR reconciliation, name the two largest reconciling items and classify each recurring vs one-off; note any unrecognized DTAs or valuation allowance.
6. Memo: one page, conclusion first: "Which airline's financing structure is riskier, and what two numbers prove it?"

Scoring rubric. Grade yourself after writing. You pass at ≥ 12/16.

File both companies in your company files, and add the best insight to your permanent notes (the toolkit and knowledge system you set up in Phase 0).

Case Lab: run case CL-1.4 (Two Airlines, One Fleet) in the app's Cases tab this week: the same leased aircraft under Ind AS 116 and ASC 842, walked from lease liability and ROU through EBITDAR, three leverage readings, and cash-flow geography to a survival verdict.

Reading & resources

• Kieso, Weygandt, Warfield, Intermediate Accounting (any recent edition). [Paid] [Intermediate]: chapters on Current Liabilities & Contingencies, Long-Term Liabilities, Leases, and Income Taxes; unmatched drill bank.
• EY, US GAAP vs IFRS: The Basics (free PDF, updated annually). [Free] [Intermediate]: the leases, provisions, and income-tax sections are this week in professional shorthand.
• Deloitte iasplus.com. [Free] [Intermediate/Advanced]: summaries of IFRS 16, IAS 37, IAS 12 (Ind AS 116/37/12 are converged mirrors; bare Ind AS texts on the MCA website).
• KPMG Leases Handbook (ASC 842). [Free] [Advanced]: free download; the lessee chapter shows the operating-lease "plug" mechanics with worked tables.
• Aswath Damodaran, "Operating Leases: Dealing with the New Standards." [Free] [Intermediate]: plus the oplease.xls converter (his NYU pages), the analyst's-eye view of capitalization and comparability.
• Farhat Lectures (YouTube). [Free] [Beginner→Intermediate]: playlists on bonds/effective interest, ASC 842 leases, and ASC 740 deferred taxes; watch after your own first pass.
• Wharton/Bushee, Introduction to Financial Accounting (Coursera). [Free] [Beginner]: the liabilities and income-tax weeks as a gentler second pass.
• CFA Institute Level I curriculum. [Paid] [Intermediate]: FSA readings "Non-Current Liabilities" and "Income Taxes."
• Filings to pull (all free). [Free] [All levels]: the IndiGo annual report (leases, EBITDAR, contingencies) and the Delta Air Lines 10-K (leases, contingencies). The Apple 10-K has the term-debt note and effective rates, and the Tesla FY2023 10-K the valuation-allowance release. The Tata Motors FY24 annual report shows the JLR DTA recognition, filed before the October 2025 demerger. A Maruti Suzuki or ITC annual report shows contingent-liabilities anatomy, and the Vodafone Idea annual report covers AGR provisions and unrecognized DTAs. Get them from SEC EDGAR and BSE/NSE company pages.
• Zerodha Varsity: Fundamental Analysis. [Free] [Beginner]: the annual-report chapters, for Indian-notes navigation.
• Going deeper: IFRS Foundation, IFRS 16 Leases: Effects Analysis (ifrs.org, free PDF). [Free] [Intermediate]: the standard-setter's own quantification of the ~US$3.3 trillion of off-balance-sheet leases and the expected effects on leverage, EBITDA, and ratios; the evidence behind the leases section above. Read it right after the twin-airlines table.
• Going deeper: Howard Schilit, Jeremy Perler & Yoni Engelhart, Financial Shenanigans (4th ed.), the liability and reserve chapters. [Paid] [Advanced]: how provisions, deferred revenue, and off-balance-sheet obligations are used to flatter results; a preview of the provisions-roll-forward forensics the earnings-games module ahead weaponizes.

The Modern Analyst's Addendum

Everything above teaches this skill from first principles, by hand; that is how you learn it, and the mastery check still tests it that way. This addendum shows how a working analyst amplifies the same skill today. It adds; it never replaces. (R1/R10)

AI-Augment this skill

{
  "skill": "Effective-interest schedules, lease capitalization under three rulebooks, the provisions/contingency recognition tests, and reading a deferred-tax note",
  "use": "This module is where the two frameworks disagree most, and disagreement is exactly what an assistant is good at surfacing and bad at settling. (1) Paste a fact pattern — a lawsuit, an onerous charter, a restructuring announcement — and ask for the IAS 37/Ind AS 37 versus ASC 450 treatment side by side with the paragraph reference and the resulting rupee number for each; the gap between the two answers is the finding, and §4.3's table is your key. (2) Ask a document tool to pull the borrowings, lease, contingent-liabilities and tax notes out of ten annual reports at once, so the five-note sweep in §4.5 runs over a peer set instead of over one filing. (3) Ask it to restate a tax-rate reconciliation as recurring versus one-off, then argue the classification back — the one-offs it wants to call recurring are usually the interesting ones. (4) Ask it to name what a note does NOT disclose (variable lease payments, the discount rate behind a provision, the composition of an unrecognized DTA); absence is the hardest thing to notice by reading.",
  "tools": ["Chat assistants — Claude, ChatGPT for IFRS-versus-US-GAAP treatment pairs on a stated fact pattern", "Document/PDF AI — NotebookLM or Claude file upload for the five financing notes across a peer set", "Coding assistants — for the amortization, lease and EBITDAR code below"],
  "prompt": "Here is a note from a filing: <PASTE>. Produce a table [Item | IAS 37/Ind AS 37 or IFRS 16 treatment | ASC 450/842 treatment | Amount recognized under each | Standard paragraph]. Then list separately: (a) every figure you took directly from my paste, (b) every figure you computed and the formula used, (c) every figure you could not obtain from the paste. Do not supply a discount rate, a probability, or a useful life I have not given you.",
  "verify": "Two reconciliations, always. First, the arithmetic: lifetime interest expense must equal total coupons plus the full discount, and lifetime lease expense must equal total lease payments — a schedule that does not close on those identities is wrong regardless of how confident its prose is. Second, the source: read the filer's own disclosed discount rate, weighted-average remaining term and provisions roll-forward rather than any rate the model volunteers, because an invented IBR moves the liability several percent (see the Quantitative lens) and an invented probability moves a provision from a footnote to the balance sheet. A model asked which framework applies will always answer; only the standard's text decides (Guardrail 1-5).",
  "diy": "The gate is unaided: you build the effective-interest schedule with issuance fees, the lease liability and ROU roll-forward under both regimes, the EBITDAR reconciliation, and the DTL/DTA schedules from a blank sheet. Nothing in this callout is admissible in that room."
}

Modern Data Analysis

By hand first. You built the discount-bond schedule, the six-year lease roll-forward and the DTL table above on paper, and you still need to be able to. The code below writes the same three recurrences once. The point isn't any single schedule. It's that the comparability adjustments from the leases section and the financing-read sweep can then run across a whole peer set in the time it takes to read one note.

Today's workflow. Every schedule this week is a recurrence over an opening balance, so each one takes three or four lines of code. That holds for a bond's carrying amount, a lease liability, a provision's discount unwind, a DTL balance. Two jobs that are prohibitive by hand then become cheap. The first is solving for a rate instead of a balance. Issuance fees mean the economic cost of a borrowing is an internal rate of return, and that isn't a disclosed yield. A root-finder gets it to the decimal. This week's discount bond nets ₹93.0263 crore after ₹2 crore of fees and prices to an effective rate of 10.8461%. That's above the 10% gross yield, and well above the 8% coupon. The second job is making a cross-border peer set comparable in the first place. Compute EBITDAR and (debt + lease liabilities) ÷ EBITDAR for every filer in a frame. The ten-point EBITDA-margin gap between the twin airlines from the leases section closes to zero, and both twins land on EBITDAR of 30.00, because the adjustment is now a column. It used to take an afternoon.

Tools & sources (IN + US). Use numpy_financial / scipy.optimize.brentq for the rate solves, and pandas for the schedules and the peer panel. For India, work from the annual report (BSE/NSE): the Schedule III borrowings note, the Ind AS 116 lease-maturity and discount-rate disclosures, the provisions roll-forward, the contingent-liabilities note and the Ind AS 12 tax note. RBI DBIE's G-sec yields and WALR series give you a sanity check on any disclosed incremental borrowing rate. For the US, use EDGAR XBRL companyfacts (OperatingLeaseLiability, FinanceLeaseLiability, LongTermDebtNoncurrent, DeferredTaxLiabilities, OperatingLeaseWeightedAverageDiscountRatePercent) plus the 10-K's Debt, Leases, Commitments-and-Contingencies and Income-Taxes notes. FRED (DGS10, BAMLC0A0CM) supplies the rate and spread context.

# Synthetic, module's own figures — R3: every number recomputed in-session
from scipy.optimize import brentq
pv = lambda pmt, r, n: pmt * (1 - (1 + r) ** -n) / r
price = sum(8 / 1.10 ** t for t in (1, 2, 3)) + 100 / 1.10 ** 3   # 95.0263
eff = brentq(lambda r: sum(8 / (1 + r) ** t for t in (1, 2, 3))
             + 100 / (1 + r) ** 3 - (price - 2.0), 1e-4, 1.0)      # fees -> 0.108461
liab = pv(10, 0.06, 6)                                             # 49.1732, dep 8.1955
closing = liab * 1.06 - 10                                         # 42.1236 at year 1 end
ebitda_ifrs, ebitda_us, rent = 30.0, 20.0, 10.0                    # the §4.2 twins
print(round(price, 4), round(eff * 100, 4), round(liab, 4), round(closing, 4))
print(ebitda_ifrs, ebitda_us + rent)                               # 30.0 30.0 — EBITDAR ties
print(round(closing / ebitda_ifrs, 2), round(closing / ebitda_us, 2))  # 1.4 vs 2.11

Verify. Reconcile before you compare. Tie the closing lease liability your code produces to the filer's own balance-sheet line, and to its disclosed weighted-average discount rate and remaining term. If you had to guess the rate, say so in the memo, because the guess is doing more work than the model. Tie the borrowings schedule to the maturity ladder and the unamortized-issuance-cost disclosure. Then check what the frame can't see. Variable lease payments are outside the liability by design, as the leases section showed. So any purely liability-based screen will systematically understate a turnover-rent retailer's obligations, and no amount of code fixes a number that was never disclosed.

Quantitative lens

Two of this week's biggest liabilities depend on a discount rate that management chooses. That makes them the right place to learn the difference between a measured liability and an estimated one. Hold this week's lease fixed at ₹10 a year for six years, and vary only the incremental borrowing rate. At 5% the liability is 50.757. At 6% it's 49.173, at 8% it's 46.229 and at 10% it's 43.553. A disclosed IBR 200 basis points higher shrinks reported lease debt by 5.99%, and a 400bp one by 11.43%, with no change at all to the contractual cash. So the disclosed rate deserves a screen of its own. Compute each peer's implied IBR against a common benchmark, either its own bond yields or a G-sec or Treasury curve plus a rating spread. A filer that discounts a long lease book at a rate well above what it demonstrably borrows at is reporting less lease leverage than it has.

The same arithmetic retires an old heuristic, and that's the more useful lesson. Before 2019, the rule of thumb capitalized leases at seven to eight times annual rent. At a 6% discount rate, the present value of a level rent stream equals 8× rent only at a term of about 11.2 years. At 8% it takes about 13.3 years. This week's six-year lease capitalizes at 4.92× rent, so the old heuristic would have overstated this liability by roughly 63%. For a twenty-year property book at 6% the multiple is 11.47×, and the heuristic understates it by a third. The heuristic was a single point on a curve, applied everywhere. It never held as a constant across rates and terms.

This has limits. The sensitivity describes the measurement, and it says nothing about the risk. A lease's danger is that it can't be cancelled and where it ranks, and neither changes when the discount rate does. A screen on implied IBR will also flag lease books that are simply different long before it flags anything questionable. A five-year equipment book and a thirty-year property book properly carry different rates. So treat cross-sectional discount-rate work as a candidate generator. You confirm or drop a finding in the maturity table and the variable-payment disclosure.

Do it in code: build the peer panel of disclosed lease liability, weighted-average discount rate and remaining term. Invert each to an implied level rent, re-discount every filer at one common curve, and read the dispersion in restated leverage instead of in the reported figures. See G1/FI1.01 and G1/QM1.02.

Where this goes next: galaxy cross-links

• G1 · FI1.01 Bond Pricing & Yield Measures: the effective-interest method from the debt-at-amortized-cost section above is bond mathematics with the borrower's pen instead of the lender's; the yield-to-maturity you solve for there is the effective rate you amortize at here, and the issuance-fee solve is a plain IRR.
• G1 · FI1.02 Duration & Convexity: why the lease and provision balances above move as much as they do when the discount rate moves, expressed once as a sensitivity instead of re-tabulated per rate.
• G1 · QM1.02 Hypothesis Testing & Regression: deciding whether a filer's implied borrowing rate or contingent-liability ratio is really out of line with its peers or merely in the tail of a noisy, small-sample distribution.
• G2 · DA1.02 Pandas Data Wrangling: the five-note financing sweep from the financing-read section above, as a frame over a peer set, where the EBITDAR and lease-adjusted-leverage columns cost nothing after the first company.

Teach it back & journal

Feynman prompt. Explain to a smart 15-year-old why two airlines flying the same planes on the same leases can report EBITDA margins ten points apart, and why neither is lying. You must use the phrase "where the rent hides," write no more than one page, include one tiny numeric example (three lines suffice), and end with the one metric that makes the airlines comparable again.

Journal reflection. Before this week, which liability would you have worried about most on a balance sheet? And which of the four worries you most now, for the next company you'll read this Saturday: debt maturities, capitalized leases, disputed taxes in the contingency note, or deferred-tax swings? Write three sentences on what changed in your model of "what can kill a company." Then add one item to your personal red-flags list. It'll seed the Phase 2 checklist.

This module's flashcards and mastery quiz are wired into the app: see the node's Quiz and Reviews.