Equity, SBC, Consolidation & the Rest

Phase 1 · Accounting: The Language of Business · Week 13 · ~20 hours of module work

You've built the machine one piece at a time so far. You started with double-entry and the three statements and how they link. Then came revenue and working capital, then long-term assets, then the liability side with its debt and leases, its provisions and its deferred taxes. What's left is everything that happens once a company stops being a simple, single, domestic entity. It pays people in its own shares. It buys and controls other companies and operates in foreign currencies. It makes retirement promises. And it holds financial instruments that have to be marked somewhere. This week "a company's accounts" becomes "a group's accounts", and that's what you'll meet in every consolidated filing from now on. Next week's Phase 1 capstone sends you into a full 10-K and a full Indian annual report. Almost every note you'll wrestle with there is taught here.

Hour budget (suggested): SBC 4.5 h · consolidation & equity method 5 h · business combinations/PPA 3 h · FX 2 h · segments, pensions, instruments 3 h · practice set and mastery check 2.5 h.

Learning objectives

By the end of the week you can:

1. Account for stock-based compensation end-to-end: measure an RSU and an option grant at grant-date fair value, build the vesting/expense schedule (straight-line and graded), post the triple entry (P&L expense → cash-flow add-back → equity build-up), and record exercise/settlement.
2. Compute dilution from options and RSUs using the treasury-stock method and explain, with numbers, why "SBC is non-cash, so ignore it" is a half-truth that destroys analysis.
3. Consolidate a parent and a partly-owned subsidiary: combine line-by-line, eliminate the investment and intercompany transactions (including unrealized profit), and place non-controlling interest correctly on both the balance sheet and the income statement.
4. Apply the equity method to a 20–50% associate (roll the investment account forward, report the one-line share of profit) and state how joint arrangements are treated (joint operation vs joint venture).
5. Build a purchase price allocation: step identifiable assets up to fair value, recognize acquired intangibles and the deferred-tax effect, derive goodwill as the residual, and trace the post-deal income statement drag. State precisely how a bargain purchase is treated under US GAAP/IFRS versus the Ind AS 103 carve-out.
6. Handle foreign currency: distinguish functional from presentation currency, translation (current-rate) from remeasurement (temporal), and locate, and interpret, the CTA that sits in OCI.
7. Use the segment note as an analytical weapon: explain the management approach, the 10% and 75% tests, and extract the real profit engine of a multi-business company.
8. Read the remaining notes at awareness level: DB vs DC pensions and funded status (including India's gratuity), the three financial-instrument buckets (amortized cost / FVOCI / FVTPL), hedge-accounting basics, and the fair-value hierarchy, including why Level 3 deserves suspicion.

Prerequisites & connections

Builds on. M1.01–M1.05, because you need to be fluent in journal entries and the three-statement loop, and SBC is the classic "one event, three statements" drill. M1.02 (basic vs diluted EPS), where this week supplies the dilution machinery. M1.03 (the equity section: share capital, securities premium/APIC, reserves, AOCI, and the NCI line you were promised an explanation of). M1.07 (goodwill and impairment mechanics), since PPA is where goodwill is born. M1.08, for deferred taxes (fair-value step-ups create DTLs) and effective-interest debt (the amortized-cost bucket is the same math).

Feeds into. M1.10, the capstone filing read, where SBC, segment, business-combination, FX, pension and fair-value notes all appear in your hand-spread. M1.11 finishes the equity method, with the premium allocation, the investment's impairment test and associate-level eliminations. M1.12 takes share-based compensation and pensions to depth, including cash-settled awards and the pension roll-forward. M1.13 gives the full translation and remeasurement treatment, with the hyperinflation rules. M2.03 (ROIC), because goodwill and intangibles decisions change invested capital, and M2.04 (FCF), where the SBC add-back debate returns with force. M2.06–2.08, since many shenanigans hide in consolidation scope, Level 3 marks and one-time acquisition gains. M3.06 and M3.10, because EV must include NCI and sum-of-the-parts valuation is built from the segment note. Phase 5 sector playbooks, for bank/NBFC instrument books and conglomerate segment analysis. On the five-questions spine, the segment note answers half of Question 1 (where is the money actually made?). SBC and capital allocation feed Question 3 (aligned, honest people?), and reading consolidated against standalone in India is a Question 4 risk tool.

Core teaching content

4.1 The map: one module, two ideas

Everything here reduces to two ideas.

Idea 1: equity is a claims ledger, and not everything reaches it through the P&L. In the balance-sheet module you built the equity section and met a mysterious sibling of retained earnings: accumulated other comprehensive income (AOCI; in India, various "Other Equity" reserves). This week explains almost every item that lives there: currency translation differences, fair-value reserves on FVOCI instruments, cash-flow-hedge reserves, pension remeasurements; and, uniquely in India, bargain-purchase gains. Think of AOCI as the attic of equity: real changes in wealth that standard-setters decided are too volatile, too unrealized, or too non-operating to run through net income. Analysts who never climb into the attic get surprised by what falls out of it.

Idea 2: the statements you read are almost always a group's statements. "Infosys" in the consolidated accounts isn't one legal company. It's a parent plus dozens of subsidiaries stitched together by consolidation rules, with associates hanging off the side as one-line investments. Ask how the parent relates to each investee. If it controls it, you consolidate 100% line-by-line and carve out NCI. If it has significant influence, you use the equity method, on one line. If it merely holds the stake, that's a financial instrument in one of three buckets. The answer changes revenue, margins, debt and every ratio you'll ever compute, without changing the underlying economics at all. A large part of becoming an analyst is learning to see through these presentation boundaries.

4.6 Foreign currency: translation, remeasurement, and CTA

4.6.1 Two different questions, two different methods

Every multinational faces two distinct FX accounting questions (IAS 21 / Ind AS 21 / ASC 830). The aim here is to get you to the point where you can tell the two methods apart and read a CTA. The full translation and remeasurement treatment comes later in this phase, including the hyperinflation rules and a grid showing what each method does to every reported ratio.

Q1: What currency does each entity actually operate in? That's its functional currency, the currency of the primary economic environment. It's the one that mainly drives the entity's selling prices, costs and financing, and facts decide it, not choice. For Infosys' Indian parent it's INR. For its German subsidiary billing in euros with euro costs, it's EUR. For Jaguar Land Rover inside Tata Motors Passenger Vehicles (the JLR-owning company since the October 2025 demerger), it's GBP.

Q2: What currency are the group statements presented in? That's the presentation currency. Tata Motors Passenger Vehicles presents in INR, and nothing stops a Dutch company presenting in USD.

Two mechanisms bridge the gaps:

Separate from both are ordinary transaction gains and losses. A receivable, payable or loan in a foreign currency is remeasured at each closing through P&L. That's the FX line you see in Indian companies' "other income/finance costs". An Indian borrower with USD debt takes a P&L hit every year the rupee slides. Bharti Airtel's USD borrowings produced years of such losses, so check its notes.

4.6.2 CTA: the number in the attic

Translate a healthy foreign subsidiary year after year and the balancing figure, the CTA, piles up in OCI/equity. Work through it line by line. Here are the quick mechanics, with the full version in Worked Example 4. Take an Indian parent's US subsidiary with opening net assets of $100m and profit of $20m, while ₹/$ moves from 83 (open) to 85 (average) to 87 (close). For the year, CTA = 100×(87−83) + 20×(87−85) = ₹440m. In rupee terms that's a ₹44 cr gain [(₹400m + ₹40m) = ₹440m], and it goes to OCI, not profit. Here's the intuition. The group's dollar net assets became worth more rupees just because the rupee weakened. That's a real change in wealth, but it isn't operating performance, so it goes in the attic. It also recycles. When the operation is disposed of, the accumulated CTA is reclassified into P&L. That occasionally produces bizarre-looking gains or losses on exits, as the P&L catches up with a decade of currency drift in one line.

This feeds straight into your analysis. (1) An Indian IT company's INR revenue growth = volume + pricing + currency. Always find constant-currency growth in the MD&A or earnings release before you credit the business. TCS and Infosys disclose it every quarter, as do US multinationals like Coca-Cola and Apple when they're fighting strong-dollar years. (2) A persistently negative CTA on a US company's books usually just means it operates in depreciating-currency countries. Every US multinational with Indian operations accumulates negative CTA on INR net assets. (3) Equity has absorbed what the P&L never showed, so book value growth at multinationals is partly a currency artifact.

4.7 Segment reporting: the analyst's best friend

Consolidation solves one problem and creates another, because it fuses a dozen businesses into one set of totals. Segment reporting (IFRS 8 / Ind AS 108 / ASC 280) pulls them apart again. It does so imperfectly, but it's the only place the company itself tells you where the money is made.

The rules follow the management approach. Operating segments are the components whose results the chief operating decision maker (CODM), the CEO or board, regularly reviews to allocate resources, so you see the business the way management slices it. A segment is separately reportable if it crosses any 10% test: 10% of combined revenue (including intersegment), 10% of absolute profit/loss, or 10% of assets. Reported segments must also cover at least 75% of external revenue, so you keep adding segments until they do. The disclosures cover segment revenue (external and intersegment), a measure of profit, usually assets/capex/depreciation, and a reconciliation to consolidated totals. On top of that come entity-wide disclosures: revenue by geography and any single customer ≥10% of revenue. That last one gives you a free detector for customer-concentration red flags.

From Phase 8 on, this note is the first stop of every teardown you'll run. Here's why:

• Find the real engine. Amazon's consolidated operating margin is unremarkable. The segment note shows AWS producing roughly 60% of operating income from under a fifth of revenue (illustrative, ~FY24 scale, so pull the 10-K). ITC's cigarettes segment supplies the large majority of segment profit, while FMCG-Others runs thin margins on big revenue. Reliance's note splits O2C, Jio and Retail, three different businesses that deserve three different multiples.
• Catch cross-subsidies and hidden losers. A fat consolidated margin can hide one segment burning cash. Set segment capex against segment profit and you see where management is actually putting capital, which is Question 3 of the five.
• Seed the sum-of-the-parts valuation, covered later in Phase 3. Each segment gets its own multiple, and the conglomerate-discount debate starts here.
• Watch the games: aggregation of dissimilar businesses into "Others", redefined segments that break your time series the year a division sours (restate history yourself), and profit measures that allocate no corporate costs.

4.10 The OCI decoder: pulling the module together

You've now met every major tenant of the equity attic. Learn this table well. It turns the statement of comprehensive income from noise into information:

Build two habits now. First, once per company, read the statement of changes in equity and the OCI note line by line. It takes five minutes and tells you about currencies, hedges, pensions and instrument books all at once. Second, when someone quotes "comprehensive income" against "net income", you can now name what's in the gap and whether it'll ever visit the P&L.

Common mistakes & how experts think differently

1. "SBC is non-cash, so FCF should add it back, full stop." The novice sees the CFO add-back and stops. The expert sees a compensation cost routed around the cash flow statement through dilution, and either deducts SBC from FCF or models the rising share count. Never neither, and never both. You can spot a careless analyst by a DCF with SBC added back and a frozen share count. That model assumes employees will eventually work for free.

2. Valuing the whole but owning a part: NCI blindness. Divide the parent's market cap by total consolidated profit, instead of profit attributable to owners, and you flatter P/E for any company with big minorities. Forget to add NCI when building EV and you double-flatter EV/EBITDA, counting 100% of the sub's EBITDA and none of the minority's claim. Experts have a reflex. If the NCI line isn't 0, they find which subsidiary, what stake, and whether the crown jewel is shared. Hindustan Zinc inside Vedanta is the canonical Indian example.

3. Reading Indian standalone statements because they came up first. Consolidated is the economic entity. Standalone is a legal entity, and it can be dressed up by selling to, or borrowing through, unlisted subsidiaries. Experts read consolidated, then diff it against standalone. A parent reporting fat standalone profits next to a starving consolidated picture is a red flag with a long Indian pedigree.

4. Treating equity-method income as ordinary earnings. It arrives without revenue, without cash (only dividends are cash), and without showing the associate's leverage. Experts do a look-through on any material associate: real share of revenue, real share of debt, cash actually received vs profit booked. A company "earning" ₹500 cr from associates that remit ₹50 cr of dividends has ₹450 cr of paper accrual in its P&L.

5. Reading goodwill as "brand value," or panicking at amortization. Goodwill is a residual: synergies + assembled workforce + overpayment, unlabeled. Experts read the PPA table to see what was separately valued (customer lists, technology, brands) and treat goodwill as a fossil record of the price paid. A goodwill-heavy balance sheet means past M&A, and an impairment is a lagging confession that tells you nothing new. Going the other way, amortization of acquired intangibles is a charge organic peers don't carry. Compare acquisitive and organic companies on EBITA or adjust explicitly, and keep your eyes open about serial acquirers.

6. Confusing the three FX effects. Transaction losses go to P&L and are real and cash-relevant (e.g., USD debt in an INR company). Translation CTA goes to OCI as an accounting restatement of foreign net assets. Constant-currency growth is an MD&A disclosure. They're three different animals. The novice sees "FX loss" and treats them all alike. The expert asks which mechanism and does cash care. A CTA swing changes no cash flow today, while a USD borrowing rolled at ₹90 changes everything.

7. Skipping the segment note because the consolidated numbers "look fine." Consolidated margins are an average that hides the engine and the anchor. Experts open the segment note first, and it's step one of the Phase 8 teardown. They check profit share vs revenue share by segment, segment capex vs segment profit, the customer-concentration disclosure, and whether segment definitions hold steady across years. Companies love to redraw the map the year a territory sours.

8. Taking every "fair value" as equally fair. Level 1 is a price. Level 3 is an opinion with a discount-rate assumption. Experts scale their trust by level, size Level 3 against tangible equity, and read the Level 3 roll-forward for gains manufactured by remeasurement. Bring the same instinct to bargain-purchase gains. They're one-time, so strip them out (under Ind AS they're invisible in P&L, so check capital reserve movements). Bring it, too, to "gain on obtaining control" step-ups when an associate becomes a subsidiary.

Worked examples

All the companies below are synthetic, with deliberately clean numbers. Their mechanics are the ones you'll meet in real filings.

Worked Example 1: An RSU grant, start to finish (with forfeiture true-up)

Nivara Ltd grants 60,000 RSUs on 1 April 2025, when the share price is ₹500, for a grant-date fair value of ₹3.0 cr in total. They cliff-vest after 3 years. Management estimates 90% of units will survive attrition.

Step 1: expected cost. 60,000 × 90% = 54,000 units are expected to vest, so the expected total expense is 54,000 × ₹500 = ₹2.70 cr, spread over 3 years.

Step 2: FY26 expense. ₹2.70 cr ÷ 3 = ₹0.90 cr. Book it as Dr Employee stock compensation expense 0.90 · Cr Share options outstanding account 0.90, and in the CFS that +0.90 is added back in CFO.

Step 3: FY27, estimate revised. Attrition is running hotter, and now only 48,000 units are expected to vest. Cumulative expense required by end-FY27 = 48,000 × 500 × (2/3) = ₹1.60 cr. You've already booked ₹0.90 cr, so FY27 expense = ₹0.70 cr. That's a catch-down. You take the true-up through the current year, with no restatement.

Step 4: FY28, actual vesting. 47,000 units actually vest. Total expense must equal actuals, which is 47,000 × 500 = ₹2.35 cr. With ₹1.60 cr booked so far, FY28 expense = ₹0.75 cr. Check it: 0.90 + 0.70 + 0.75 = 2.35.

Step 5: the point. On vesting day the share price is ₹800. The value actually delivered to employees is 47,000 × 800 = ₹3.76 cr, against ₹2.35 cr ever expensed. That ₹1.41 cr gap never touches any P&L. Shareholders pay it through dilution. Until vesting, the ~47,000 pending shares were already inflating the diluted share count nearly one-for-one, since RSUs carry no strike. So here's the second of the half-truth's four truths, in numbers. Grant-date accounting understates the transfer whenever the stock has risen.

Worked Example 2: Consolidation with unrealized intercompany profit (Meridian–Kaveri, year 2)

This picks up the 80%-control example from earlier. In FY27 Kaveri's revenue is ₹340 cr, of which ₹100 cr is sold upstream to Meridian at a 25% margin on the transfer price (cost to Kaveri ₹75 cr). By 31 March 2027 Meridian has resold only 60%, and goods with a transfer value of ₹40 cr are still in its closing inventory. Kaveri's PAT is ₹70 cr, with no dividend. Meridian's revenue is ₹880 cr and its PAT ₹120 cr, with no dividend income this year. The intercompany trade balance outstanding is ₹15 cr. Meridian owns 80%, and opening NCI is ₹58 cr.

Step 1: unrealized profit. Profit embedded in unsold goods = 25% × ₹40 cr = ₹10 cr. Nobody in the group has earned this. It moved goods from one pocket to another and marked them up.

Step 2: consolidated income statement. All figures are in ₹ cr:

The sale went upstream, from subsidiary to parent, so the eliminated profit belongs to Kaveri's income and NCI bears its 20% share. NCI gets 20% of the adjusted ₹60 cr instead of the reported ₹70 cr. Now the downstream contrast. Had Meridian been the seller, NCI = 20% × 70 = ₹14 cr, and the whole ₹10 cr elimination would fall on the owners, taking their share from 168 to 120 − 10 + 56 = 166.

Step 3: balance sheet effects. Consolidated inventory is ₹10 cr lower, carried at group cost of ₹30 cr instead of the ₹40 cr transfer value. Receivables and payables each fall by ₹15 cr. NCI closing = 58 + 12 = ₹70 cr. One refinement, for awareness only. Tax really was paid on the intra-group sale, so a deferred tax asset of 25% × 10 = ₹2.5 cr arises on consolidation. Next year, when the goods finally sell to outsiders, the ₹10 cr is "released" into consolidated profit. Eliminations defer profit. They don't destroy it.

Worked Example 3: Partial vs full goodwill, then a bargain purchase three ways

Zenith Ltd acquires 75% of Orchid Ltd. Orchid's identifiable net assets have a fair value of ₹800 cr, and the quoted price implies the 25% minority stake is worth ₹210 cr.

Case A: price ₹660 cr.

• Partial goodwill (IFRS 3 / Ind AS 103 option): NCI = 25% × 800 = ₹200 cr, so goodwill = 660 + 200 − 800 = ₹60 cr.
• Full goodwill (the other IFRS/Ind AS option; mandatory under US GAAP): NCI at fair value = ₹210 cr, so goodwill = 660 + 210 − 800 = ₹70 cr. Goodwill and NCI are each ₹10 cr fatter, because the minority's share of goodwill is now on the books. It's the same deal with two balance sheets. Remember that before you compare goodwill/assets ratios across US and Indian companies.

Case B: distress sale at ₹560 cr (NCI proportionate, ₹200 cr). Consideration + NCI = 760, which is below 800, leaving an excess of ₹40 cr. Here's how it's routed:

• IFRS 3 / ASC 805: reassess identification and measurement first, since most "bargains" are mismeasurement. If the excess survives, it's a gain of ₹40 cr in profit or loss on the acquisition date, attributed to the acquirer. EPS jumps, and analysts strip it out as one-time.
• Ind AS 103: never P&L. With clear evidence of why this was a bargain (documented distress), recognize the ₹40 cr in OCI on the acquisition date and accumulate it in equity as capital reserve. Without clear evidence, credit it directly to capital reserve, not even through OCI. Either way Indian EPS shows nothing. The only trace is a capital-reserve movement in the statement of changes in equity. Two identical deals in Mumbai and New York produce a ₹40 cr earnings gap purely because of the carve-out.

Worked Example 4: Translating a foreign subsidiary and deriving CTA

Deepak Industries, an Indian parent that presents in ₹, owns DI Americas Inc. DI Americas earns, spends and borrows in dollars, so its functional currency is USD. In FY26 it starts with net assets of $100m, makes a profit of $20m and pays no dividends. Exchange rates: 1 April ₹83/$ · average ₹85/$ · 31 March ₹87/$.

Step 1: translate the income statement at the average rate. $20m × 85 = ₹170 cr, and this flows into consolidated profit.

Step 2: translate closing net assets at the closing rate. $120m × 87 = ₹1,044 cr.

Step 3: what equity would be without CTA. Take opening net assets as first translated ($100m × 83 = ₹830 cr) and add translated profit of ₹170 cr. That gives ₹1,000 cr.

Step 4: CTA = the plug. 1,044 − 1,000 = +₹44 cr, credited to OCI (foreign currency translation reserve). Always break it down. Repricing opening net assets gives $100m × (87 − 83) = ₹40 cr. Repricing profit from average to closing gives $20m × (87 − 85) = ₹4 cr. Together that's ₹44 cr, matching the plug. The rupee weakened, so dollar net assets are worth more rupees. That's a gain, but it goes to the attic, not the P&L.

Contrast: remeasurement. Suppose DI Americas were a mere warehousing arm whose functional currency is really INR. Then there's no translation. Its USD monetary items are remeasured through P&L. Say net monetary assets were $30m. The remeasurement gain is 30 × (87 − 83) = ₹12 cr, straight into profit, with non-monetary assets frozen at historical rates. Same subsidiary, same rates. Earnings volatility appears or disappears on the functional-currency judgment alone. Now you know why the notes state that judgment explicitly. You also know why, on disposal of DI Americas, the accumulated CTA (years of ₹44-crore-type entries) would suddenly recycle through the P&L.

Worked Example 5: Diluted EPS when options and RSUs are both outstanding (US $)

Helioscore Inc (synthetic), a US software firm, reports net income attributable to common shareholders of $540m and a basic weighted-average count of 300m shares. Three equity-award pools are live at year-end. There are 20m employee options at a $40 strike, in the money against an average market price for the year of $80. There are 5m options from an older grant at a $100 strike, above the average price. And there are 8m unvested RSUs, which have no strike.

Step 1: basic EPS. 540 ÷ 300 = $1.80. That's what a lazy screen quotes, but per-share value work has to use the diluted figure.

Step 2: the in-the-money options (treasury-stock method). Assume exercise. Proceeds = 20m × $40 = $800m, which buy back shares at the $80 average, so 800 ÷ 80 = 10m are repurchased. Incremental = 20 − 10 = 10m, or directly 20 × (1 − 40/80) = 10.

Step 3: the out-of-the-money options. Strike $100 ≥ average $80. These options are antidilutive and get excluded entirely. Counting them would lower the share count and raise EPS, which the standard forbids you to show. TSM can only ever add shares.

Step 4: the RSUs. Unvested RSUs carry no strike, so assumed proceeds are ~nil and they dilute almost one-for-one, adding 8m shares. The filed diluted count also treats unrecognized compensation cost as assumed proceeds under ASC 260/IAS 33. It's a small refinement, so use the filing's number in practice.

Step 5: diluted EPS. Diluted shares = 300 + 10 + 8 = 318m, and 540 ÷ 318 = $1.70.

The point. The roughly 6% haircut from basic $1.80 to diluted $1.70 is your ownership being shared with employees, and it's only the vested/in-the-money slice. Count every live award, (20 + 5 + 8) = 33m against 300m shares, and the overhang is ~11%. That's well past the 2% "background noise" line and into "real claim on future profits" territory, which no single year's SBC expense reveals. Always value on the diluted count, and read the overhang in the SBC note.

Worked Example 6: From associate to subsidiary: the "gain on obtaining control" (India ₹)

Vaani Ltd (synthetic) already owns 30% of Tejas Ltd (synthetic), carried by the equity method at ₹300 cr. On 1 October 2025 Vaani buys a further 45% for ₹540 cr cash, lifting its holding to 75%. It now controls Tejas, so the one-line associate flips to full consolidation. That's the equity-method warning from earlier, in numbers. On that date the fair value of Tejas's identifiable net assets is ₹1,000 cr, and NCI is measured at its proportionate share.

Step 1: remeasure the stake you already held. Obtaining control is accounted for as if you sold the old 30% and bought it back at fair value. The price paid for 45% (₹540 cr) implies a 100% value of 540 ÷ 0.45 = ₹1,200 cr, so the previously held 30% is worth 0.30 × 1,200 = ₹360 cr. Remeasuring it up from its ₹300 cr carrying value books a gain on obtaining control of ₹60 cr through P&L. It's a one-time, non-operating item.

Step 2: non-controlling interest. NCI = 25% × ₹1,000 cr = ₹250 cr.

Step 3: goodwill. Every term of the goodwill formula from the business-combinations section is live now. Goodwill = consideration ₹540 cr + fair value of the previously held stake ₹360 cr + NCI ₹250 cr − FVINA ₹1,000 cr = ₹150 cr.

The point. There are two traps here. (i) The ₹60 cr "gain on obtaining control" is an artifact of the remeasurement. No cash was earned, so strip it from earnings power the same way you would a bargain-purchase gain. (ii) From 1 October, Tejas's entire revenue, cost and debt consolidate line-by-line, where before only a single equity-income line appeared. Year-over-year growth looks explosive for a purely accounting reason. Flag it as a comparability break, not performance.

Practice set

Work each problem fully before opening its solution (working agreement #3). Problems 1–4 are guided, 5–9 independent, 10–12 timed.

P1 (guided: RSU schedule). On 1 April 2025, Suvarna Tech grants 2,00,000 RSUs when its share trades at ₹450. Vesting: 25% per year over 4 years; the company uses straight-line attribution and expects negligible forfeitures. (a) Total grant-date fair value and annual expense. (b) The journal entry each year and where each statement feels it. (c) The share price doubles to ₹900 during year 2: what happens to the expense? (d) Where does an analyst see the cash reality?

Solution. (a) 2,00,000 × ₹450 = ₹9.0 cr, which is ₹2.25 cr a year for four years. (b) Dr Employee stock compensation expense 2.25 cr · Cr Share options outstanding (equity) 2.25 cr. Profit falls by 2.25 in the P&L, the same 2.25 is added back inside CFO as non-cash, and the equity reserve on the balance sheet builds by 2.25/yr. That's the triple entry. (c) Nothing. Grant-date fair value is frozen, and equity-settled SBC is never remeasured for price moves. The economic transfer at vesting has doubled, and the P&L doesn't see it. (d) You see it in dilution, since the diluted share count includes unvested RSUs nearly 1:1. You also see it in any cash the company later spends buying back shares to offset issuance (CFF), plus tax-withholding cash paid on net settlement.

P2 (guided: graded vesting). Vikram Auto grants 90,000 options (grant-date fair value ₹200 each) vesting in three equal annual tranches of 30,000. Compute the year-by-year expense under (a) Ind AS 102 / IFRS 2 and (b) US GAAP with the straight-line election. Comment.

Solution. Total = 90,000 × 200 = ₹1.8 cr. (a) Under graded attribution each tranche is its own award. Year 1 takes all of tranche 1 (30,000×200 = ₹60 lakh), half of tranche 2 (₹30 lakh) and a third of tranche 3 (₹20 lakh), for ₹1.10 cr. Year 2: 30 + 20 = ₹0.50 cr. Year 3: ₹0.20 cr. (b) Straight-line gives ₹0.60 cr each year. The total is the same ₹1.8 cr, but Ind AS front-loads ~61% into year 1. So an Indian company's margin dip in a big grant year, set against a US straight-liner, is an accounting artifact. Normalize before you compare.

P3 (guided: treasury-stock method). Meghdoot Ltd: net income ₹840 cr; weighted basic shares 200 cr; 12 cr employee options outstanding, strike ₹70; average market price for the year ₹120. (a) Incremental shares and diluted EPS. (b) Rework if the average price were ₹60.

Solution. (a) Assumed proceeds = 12 × 70 = ₹840 cr, which buy back 7 cr shares at ₹120, so incremental = 12 − 7 = 5 cr (or 12 × (1 − 70/120) = 5). Diluted shares are 205 cr. Basic EPS = 840/200 = ₹4.20, and diluted = 840/205 = ₹4.10. (b) The ₹70 strike is above the ₹60 average, so the options are out of the money. They're antidilutive and excluded, and diluted EPS = basic = ₹4.20. TSM can only ever add shares. Exercising below-market options would "reduce" the count, which the standard forbids you to show.

P4 (guided: SBC and free cash flow). Kalpana Software reports: CFO ₹1,150 cr (which includes a ₹240 cr SBC add-back), capex ₹350 cr, and a ₹260 cr buyback described in the call as "primarily offsetting employee-plan dilution." (a) Reported FCF. (b) FCF treating SBC as a cash cost. (c) The other consistent treatment. (d) Why "(b) plus a growing share count" would be wrong.

Solution. (a) 1,150 − 350 = ₹800 cr. (b) 800 − 240 = ₹560 cr. You treat SBC as if paid in cash (Damodaran's default), then value on the current diluted count. Notice the cross-check. FCF after the anti-dilution buyback = 800 − 260 = ₹540 cr, which comes close to (b). The buyback is the dilution cost turning into cash in front of you. (c) Keep FCF at ₹800 cr but explicitly grow the share count for future issuance, and value outstanding awards separately. (d) Deducting SBC and growing the count charges the same cost twice. It's the mirror image of the classic sin of doing neither.

P5 (independent: NCI and goodwill both ways). Prisha Ltd pays ₹560 cr for 70% of Tarang Ltd. Fair value of identifiable net assets ₹600 cr; fair value of the 30% NCI ₹230 cr. Compute goodwill and NCI under (a) the proportionate method and (b) the fair-value method. (c) Which does US GAAP permit?

Solution. (a) NCI = 30% × 600 = ₹180 cr, and goodwill = 560 + 180 − 600 = ₹140 cr. (b) NCI = ₹230 cr, and goodwill = 560 + 230 − 600 = ₹190 cr. The extra ₹50 cr is the minority's share of goodwill, now grossed up on both sides. (c) US GAAP mandates (b), full goodwill. IFRS/Ind AS allow either, deal by deal. Take care whenever you screen goodwill intensity across regimes.

P6 (independent: consolidated P&L with attribution). Parent P owns 80% of its subsidiary, S. P has revenue of ₹1,200 cr and PAT of ₹150 cr, which includes ₹8 cr of dividends received from S during the year. S has revenue of ₹400 cr and PAT of ₹50 cr, and pays total dividends of ₹10 cr. S also sold P goods for ₹60 cr (upstream) at a 20% margin on transfer price, and goods with transfer value ₹15 cr remain unsold in P's inventory. Compute consolidated revenue, consolidated PAT, NCI's share and owners' share.

Solution. Unrealized profit = 20% × 15 = ₹3 cr. Revenue = 1,200 + 400 − 60 = ₹1,540 cr. For PAT, eliminate the dividend from S (check: 80% × 10 = 8), then add S's profit less the unrealized part. That's 150 − 8 + 50 − 3 = ₹189 cr. The sale was upstream, so NCI shares the elimination: NCI = 20% × (50 − 3) = ₹9.4 cr. Owners get 189 − 9.4 = ₹179.6 cr. Had the sale been downstream, NCI = 20% × 50 = ₹10 cr and owners ₹179 cr. Only the attribution moves. Total profit stays the same.

P7 (independent: equity method). Ravindra Ltd buys 25% of an associate for ₹500 cr. Its share of book net assets is 25% of the ₹1,600 cr = ₹400 cr, and the ₹100 cr premium is embedded goodwill that isn't booked separately. During the year the associate earns PAT ₹240 cr, pays total dividends ₹80 cr, and carries ₹900 cr of debt. Give (a) the P&L line, (b) cash received, (c) the closing investment balance, (d) the debt shown on Ravindra's balance sheet from the associate, and (e) the one-sentence quality-of-earnings comment.

Solution. (a) Share of profit of associates = 25% × 240 = ₹60 cr (one line, after tax). (b) 25% × 80 = ₹20 cr. (c) 500 + 60 − 20 = ₹540 cr. (d) Nil. The equity method shows no associate debt. The look-through exposure is ≈ 25% × 900 = ₹225 cr, and a debt ratio computed off the balance sheet will never see it. (e) ₹60 cr of earnings is backed by ₹20 cr of cash, so ₹40 cr is accrual whose value depends entirely on the associate's reinvestment quality.

P8 (independent: PPA build). Nalanda Consumer pays ₹1,250 cr for 100% of a target with book net assets ₹500 cr. The PPA finds PP&E undervalued by ₹150 cr (remaining life 10 years), developed technology worth ₹180 cr (6-year life) and an inventory step-up of ₹70 cr (sells within a year). Tax rate 25%. Compute (a) goodwill and (b) the year-1 pre-tax earnings drag versus the target's old books, and (c) state goodwill's ongoing P&L treatment.

Solution. (a) Step-ups = 150 + 180 + 70 = 400. DTL = 25% × 400 = ₹100 cr. Fair value of identifiable net assets = 500 + 400 − 100 = ₹800 cr, so goodwill = 1,250 − 800 = ₹450 cr. (b) Add the inventory flush of ₹70 cr (one-time, in COGS), technology amortization of 180/6 = ₹30 cr and extra depreciation of 150/10 = ₹15 cr. That's a ₹115 cr pre-tax drag, softened by a ~₹28.75 cr deferred-tax credit as the DTL unwinds. (c) Goodwill isn't amortized under US GAAP/IFRS/Ind AS. It gets an annual impairment test only, and any later impairment is a lagging admission about the price paid.

P9 (independent: bargain purchase, three ways). Acquirer pays ₹800 cr; fair value of identifiable net assets acquired is ₹950 cr. State the treatment of the ₹150 cr excess under US GAAP, IFRS, and Ind AS, and the analyst handling in each case.

Solution. Every framework says reassess first. Recheck that assets and liabilities are complete and correctly measured, because most bargains are measurement errors. If the excess is confirmed, US GAAP and IFRS recognize a ₹150 cr bargain-purchase gain in P&L at the acquisition date (First Citizens–SVB style), and the analyst strips it from earnings power as non-recurring. The Ind AS 103 carve-out never uses P&L. With clear evidence of the bargain's cause, the excess goes to OCI and accumulates as capital reserve. Without clear evidence, credit it directly to capital reserve. Indian EPS and screeners show nothing, so scan the statement of changes in equity for capital-reserve additions to catch it.

P10 (timed, 10 min: FX translation). An Indian parent's US subsidiary (functional USD): opening net assets $50m, profit $10m, no dividends. Rates: opening ₹82/$, average ₹84/$, closing ₹86/$. Compute the year's CTA, state where it appears, and when it would ever hit the P&L.

Solution. CTA = 50 × (86 − 82) + 10 × (86 − 84) = ₹200m + ₹20m = ₹220m = ₹22 cr gain. To check it, closing net assets are $60m × 86 = ₹5,160m, against opening ₹4,100m + profit ₹840m = ₹4,940m. That difference of ₹220m equals the CTA. It appears in OCI, accumulating in the foreign-currency translation reserve within equity. It touches the P&L only if the subsidiary is disposed of, when the accumulated balance recycles.

P11 (timed, 10 min: segment teardown). A conglomerate reports (₹ cr, all revenue external): Segment A revenue 6,200, EBIT 620 · Segment B revenue 2,800, EBIT 700 · Segment C revenue 800, EBIT 90 · unallocated corporate costs (110) · consolidated EBIT 1,300. (a) Find the segment margins and the profit engine. (b) Decide whether C must be separately reported under the 10% tests (assets unknown). (c) Check whether A+B alone satisfy the 75% rule. (d) If C instead lost ₹150 cr, does the answer to (b) change?

Solution. (a) Margins are A 10%, B 25% and C 11.25%. The segment EBIT pool is 1,410, and B contributes ~50% of segment profit on ~29% of revenue. B is the engine. A is a low-margin volume business diluting the optics. (b) Revenue test: 800/9,800 = 8.2% < 10%. Profit test: threshold = 10% × 1,410 = ₹141 cr > 90. C fails both disclosed tests (assets unknown), so it's not required separately. (c) A+B external revenue = 9,000/9,800 = 91.8% ≥ 75%. C may be folded into "all other segments", and you lose a clean read on it. (d) Yes. The profit test uses the greater of combined profits (620+700 = 1,320) or combined losses in absolute terms, so the threshold is ₹132 cr. Since |−150| > 132, C becomes reportable. Losses can force a segment out of hiding.

P12 (timed, 8 min: classification lightning round). Assign each holding to its measurement bucket (amortized cost / FVOCI / FVTPL), noting any recycling quirk, and give its likely fair-value level. (1) ₹200 cr of 10-year G-Secs in a hold-to-collect book. (2) A 3% strategic stake in a listed supplier with the FVOCI election. (3) A NIFTY-futures trading book. (4) An interest-rate swap designated as a cash-flow hedge of floating-rate debt. (5) Unlisted startup equity, no election. (6) AAA corporate bonds held to collect and sell. (7) Trade receivables. (8) Convertible debentures held as an investment.

Solution. (1) Amortized cost (hold-to-collect + SPPI), and the disclosed FV is Level 1. (2) FVOCI-equity. Dividends go to P&L and fair-value moves to OCI, never recycled. Level 1. (3) FVTPL, Level 1. (4) Fair value on the BS, with the effective portion of changes going to OCI (cash-flow hedge reserve) and recycled when the hedged interest hits P&L. It's Level 2, valued off observable curves. (5) FVTPL, the default for equities (FVOCI only by election). It's Level 3, management's model, so apply the fair-value-hierarchy suspicion from earlier. (6) FVOCI-debt (hold-to-collect-and-sell + SPPI). FV moves go to OCI and are recycled on sale. Level 1 or 2, depending on quote depth. (7) Amortized cost, with ECL provisioning, as in the revenue, receivables and inventory work. (8) The conversion feature fails SPPI, so the entire instrument is FVTPL, typically Level 2–3.

P13 (guided: defined-benefit gratuity, funded status and P&L geography). Ananya Manufacturing (synthetic) runs a statutory gratuity plan under Ind AS 19. At year-end the defined-benefit obligation is ₹800 cr and plan assets at fair value are ₹680 cr. During the year current service cost is ₹55 cr and the discount rate is 7%. An actuarial loss of ₹40 cr arose from lowering the discount-rate assumption. (a) What appears on the balance sheet, and at what amount? (b) Place each cost piece: current service cost, net interest, remeasurement. (c) Does the actuarial loss ever reach the P&L? (d) Give the one-line analyst treatment.

Solution. (a) Funded status = 680 − 800 = a net defined-benefit liability of ₹120 cr, shown net on the face of the balance sheet. (b) Current service cost of ₹55 cr goes to operating expense. Net interest on the net liability = 7% × 120 = ₹8.4 cr, which goes to finance cost. The ₹40 cr actuarial loss goes to OCI. (c) No. Under Ind AS 19/IAS 19, remeasurements never recycle to P&L. US GAAP amortizes them back over time through the corridor. (d) Treat the ₹120 cr deficit as debt-like and add it to EV, since it ranks ahead of shareholders. Then sanity-check the discount-rate and salary-growth assumptions before trusting the number.

P14 (guided: PPA and full goodwill, a US deal). Beacon Systems Inc (synthetic, reporting under US GAAP) buys 80% of a target for $1,200m. The target's book net assets are $600m. The PPA finds PP&E undervalued by $120m (8-year life), developed technology worth $200m (5-year life), and an inventory step-up of $80m (sold within the year). The 20% NCI has a fair value of $260m, and the tax rate is 21%. (a) Deferred tax liability and fair value of identifiable net assets. (b) Goodwill. (c) Year-1 pre-tax earnings drag versus the target's old books, and the offsetting item. (d) How would Ind AS/IFRS change the goodwill on this same deal?

Solution. (a) DTL = 21% × (120 + 200 + 80) = 21% × 400 = $84m, and FVINA = 600 + 400 − 84 = $916m. (b) US GAAP mandates full goodwill, so NCI enters at its $260m fair value. Goodwill = 1,200 + 260 − 916 = $544m. (c) Drag = inventory flush $80m + technology amortization 200/5 = $40m + extra depreciation 120/8 = $15m = $135m pre-tax, softened by a ~$28.35m deferred-tax credit as the DTL unwinds (21% × 135). (d) Under Ind AS/IFRS the acquirer may instead elect NCI at proportionate share (20% × 916 = $183.2m), which gives goodwill of only $467.2m. Same deal, ~$77m less goodwill and less NCI. Take care whenever you screen goodwill intensity across regimes.

P15 (independent: remeasurement, the temporal method). Sagarika Exports (synthetic), an Indian parent presenting in ₹, runs a small US procurement office whose functional currency is judged to be INR. It's an extension of the parent, not self-contained. At year-end the office holds net monetary assets of $25m and non-monetary assets (equipment, prepaid rent) of $10m booked when the rate was ₹80/$. The closing rate is ₹85/$. (a) The remeasurement gain or loss and where it goes. (b) How the non-monetary assets are carried. (c) If the functional currency had instead been USD, where would the difference have gone?

Solution. (a) Only monetary items remeasure. That gives a gain of 25 × (85 − 80) = ₹125m = ₹12.5 cr, straight into P&L. (b) Non-monetary assets stay at the historical rate, so $10m × 80 = ₹80 cr, untouched by the currency move. (c) If USD were functional, this would be a self-contained foreign operation under the translation (current-rate) method. The balancing difference would go to OCI as CTA, not to profit. Same office, same rates. The functional-currency judgment alone decides whether earnings wobble.

P16 (independent: three buckets, three recycling fates). Trilok Capital (synthetic) buys a ₹200 cr investment. Over the year its fair value rises ₹30 cr and it receives ₹6 cr of dividends/coupons. State the P&L impact, the OCI impact, and what happens to the accumulated gain on eventual sale, if the holding is classified as (a) an equity investment under the FVOCI election; (b) a debt security at FVOCI; (c) FVTPL.

Solution. (a) FVOCI-equity (election): the ₹6 cr of dividends hit P&L and the ₹30 cr fair-value gain goes to OCI. On sale it never recycles. The accumulated gain simply moves within equity to retained earnings. (b) FVOCI-debt: interest hits P&L, the ₹30 cr fair-value gain stays in OCI while held, and on sale it recycles into P&L. (c) FVTPL: the entire ₹30 cr fair-value change hits P&L this period, and so does the ₹6 cr of income. You get three different reported-earnings paths from identical economics. Classification defers recognition. It doesn't change the cash.

P17 (independent: NCI and EV hygiene). Helios Corp (synthetic, US) reports consolidated PAT of $900m, of which $150m is attributable to NCI (from a 70%-owned subsidiary). Consolidated EBITDA is $3,000m. Market cap is $30,000m, net debt $4,000m, and the fair value of the NCI is $2,500m. (a) The P/E an analyst should quote, versus the flattered version. (b) Enterprise value and EV/EBITDA done correctly. (c) The EV/EBITDA obtained by forgetting NCI, and why it is wrong.

Solution. (a) Market cap is a claim only on owners' profit, which is 900 − 150 = $750m. So P/E = 30,000 / 750 = 40×. Dividing by total PAT gives 30,000 / 900 = 33.3×, the flattering error that treats the minority's profit as yours. (b) Consolidated EBITDA already includes 100% of the subsidiary, so EV must add the minority's claim. EV = 30,000 + 4,000 + 2,500 = $36,500m, and EV/EBITDA = 12.17×. (c) Omitting NCI gives EV of $34,000m and a multiple of 11.33×. You've counted 100% of the subsidiary's EBITDA in the denominator but none of the minority's claim in the numerator, so the multiple is double-flattered.

P18 (timed/stretch: 12 minutes, six parts; mixed currencies). (i) Define non-controlling interest in one line. (ii) 15m options, strike $60, average price $100: find the incremental shares under the treasury-stock method. (iii) Consideration ₹700 cr + NCI ₹180 cr, fair value of identifiable net assets ₹760 cr: is it goodwill or a bargain, and how much? (iv) DBO ₹500 cr, plan assets ₹430 cr: state what shows on the balance sheet. (v) A US subsidiary (functional USD) has opening net assets of $60m, profit of $10m and no dividends, with rates of ₹80 open / ₹83 average / ₹85 close. Compute the year's CTA. (vi) Name the OCI reserve that recycles to P&L on disposal of a foreign operation.

Solution. (i) The equity in a consolidated subsidiary belonging to shareholders other than the parent. It's a component of group equity, not a liability. (ii) 15 × (1 − 60/100) = 6m shares. (iii) 700 + 180 = 880 > 760, so it's goodwill of ₹120 cr, a positive residual and not a bargain. (iv) A net defined-benefit liability of ₹70 cr (430 − 500), shown net. (v) CTA = 60 × (85 − 80) + 10 × (85 − 83) = 300 + 20 = ₹320m = ₹32 cr, credited to OCI (foreign currency translation reserve). (vi) The CTA / foreign currency translation reserve. Scoring: 6/6 in time means you're fluent. At ≤4, redo Worked Examples 5–6 and P13–P15 before the mastery check.

Applied mini-project: the group-anatomy audit

Task. Dissect one US and one Indian group through the notes this week taught. Budget ~3.5–4 hours. Every number in your output must carry a page/note reference to the filing.

Filings to pull.

1. Microsoft FY2024 10-K: SEC EDGAR (sec.gov → EDGAR full-text search → "Microsoft" → 10-K filed July 2024). You need: the business-combination note (Activision Blizzard), the stock-based compensation note, the segment note, and the fair-value measurements/investments note.
2. Reliance Industries latest consolidated annual report: ril.com → Investor Relations → Financial Reporting (or the BSE/NSE annual-report page). You need: the consolidated balance sheet and P&L (NCI lines), the segment note, the other-equity/OCI schedules, and the list of subsidiaries with holding percentages.

Deliverables.

1. PPA extraction (Microsoft–Activision): rebuild the allocation table with total consideration, major asset classes at fair value, each intangible category with its amortization life, and goodwill. Compute goodwill as a % of consideration and the annual amortization drag the deal added.
2. SBC panel (Microsoft): total SBC expense and SBC/revenue. Locate the CFO add-back. Find the unvested awards and the unrecognized compensation. Compute FCF two ways (reported CFO − capex, then the same minus SBC) and state the gap in dollars and %.
3. NCI panel (Reliance): NCI on the balance sheet and the P&L attribution split. From the subsidiaries annexure, name the two group companies that plausibly drive most of the NCI and the stake sales that created them. Then compute profit-attributable-to-owners ÷ total profit for the year.
4. Segment panels (both): revenue share vs profit share per segment; name each company's profit engine in one line each.
5. OCI safari (Reliance): list every OCI component reported, classify each as recycling or non-recycling, and identify the year's largest mover.
6. The memo (≤2 pages): "What the consolidated headline hides": your five most decision-relevant findings across both companies, conclusion first.

Scoring rubric. Be strict when you score yourself: 0 = missing/wrong, 1 = partial, 2 = complete and correct. You pass at ≥ 12/16.

Reading & resources

• EY, US GAAP vs IFRS: The Basics. [Free] [Intermediate]: sections on business combinations, consolidation, share-based payment, foreign currency, financial instruments; the fastest cross-standard reference alive.
• Deloitte IAS Plus (iasplus.com). [Free] [Intermediate]: one-page summaries of IFRS 2, IFRS 3, IFRS 10/11/12, IAS 21, IAS 28, IFRS 8, IAS 19, IFRS 9, IFRS 13. Bookmark; use for life.
• ICAI Educational Material on Ind AS 103 and Ind AS 110 (icai.org → Ind AS resources). [Free] [Advanced]: the authoritative walk-through of the Indian carve-outs, including the bargain-purchase capital-reserve routing and common-control pooling.
• Mauboussin & Callahan, "Stock-Based Compensation: Unpacking the Issues" (Morgan Stanley Counterpoint Global, 2023). [Free] [Intermediate]: the best single analyst treatment of the SBC debate.
• Damodaran on SBC. [Free] [Intermediate]: Musings on Markets post "Stock-based Employee Compensation: Value and Pricing Effects" and the SBC segment of his free NYU valuation classes; the subtract-from-FCF treatment argued from first principles.
• Kieso, Weygandt & Warfield, Intermediate Accounting. [Paid] [Intermediate]: ch. 16 (dilutive securities and EPS), ch. 17 (investments).
• Hoyle, Schaefer & Doupnik, Advanced Accounting. [Paid] [Advanced]: the consolidation chapters (equity method, consolidation mechanics, intercompany transactions, FX translation; ch. 1–7 in recent editions). Go here when you want the full machinery beyond this week's depth.
• Farhat Lectures (YouTube). [Free] [Beginner→Intermediate]: playlists "Business Combinations," "Consolidation," "Earnings per Share," "Share-Based Compensation," free lecture-length worked examples.
• Wharton (Bushee), Introduction to Financial Accounting, Coursera. [Free] [Beginner]: final weeks reinforce statement articulation with these advanced items.
• CFA Level II FSA readings. [Paid] [Advanced]: "Intercorporate Investments," "Employee Compensation: Post-Employment and Share-Based," "Multinational Operations," the professional-exam treatment of this same material.
• Primary filings to keep as specimens: Microsoft FY24 10-K (PPA, SBC, segments; EDGAR) [Free]. Infosys latest annual report, ESOP note + segment note (infosys.com) [Free]. Reliance Industries consolidated AR (NCI, segments) [Free]. Vedanta AR (NCI: Hindustan Zinc) [Free]. Tata Motors Passenger Vehicles AR (JLR translation, FCTR; Tata Motors ARs up to FY2024-25 for the pre-demerger years) [Free]. First Citizens BancShares 2023 filings (a real bargain-purchase gain) [Free]. All [Intermediate] as reading.
• Going deeper: the primary standards with their illustrative examples (IFRS 3, IFRS 10/11/12, IAS 21, IAS 19, IFRS 9/13 on ifrs.org; the Ind AS equivalents and carve-out text on mca.gov.in). [Free] [Advanced]: slow but authoritative. They're the only place you'll find the edge cases and the Basis for Conclusions, the why behind each rule. Go here when a one-page summary leaves you unsure why a treatment exists.
• Going deeper: Stephen Penman, Financial Statement Analysis and Security Valuation. [Paid] [Advanced]: the reformulation chapters (operating vs financing items, minority interest, comprehensive income and "dirty surplus") turn this week's presentation rules into inputs a valuation can actually use; the rigorous bridge from here into Phases 2–3.

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": "Stock-based compensation and dilution, consolidation and non-controlling interests, purchase price allocation, currency translation, segment reporting and the OCI decoder",
  "use": "This module is a note-reading module — nine topics, all of which live in footnotes — which makes it the best place in Phase 1 for first-pass document triage, and the worst place to accept a summary. (1) Point a document tool at the SBC note, the group-structure and NCI note, the segment note and the OCI statement across a peer set and ask for one table per topic with the page reference for every cell; the reference is the point, because a cell you cannot find is a cell you must not use. (2) Paste an acquisition's purchase-price-allocation table and ask what the goodwill implies about the multiple paid and which intangibles were recognized versus left inside goodwill — then argue with the answer using the PPA arithmetic of §4.5. (3) Ask it to read an 'adjusted EBITDA' reconciliation and list every add-back that is a real economic cost; SBC will be on the list, and the point is to see how the adjustment was built. (4) Ask it to build the group tree — subsidiary, associate, joint venture, consolidation basis, ownership percentage — from the related-party and subsidiaries schedules, which is tedious by hand and mechanical for a tool.",
  "tools": ["Document/PDF AI — NotebookLM or Claude file upload for the SBC, subsidiaries, segment and OCI notes across a peer set", "Chat assistants — Claude, ChatGPT for PPA and consolidation fact patterns under Ind AS 103/110 versus ASC 805/810", "Coding assistants — for the option-valuation, treasury-stock and dilution-distribution code below"],
  "prompt": "From the attached filing, extract: (1) the SBC note — expense by function, grant-date fair value, the option-pricing inputs (volatility, expected life, risk-free rate, dividend yield), awards outstanding, and total unrecognized cost; (2) the diluted-share reconciliation; (3) the subsidiaries and associates schedule with ownership percentages and consolidation basis. Give a page or note reference for every single figure. List separately any figure that is NOT disclosed in this filing. Do not compute an option value, a volatility, or a share count that the filing does not state — I will compute those myself.",
  "verify": "Reconcile every extraction to the statement it must tie to. SBC expense must equal the CFO add-back line; the diluted share count must reproduce the filed diluted EPS to the reported precision; the subsidiaries schedule's ownership percentages must be consistent with the NCI balance on the balance sheet; OCI components must sum to the movement in the corresponding reserve. Then rebuild the treasury-stock arithmetic yourself from the disclosed strike and average price, because a tool that mis-reads one strike price produces a plausible dilution number that is simply wrong. A model asked to summarise a footnote will produce a summary; only the note establishes what the company actually said (Guardrail 1-5).",
  "diy": "The gate is unaided: you compute SBC expense under both attribution methods, the treasury-stock incremental shares, a full consolidation with NCI, a PPA to goodwill, and you decode an OCI statement line by line from a blank sheet. Nothing in this callout is admissible in that room."
}

Modern Data Analysis

By hand first. You built the graded-versus-straight-line attribution table, the exercise-day journal entry, the treasury-stock method and the consolidation eliminations above, and the gate still asks for all of them on paper. Code adds one thing. The option value itself stops being a disclosed number you accept and becomes one you can reprice under your own assumptions, and that's the only way to test the note.

Today's workflow. Three things move from tedious to trivial. First, repricing the awards. A Black–Scholes call is one function, so you can take the filing's own disclosed inputs, check the grant-date fair value it reports, then re-run it at your own volatility. On a synthetic at-the-money four-year option with a 7% risk-free rate and no dividend, the value runs 29.33% of the share price at 20% volatility, 35.04% at 30%, and 41.06% at 40%. The "very roughly 25–40% of the share price" rule of thumb from earlier comes from runs like these. It also shows that expected volatility, the input a company has the most latitude over, moves the reported expense by a third. Second, the attribution schedules. The graded-vesting spike of ₹62.5 lakh in year 1 against a straight-line ₹30 lakh is a four-line loop, so you can restate an Ind AS-versus-US-GAAP margin comparison onto one basis instead of hand-waving it. Third, the panel work: SBC ÷ revenue, SBC ÷ CFO, award overhang ÷ shares outstanding, and buyback ÷ SBC across a whole sector. Across a sector, the earlier buyback-tell observation stops being an anecdote and becomes something you can see.

Tools & sources (IN + US). Use numpy/scipy for the option and simulation work and pandas for the panel. For India, use the ESOP/RSU note and the SEBI (Share Based Employee Benefits) disclosures, the subsidiaries-and-associates schedule, the Ind AS 108 segment note and the OCI statement in the annual report (BSE/NSE). Add the stock exchanges' shareholding-pattern filings for the dilution history and NSE option data for a market-implied volatility cross-check. For the US, use EDGAR XBRL companyfacts (ShareBasedCompensation, AllocatedShareBasedCompensationExpense, MinorityInterest, Goodwill, WeightedAverageNumberOfDilutedSharesOutstanding), the 10-K SBC, segment and business-combination notes, the DEF 14A proxy for the award design behind the numbers, and FRED (DGS3, DGS5) for the risk-free input.

# Synthetic, module's own figures — R3: every number recomputed in-session
import numpy as np
from math import log, sqrt, exp, erf
N = lambda x: 0.5 * (1 + erf(x / sqrt(2)))
def bs_call(S, K, T, r, sig, q=0.0):                       # grant-date fair value
    d1 = (log(S / K) + (r - q + sig * sig / 2) * T) / (sig * sqrt(T))
    return S * exp(-q * T) * N(d1) - K * exp(-r * T) * N(d1 - sig * sqrt(T))
print([round(bs_call(100, 100, 4, 0.07, s), 2) for s in (0.20, 0.30, 0.40)])  # 29.33 35.04 41.06
tot, tranches = 120.0, 4                                   # ₹120 lakh over 25%-a-year vesting
graded = [round(sum(tot / tranches / (j + 1) for j in range(tranches) if j + 1 >= y), 2)
          for y in range(1, tranches + 1)]                 # [62.5, 32.5, 17.5, 7.5]
print(graded, sum(graded), tot / tranches)                 # ties to ₹120 lakh; SL = 30.0/yr
print(round(10.0 * (1 - 250 / 400), 2))                    # treasury-stock method: 3.75 cr

Verify. A repriced option value is your number, not the company's, and the memo has to say so. Before you use it, reproduce the filed fair value from the filed inputs. If your Black–Scholes doesn't land on the disclosed grant-date value, you've got the wrong expected life or the wrong dividend treatment, or the company used a lattice model. Every downstream figure inherits that error. Tie SBC expense to the CFO add-back and the equity credit before computing any ratio, and tie your rebuilt diluted count to the filed one. Keep the two consistent treatments from earlier as separate in code as on paper. Subtract SBC from FCF or grow the share count, never both.

Quantitative lens

The treasury-stock method gives one number for a claim that's really a distribution, and here you can see it. Incremental shares are N × (1 − K/P) when the option is in the money and zero when it isn't. So dilution is a kinked, non-linear function of a future price nobody knows. Take a synthetic company with 100 crore shares out, 10 crore options struck at ₹250, a share price of ₹400, three years to the horizon, 40% volatility and a 7% risk-free rate. The filed treasury-stock number at today's price is 3.75 crore incremental shares (3.61% dilution), and it's the right disclosure. It isn't the expected outcome, though. Simulate 200,000 lognormal price paths at those parameters. In 26.24% of them the options finish below the strike and dilute nothing at all. The median path dilutes 3.55 crore shares (3.43%). The 95th percentile dilutes 7.94 crore (7.36%) and the 99th 8.72 crore (8.02%). Mean incremental shares come to 3.47 crore, below the 3.75 crore point estimate, because the probability mass sitting at zero more than offsets the upside tail. Evaluating the same formula at the three-year forward price of ₹493.5 gives 4.93 crore, which overstates it. That's three defensible-looking single numbers, all different. None of them is the distribution.

That gives you the right sentence for a memo. Employee options carry a one-in-four chance of nothing and a one-in-twenty chance of losing more than 7% of the company, which is a different thing from a 3.6% haircut. A point estimate hides the second half of that sentence. It's also the fair way to read an overhang figure. A 10% overhang struck deep out of the money is a different animal from a 10% overhang struck at yesterday's price.

These limits are real. This is a lognormal random walk with a constant volatility and no early exercise. The option-pricing literature criticises that model, and it's the same model the filings themselves use. It says nothing about new grants, which are the larger source of future dilution for a growing company. It says nothing about exercise behaviour either, and that's empirically far from optimal. The volatility input dominates the answer, and it's a choice. Use the distribution to size a range and to kill the illusion of precision in the point estimate. Never treat it as a forecast of a share count.

Do it in code: simulate the price paths, apply the treasury-stock formula path by path, and report the dilution percentiles and the probability of zero dilution alongside the filed number; then layer an assumed annual grant rate on top and watch the tail thicken. See G1/DV1.02 and G1/QM1.03.

Where this goes next: galaxy cross-links

• G1 · DV1.02 Options Pricing & the Greeks: where the grant-date fair value from the measurement section above stops being a disclosed input and becomes something you can price, hedge and stress; the vega of an employee option is the most important number the SBC note doesn't print.
• G1 · QM1.03 Time Series, Monte Carlo & ML: the simulation machinery behind the dilution distribution above, including how badly a constant-volatility lognormal walk behaves and what to do about it.
• G2 · DA1.02 Pandas Data Wrangling: the SBC, overhang and buyback panel across a sector, which is what turns the earlier "buyback tell" from an anecdote about two companies into a distribution with outliers in it.
• G2 · DA1.04 Exploratory Data Analysis & Visualization: the segment note above as small multiples across years and reportable segments, where a deteriorating division shows up as a shape long before it shows up in the consolidated total.

Teach it back & journal

Feynman prompt. Write a one-page explainer for a smart 15-year-old that answers two questions with one story each: (1) "If a company pays its engineers only in its own shares, why do its owners get poorer even though no money left the bank?"; you must use the words dilution, grant date, and buyback. (2) "You own 80% of a lemonade-stand chain. Why do your family accounts show 100% of its lemons, 100% of its debts, and a line that says 'one-fifth of this isn't ours'?" No jargon beyond the three required words. If the 15-year-old couldn't re-explain NCI to a friend, rewrite it.

Journal reflection. Tonight's entry has three parts. (1) Boundaries. This week showed the same economics wearing three costumes: consolidated, equity-method, and fair-value-instrument. Where else in my analysis might a presentation boundary (not an economic change) be driving what I see? (2) Pull up one company I follow and read only its segment note. What did the consolidated numbers hide, and does it change my one-line view of the business? (3) Record one number from this week that really surprised me (the graded-vesting front-load, the invisible associate debt, the Ind AS bargain-purchase routing, whatever it was) and why my prior intuition was wrong.

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