Not one statement, but several - the statutory formats and why they differ

Everything you learned in Part One is true - for one kind of company. Open a manufacturer's annual report and the lines are where you expect them: revenue at the top, a gross margin below it, inventory and plant on the asset side, borrowings and payables on the funding side. Now open a bank's report with exactly that training, and almost nothing maps. There is no "revenue from operations" - there is interest earned. There is no gross margin at all, no inventory, no working capital. Deposits, which feel like a liability, are the raw material the business runs on; loans, which sound like borrowing, are the asset. And debt-to-equity, the ratio you would reach for first, is meaningless, because leverage of ten times is the design, not a warning. illustrative

This is not a bank being difficult. It is a bank filing under a different statutory format because it is a fundamentally different business - and it is not only banks. Four families of statement carry most of the exchange, and a reader who does not know which one they are holding will silently apply the wrong template and reach a confident, wrong conclusion. This module is the pivot of the whole guide: Part One taught you to read one format extremely well; this part teaches you it was one format among several, and that the skill you now need is not more ratios but knowing, the moment you open a report, which document you are reading, which familiar lines have vanished, and which unfamiliar lines actually run the business.

A statement's format follows the business's economics. A manufacturer buys materials, holds inventory, sells goods and waits to be paid, so its statement carries inventory, gross margin and working capital because those things are real for it. A bank has none of them: it takes deposits and makes loans, its "cost of goods" is the interest it pays depositors, its "product" is the interest it earns on loans. Forcing a bank into a manufacturer's format would hide the business entirely, so the law gives it a different one. The statutory format is not bureaucratic pedantry - it is the shape the accounts must take for the business to be legible at all.

There are four families to know. Division II of Schedule III is the manufacturer-and-most-companies format you already read. The RBI's banking formats - prescribed by the Reserve Bank of India under the Banking Regulation Act - govern scheduled commercial banks. Division III of Schedule III governs NBFCs (non-banking financial companies - lenders that make loans but, unlike banks, cannot take public deposits) and housing finance companies. IRDAI formats - from the Insurance Regulatory and Development Authority of India - govern insurers, and split the accounts into a policyholders' account and a shareholders' account, because the two pools of money must not be confused. A handful of others exist - SEBI's trust formats for REITs and InvITs, the listed trusts that pool investors' money to own income-producing property or infrastructure - but these four carry the exchange.

The trick that makes all of this manageable is to stop reading line names and start reading slots. Every business, whatever its format, answers the same handful of questions, and each answer sits in a slot even when the line has a different name or is missing entirely.

The top line - what does the business sell? Revenue from operations for a manufacturer; interest earned for a bank; interest income for an NBFC; gross written premium for a life insurer. For a developer, reported revenue exists but is nearly useless, recognised only on completion - you read pre-sales instead.

The real margin - what is the true measure of profitability? A manufacturer has a gross margin. A bank has none; its equivalent is net interest income - interest earned minus interest paid - expressed as a net interest margin. A life insurer's real margin is the value of new business, because profit emerges over decades and this year's reported profit barely describes it.

The core asset - what does the business own that earns? A manufacturer's inventory and plant; a bank's advances; an NBFC's loan book; an insurer's pool of investments backing policyholder liabilities; a developer's "inventory" of land and half-built projects that sit for years.

The funding and raw material. Equity and borrowings fund a manufacturer. For a bank, deposits are both funding and raw material, which is why they are not a warning. For an NBFC, borrowings play that role, and because there are no deposits, the cost of those borrowings and the risk of not rolling them over dominate everything.

What is simply absent. This is the most useful slot of all. A bank has no gross margin, no inventory, no working capital, no meaningful debt-to-equity. A developer has no meaningful annual revenue. An IT firm has almost no inventory or capital employed. Knowing which of your familiar lines does not exist stops you computing a ratio that means nothing and trusting the answer.

The widget above switches between a bank, an NBFC, a life insurer, a developer and an IT firm, holding the five slots fixed on the left while the lines that fill them change completely - and some slots go empty, turning red. That is the whole method of Part Two in one tool: read the slots, not the line names. Module 112 later turns it into a general procedure you can run on a sector the widget does not even list.

The inversions here are structural, not matters of degree. A line does not merely mean something a little different in another sector - it ceases to exist, or it means the opposite. This is the sharpest form of "same number, different verdict," because sometimes there is no number at all.

The bank is the clearest structural inversion, and it is worth stating plainly because the rest of this part builds on it: two of the most fundamental things you learned in Part One - that revenue sits at the top of the P&L, and that high leverage is dangerous - are simply false for a bank. There is no revenue line, and high leverage is the design. Read a lender with a manufacturer's instincts and you are not slightly off; you are reading a different document than the one in front of you.

The other three each reward reading the sector's own lines rather than the ones you brought. On a life insurer, skip past the shareholders' account to the value of new business and to persistency - whether policyholders keep paying - because that is what reveals a product genuinely sold rather than mis-sold. On a developer, a near-zero revenue year with strong pre-sales is a good year the P&L conceals. On an IT firm the lines look normal, because the format is familiar, but the asset that earns - the workforce - appears nowhere on the balance sheet, so a pristine balance sheet tells you almost nothing about whether the business is winning.

Knowing the format tells you which lines to read and which to ignore; it does not tell you whether the business is any good. A bank read correctly on NIM (net interest margin), NPAs (non-performing assets - loans the borrower has stopped repaying) and capital adequacy can still be a poor bank; a developer read correctly on pre-sales and collections can still be building in the wrong city at the wrong time. The format is the grammar of the statement - it lets you read the sentences - but a grammatically perfect reading of a bad business is still a bad business. The modules that follow supply the quality judgement for each sector; this one supplies only the ability to read the page.

Nor does one format map cleanly to a business that spans several. A conglomerate that owns a bank, a factory and an insurance arm files a consolidated statement blending three incompatible formats into one, and no single template reads it. Knowing the four families helps, but it also warns you that some companies cannot be read from one statement at all - the standalone entities each file their own format, and the consolidated view is a genuine mongrel that the holding-company modules take up later. Knowing that a bank has net interest income instead of gross margin is the start, not the end; what a healthy NIM is, how provisioning conceals or reveals the truth, what an asset-liability mismatch looks like before it kills an NBFC - those are the substance, and they need their own modules.

How it comes up. The format question surfaces whenever someone imports the wrong metric, and a good analyst calls it out. On a bank's call it sounds like a correction of a peer comparison: "A brokerage note compared your debt-to-equity unfavourably with a manufacturer's. Can you help investors understand why that comparison is meaningless for a bank, and which leverage and capital measures they should use instead?" The question invites management to reframe the business in its own format rather than a borrowed one.

A good answer, verbatim-style.

"It's a fair thing to clear up. Deposits are our raw material, not debt in the corporate sense - we take them in and lend them out, so a balance-sheet leverage of nine or ten times is normal and healthy for any bank. The measure that actually constrains us is capital adequacy, which is regulated: we're at 16.5% against an 11.5% requirement, so there's real buffer. On asset quality, gross NPAs are 2.1% with 75% provision coverage. Those three - capital adequacy, NPAs, coverage - are how you should read our safety, not debt-to-equity." illustrative

It rejects the wrong metric, names the raw material, and hands the listener the three measures that actually apply, with numbers. It is teaching the correct format.

An evasive answer, verbatim-style.

"We're very comfortable with our balance sheet strength and our capitalisation is among the best in the industry. We've always maintained a conservative approach to risk and our metrics compare well with peers. Investors shouldn't be concerned about leverage - banking is a well-regulated industry and we operate well within all norms."

It reassures and says nothing usable. It never explains why debt-to-equity is the wrong lens, gives no capital-adequacy or NPA figure, and names none of the metrics that do apply. "Compares well with peers" and "within all norms" are assertions, not the reframing the question asked for - and a listener still holding a manufacturer's template learns nothing. The follow-up that settles it - "What are your exact capital adequacy, gross and net NPA, and provision coverage, and how have each moved over eight quarters?" - forces the format into specific numbers. When it is not asked, the silence is the tell: either the trend is uncomfortable, or the analysts covering the stock are themselves reading it with the wrong template.

The first trap is applying a manufacturer's ratios to a financial company and trusting the output. Debt-to-equity on a bank, gross margin on a lender, working capital on an insurer - each produces a number, and a number feels like an answer. But the ratio was built for a business with inventory and trade credit, and the financial company has neither, so the figure is not merely imprecise, it is meaningless. The screener that ranks a bank as "highly leveraged" and a developer as "no revenue" is applying the wrong template mechanically, and a reader who trusts the screener inherits the error.

The second trap is subtler: assuming that because a company files under Division II, like a manufacturer, its statement reads like a manufacturer's. An IT services firm and an asset manager both file Division II, but for both the balance sheet is nearly irrelevant - the asset is people, or a fee stream, not plant and inventory. Reading capital employed and asset turnover for a business that has almost no capital produces return ratios that look extraordinary and mean far less than they appear. The format is a guide to the statement, not a guarantee that the familiar lines carry their familiar meaning - which is the same reason a growing life insurer or a busy developer looks like it is failing when the P&L merely lags the real activity: the healthiest year of new business can show the lowest profit, and the strongest year of selling almost no revenue.