How to Value Banks and Gold-Loan Companies: ROE, Price-to-Book and Muthoot's Q2 FY27

Banks, lenders and insurers make up roughly a third of the Nifty 50 by weight, and they are the companies where the standard stock-page ratios mislead you most. Open Bank of Maharashtra on screener.in in late October 2026 and it shows a return on capital of about 6% sitting next to a return on equity of 22.6%. One of those numbers is meaningless for a bank, and if you don't know which, you will draw the wrong conclusion about every lender you look at.

This post teaches you how to value a lender from first principles. We start with the one formula that sits under every valuation, build up return on equityROE (Return on Equity)Profit generated for every rupee or dollar of shareholders' money invested. ROE above 15% is generally good. Watch out: high ROE driven by heavy debt is misleading — check ROCE alongside it.See all terms in the glossary and price-to-book step by step, compare five Indian banks with the same arithmetic, and then work through a gold-loan company in detail using Muthoot Finance's Q2 FY27 investor presentation, published on 31 October 2026.

If you are new to valuation, our guide to P/E ratios and Valuation 101 cover the basics this post builds on. When we ran our quality screen on the Nifty 200, lenders needed a separate set of rules. This is the reasoning behind those rules.

Nothing in this post is a recommendation to buy, sell or hold any stock. It is an analysis written to teach a method. ITC, the five banks, Muthoot Finance and Manappuram appear only as worked examples, and every "fair price" or scenario return below is the output of stated assumptions, not a view on what any share will do. Do your own research, and speak to a SEBI-registered investment adviser before acting on anything.

All market figures are approximate and as of late October 2026. They will have moved by the time you read this. The method will not.

The formula underneath every valuation

As a shareholder you make money in two ways: cash paid to you as dividends, and the share price going up. Since price equals the P/E ratioP/E Ratio (Price to Earnings)Tells you how many years of current earnings you're paying for when you buy a stock. A P/E of 25 means you pay 25 years of today's profits upfront. High P/E = market expects strong growth; low P/E = slow growth or high risk.See all terms in the glossary multiplied by earnings per share (EPS), the price can only rise if EPS rises or the multiple rises. That gives you a three-part formula:

ITC makes a clean example. In late October 2026 it traded near ₹266 with trailing EPS of ₹15.84 (a P/E of 16.8) and a dividend of about ₹14.5 a share, a 5.5% yield. Suppose profit grows 6.5% over the next year and the market keeps paying 16.8 times earnings.

If profit grows only 4%, the price reaches about ₹277 and the gain is ₹11 plus ₹14.5, about 9.5%. Each point of growth adds roughly a point of return.

You can also run the formula backwards to see what a price assumes. If you want 12% a year from ITC, the dividend already supplies 5.5%, so the other 6.5% must come from growth. ITC's EPS went from ₹8.04 in FY15 to ₹16.51 in FY26, about 7% a year. At ₹266, the price only needs ITC to keep doing what it has done for a decade (you can run this for any company with our required growth calculator). On that arithmetic the price looks fair rather than a bargain, and the open question becomes whether the February 2026 cigarette excise hike is a one-time step down or the start of slower growth (we covered that in our ITC Q1 FY27 results breakdown).

The third lever is the multiple itself. If profit grows 6.5% and the market regains confidence enough to pay 20 times, the price becomes ₹337. If the P/E slips to 14, it becomes ₹236, a loss despite the growth. Over one or two years this lever dominates and looks random. Over five or more, growth and dividends take over, which is why patient investors focus on the parts they can estimate.

One Indian wrinkle: dividends are taxed at your income-tax slab rate, while long-term capital gains are taxed at a lower rate. For high-yield stocks, the post-tax return can be noticeably lower than this simple sum.

ROE and ROCE, and how debt flatters one of them

P/E tells you what you pay for the profit. ROE and ROCEROCE (Return on Capital Employed)Measures how efficiently a company generates profit from all the capital it uses — both equity and debt. ROCE above 20% is generally excellent. The best businesses compound ROCE above 25% for decades.See all terms in the glossary tell you how good the business is at turning money into profit.

Take two shops. One owner puts ₹10 lakh into a tea stall and makes ₹3 lakh a year, a 30% return. Another puts ₹1 crore into a restaurant and makes ₹5 lakh, a 5% return. The restaurant makes more money in total, but the tea stall is the far better business. Another ₹10 lakh would earn the stall owner ₹3 lakh more and the restaurant owner only ₹50,000.

ROE measures the return on the owners' money:

Shareholders' equity is the same thing as book valueBook ValueA company's assets minus its liabilities, as recorded on the balance sheet — its net worth on paper. For most of Berkshire's history, growth in book value per share was Buffett's headline measure of value creation.See all terms in the glossary: everything the owners put in, plus all the profit kept in the business over the years. ITC's EPS of ₹15.8 against book value of ₹57.9 per share gives an ROE of about 27%.

ROCE measures the return on all the money in the business, borrowed money included. It divides operating profit before interest and tax by equity plus debt.

The difference matters because borrowing can make ROE look better without the business improving. A business needs ₹100 and earns ₹10 of operating profit, a mediocre 10% return. If the owners supply all ₹100, ROE is about 10%. If they supply ₹20 and borrow ₹80 at 8%, interest costs ₹6.40, leaving ₹3.60 on ₹20 of equity: an 18% ROE. Same shop, same customers, same 10% ROCE. The debt flattered the ROE and added risk. ROCE cannot be inflated by borrowing, which is why for most companies you want both numbers side by side.

For ITC the two line up. It has almost no debt, so the gap between its 38.9% ROCE and its ROE is mostly tax: 38.9% × (1 − 25%) ≈ 29%.

Why ROE drives growth

ROE also tells you how fast a company can grow using its own money:

A business earning 30% ROE that reinvests half its profit can grow about 15% a year. One earning 8% that reinvests the same half grows about 4%. ITC has a high ROE but pays out around 90% of its profit, so it grows slowly and hands the quality back to you as dividends.

How ROE connects to price

There is one more identity worth knowing, because it is the bridge to valuing banks. Divide price-to-book by ROE and you get P/E:

ITC's price-to-book of 4.59 divided by a 27% ROE gives about 17, which matches its 16.8 P/E. The business earns 27% on its book, but you pay 4.6 times book, so your return on the price you paid is about 27 ÷ 4.6 ≈ 6% before growth. ROE tells you how good the business is. The price tells you whether it is a good deal.

Why a bank breaks the usual ratios

For a normal company, debt is one way to pay for the factory. For a bank, borrowed money is what the factory processes.

A grocer buys rice at ₹40 a kilo and sells it at ₹45. A bank "buys" money at 6% (by paying interest on deposits) and "sells" it at 10% (by charging interest on loans). Deposits are the bank's rice, not debt in the usual sense. That is why ROCE and debt-to-equity are meaningless for banks: they count the rice as a liability and conclude the grocer is drowning in borrowing.

Here is a simplified bank with ₹100 of loans.

The bank keeps ₹1.20 for every ₹100 it lends. That is its return on assets (ROA), 1.2%, about one paisa per rupee. What makes banking a decent business is leverage: ₹10 of owners' money supports ₹100 of loans, so the ₹1.20 is earned on ₹10 of equity.

So for banks, you swap the usual dashboard for this one.

NBFCs and the other kind of crisis

A non-banking financial company (NBFC) lends like a bank but cannot take ordinary savings deposits, so it borrows from banks, bond investors and mutual funds instead. Many NBFCs lend for 5 to 15 years while borrowing for 3 to 12 months, which means they must keep rolling their borrowing over. If lenders suddenly stop, the NBFC can be in trouble even when every one of its borrowers pays on time. That is a liquidity crisis, not a bad-loan crisis.

India saw exactly this in September 2018. IL&FS, an infrastructure lender rated AAA, defaulted. Mutual funds panicked and stopped buying short-term paper from NBFCs in general, and healthy NBFCs found their borrowing costs jumping or their refinancing gone. When you look at an NBFC, check how well its loan maturities match its borrowing maturities, how much it relies on commercial paper (very short-term borrowing), how much cash it holds, and who its parent is.

Book value, and what a fair price-to-book looks like

Book value is what would be left for shareholders if the company sold everything at the values in its accounts and paid off everything it owes. A flat worth ₹1 crore with a ₹60 lakh home loan has a "book value" of ₹40 lakh: the part that is really yours.

For ITC, book value badly understates the business, because its brands and distribution network are not on the balance sheet at anything like their real worth. For a bank, almost everything it owns is loans recorded close to their real value. That makes price-to-book (P/B) the natural yardstick for lenders, in the way P/E is for consumer companies.

Think of the book as a pot of ₹100 that earns its ROE every year. With a 12% ROE and all profit paid out, your return depends entirely on what you paid for the pot:

A bank earning 9% ROE would need to be bought at about 0.75 times book to give you 12%. Worse, the more such a bank grows, the more value it destroys, because every rupee it reinvests earns less than you require.

Adding growth, and why high-ROE banks deserve more

Now take a bank with a 20% ROE that keeps 75% of its profit and pays out the rest. Its book grows 15% a year. We modelled holding it for 10 years, collecting the dividends, and selling at a modest 2 times book at the end. The book grows from ₹100 to about ₹405 over the decade.

For someone who wants 12% a year, a reliable 20%-ROE bank is worth about 3 times book. You are paying for a pot that keeps compounding at 20%. With any bank, two questions settle most of the valuation: how many times book am I paying, and can this ROE last ten years or more?

Five Indian banks through the same lens

Dividing ROE by P/B gives you the starting return on your price, before any growth.

Bank of Maharashtra looks like the obvious winner on this table, and that is the trap. A 22.6% ROE for a public-sector bank is unusually high, helped by low credit costs at a good point in the credit cycle. Run the same pot model over five years with a less kind assumption: ROE averages 14% instead of 22.6%, the bank keeps 75% of profit, and the market re-rates it from 1.68 times book to 1.0 times. The return works out to about 2% a year. If ROE instead averages 18% and the multiple eases only to 1.4 times, it is about 12%.

Karnataka Bank shows the opposite case. Its ROE is unremarkable, but at 0.85 times book you still start with about 14.7% on your price. Its risk is that a bad credit cycle pushes ROE to 10% and the market marks it down to 0.6 times book, in which case the same model gives under 4% a year.

HDFC Bank sits in between. Its ROE has dipped since the 2023 merger with its parent HDFC Ltd (we covered the business in our HDFC Bank analysis), so on today's figures it looks expensive. The bet there is that ROE recovers toward its long-run 16% to 17%, which is a question about the bank, not about its ratio.

Most banks have a similar best case. Where they differ is the bad case. Cheap-looking banks are usually cheap because their ROE is at a cyclical high (Bank of Maharashtra) or not yet proven (IDFC First). For compounding you can rely on, the question is whose ROE you can trust for ten years at a P/B you can live with.

Where gold-loan companies fit

Financial companies are not one family, and each needs a different lens.

A gold lender is a pawnbroker run at enormous scale. A customer brings in ₹1 lakh of jewellery and borrows, say, ₹60,000 against it. The gold goes into the lender's vault. Repay within a few months and the jewellery comes back; fail to repay and the lender auctions it.

Borrowers pay 18% to 24% a year because the loan arrives in about 30 minutes, with no credit score and very little paperwork. The lender is safe because it holds gold worth more than the loan. Loan-to-value (LTV) measures that cushion: a 60% LTV means the gold is worth about 1.7 times the loan.

High yields and tiny losses give gold lenders an ROA of 5% to 6%, against 1% to 2% for banks. They also borrow far less, with leverage of 4 to 5 times against 10 to 12 for banks. Put those together and you get an ROE of roughly 6% × 4.5 ≈ 25% or more, without heavy borrowing.

The hidden driver is the gold price

When gold rises 30%, a customer pledging the same 20 grams can borrow 30% more. The loan book grows 30% without a single new customer. When gold falls, the reverse happens just as automatically. Part of a gold lender's "growth" is really the gold price, which can reverse (our post on how gold, the dollar and the rupee connect explains what drives it). That is why you track tonnes of gold held, not just rupees lent.

Funding works in the gold lender's favour. Gold loans come back within 3 to 12 months, while much of the lender's own borrowing is longer. In a credit freeze, a gold lender can stop lending, collect repayments and shrink. That makes it structurally the opposite of IL&FS.

The real risks are a sharp fall in gold (the stocks trade partly as gold proxies), competition from banks and large NBFCs squeezing yields, regulation (the RBI tightened gold-loan rules in April 2026 and has acted against individual lenders before), and riskier side businesses such as microfinance, which hurt Manappuram Finance in FY25 and FY26.

Reading Muthoot Finance's Q2 FY27 results

Muthoot Finance is India's largest gold-loan NBFC. Its Q2 FY27 investor presentation, covering the quarter to 30 September 2026, had a strong headline.

AUM means assets under management, here simply the total gold loans outstanding. Standalone figures cover Muthoot alone; consolidated adds its subsidiaries (Muthoot Money, Belstar microfinance, Muthoot Homefin and a Sri Lankan arm), which together contribute about 12% of group profit. The ROE checks out against our formula: 6.62% ROA multiplied by leverage of about 4.5 gives roughly 29%.

Underneath the headline, two things mattered more.

Tonnes against price

The presentation gives the gold the company holds and the gold price each quarter.

Loan growth can be split into three parts that multiply together: more tonnes, a higher gold price, and more rupees lent against each gram. From September 2025 to September 2026, tonnes fell 2.4% (204 ÷ 209) and the gold price rose 28.2% (13,547 ÷ 10,566). AUM grew 43.3% (1,790 ÷ 1,249). Dividing out the first two, 1.433 ÷ (0.976 × 1.282) ≈ 1.145, so lending per gram rose about 15%.

In other words, almost all of the 43% growth came from a higher gold price and lending more against each gram, not from holding more gold. The encouraging sign is the most recent quarter, where tonnes rose from 197 to 204 and active customers grew 4%.

The safety cushion has shrunk

A year ago, gold had to fall 43% before the average loan was worth more than its collateral. Now it is 35%. The change followed the RBI's April 2026 tiered LTV rules, which allow higher lending against smaller loans. That is good for growth and worse for resilience to a gold fall.

Funding looked safe. Banks supplied 46% of borrowings, listed bonds 34%, overseas bonds 16%, and commercial paper only 3%. With about ₹13,500 cr of cash, AA+ credit ratings and short-dated loans, the IL&FS kind of risk is low.

Two other numbers in the deck reward a closer look. The interest spread is the loan yield (19.41%) minus the cost of borrowing (8.64%), which gives 10.77%. NIM, at 11.93%, is higher than the spread because part of the loan book is funded by shareholders' equity, which costs no interest. And operating expenses fell to 2.10% of average loans, from about 3.5% in FY22. When the gold price lifts the loan book, the same branches and staff handle more rupees. That operating leverage reverses if gold falls.

Stages, provisions and the jar of money set aside

That callout used words that sound technical and are not, once you see what they describe.

Under Ind AS 109, the accounting standard Indian lenders follow, every loan sits in one of three stages.

Lenders must set money aside for the losses they expect, not only the ones they have already suffered. That set-aside is called the expected credit loss (ECL) provision. Stage 1 loans need a small provision. Stage 2 and Stage 3 loans need provision for the loan's whole expected lifetime of losses, which is why a jump in Stage 2 forces a large top-up even before anyone defaults.

The easiest way to keep the vocabulary straight is to picture a jar.

Impairment is easiest to grasp at small scale. You lend a friend ₹1,000 and start to suspect you will only get ₹800 back. Before your friend has actually failed to pay, you note down that you will probably lose ₹200. That ₹200 reduces your profit today. At Muthoot, the extra ₹464 cr of impairment over Q1 cost roughly ₹350 cr of after-tax profit, even though very little money was actually lost.

Why low coverage can be fine for gold loans

A bank lending to a company might recover ₹30 of every ₹100 after years in court, so it provides ₹70 up front: 70% coverage. Muthoot holds gold worth roughly ₹1,00,000 against a ₹60,000 loan and can auction it within weeks, so its expected loss is close to zero. The gold is the real protection, and a 12% coverage ratio that would alarm you at a bank is reasonable here.

That only holds while gold stays high. Expected loss is roughly the amount owed minus what the gold will fetch. Take a bad loan where the borrower, with interest added, owes ₹72,000:

With only 12% set aside, a deep enough fall in gold forces provisions to be built quickly out of current profit, at exactly the moment loan growth and margins are also shrinking.

When losses turn out smaller than expected

Provisions are not cash spent. The money stays in the business, and if losses come in lower than expected the provision is released ("written back"), adding to profit later. Over a full cycle, profit is not lost, only shifted in time. After COVID, many Indian banks released earlier over-provisions and reported a burst of profit. Muthoot already carries a deliberate ₹295 cr "excess provision" cushion on top of what the rules require.

Because management has some judgment over provisions, they can be used to smooth profits. To see through that, use pre-provision operating profit (PPOP):

On this view the core business grew faster than reported profit. Treat that with care, though. Part of the PPOP jump came from a higher loan yield, and on bullet loans (explained below) interest on the overdue loans is still being booked as income. If those borrowers repay, Muthoot gets both the interest and a provision write-back. If they don't, both reverse.

CRAR, the owners' cushion

Lend ₹100 with ₹20 of your own money and ₹80 borrowed, and you can absorb ₹20 of losses before your own lenders lose anything. "Risk-weighted" means riskier loans count for more. Muthoot's capital to risk-weighted assets ratio (CRAR) was 19.81%, of which Tier 1 capital (pure equity) was 19.04%, comfortably above the RBI's 15% minimum for NBFCs. Fast growth uses capital up: Manappuram's CRAR fell from 30.9% to 21.3% as its loan book grew.

What a gold crash does to the lender

By the end of October 2026, international gold had fallen about 29% from its peak, from roughly $5,430 to $3,875 an ounce. The Indian price fell less, partly because import duty was raised from 6% to 15% earlier in the year. If the Indian price follows, the sequence for a lender runs roughly like this.

The cushion shrinks first. Starting from Muthoot's 65% average LTV, a 10% fall in Indian gold takes LTV to about 72% (65 ÷ 0.9), a 20% fall to about 81%, a 30% fall to about 93%, and a 35% fall to 100%, where the average loan equals its collateral.

Borrowers are then asked for more. Above about 75% LTV, lenders ask for part-repayment or extra gold, and most borrowers comply because the collateral is family jewellery. Some loans go to auction anyway. The lender only loses money when the gold sells for less than the loan plus interest.

Growth reverses automatically. On renewal, the same 20 grams supports a smaller loan, so the book shrinks with the same customers. Profit is squeezed from both ends, as income falls with the smaller book while branch and staff costs stay put and credit costs rise. And the stock is hit twice: once through lower ROE, and again through a lower P/B multiple as the market treats it as a gold proxy.

Bullet loans make the cushion thinner than it looks

Most gold loans are bullet loans: the borrower pays nothing, or only interest, during the loan, and everything falls due at the end. Pledge gold worth ₹1,00,000, borrow ₹60,000 at 19% for twelve months, pay nothing monthly, and at the end you owe ₹71,400. The LTV has crept from 60% to 71% with no change in the gold price. If gold also falls 20%, to ₹80,000, you owe 89% of its value.

So the real cushion is thinner than the reported LTV. The lender books interest as income each month even though the cash arrives only at the end, which means more overdue loans translate into more reported profit that has not been collected. Trouble tends to show up late, often only at maturity, and renewals can hide stress by rolling unpaid interest into a fresh loan.

What Muthoot's price assumes, worked through slowly

This section shows how to test what a price assumes. It is an illustration of the method, not a verdict on the stock. Three questions matter: what is a normal ROE, how fast can the book grow without help from gold, and what are you paying?

FY26 and the first half of FY27 were lifted by the gold rally. Through a cycle, 18% to 20% looks like a fair estimate of normal. Real growth, from more customers and more grams rather than a higher gold price, is perhaps 5% to 8% a year.

Now the pot arithmetic, one step at a time, at a share price of about ₹2,400.

1. Book value is the pot. Muthoot has ₹1,046 of owners' money per share working inside the business.
2. ROE is what the pot earns. At a normal 20%, that is ₹1,046 × 20% = ₹209 of profit per share a year.
3. P/B is what you pay for the pot. ₹2,400 ÷ ₹1,046 = 2.3 times book.
4. Your return depends on your price. Pay ₹1,046 and you earn 20%. Pay ₹2,400 and you earn ₹209 ÷ ₹2,400 = 8.7%, before growth.
5. Growth fills the gap. To grow 6%, the pot needs ₹1,046 × 6% = ₹63 kept back each year. The other ₹146 could come to you: ₹146 ÷ ₹2,400 = 6.1% in cash, plus 6% growth, gives about 12% a year.

At ₹2,400 the price assumes a normal ROE of about 20% and long-run growth of about 6%. On those assumptions the price and the business roughly line up; change the assumptions and the answer changes. The same logic collapses into one formula for the P/B that delivers a required return:

Then the bad case. Suppose a gold crash holds ROE at 14% for several years. Profit is ₹1,046 × 14% = ₹146. Growing 5% needs ₹52 kept back, leaving ₹94 for shareholders. For a 12% return made of 5% growth and 7% cash, ₹94 must be 7% of the price you pay: ₹94 ÷ 7% ≈ ₹1,340, or about 1.3 times book. That is roughly 44% below ₹2,400.

None of this tells you what to do with the stock. What the exercise does show is the kind of thinking any analysis should end with: list what must be true for the numbers to hold (here, a normal ROE near 20%, no lasting gold collapse, and Stage 2 clearing), note which of those the next few quarters will reveal, and see how far the answer moves when one of them fails. Those are general principles of analysis, not suggestions about Muthoot.

When the profit stays inside the company

All of this assumed profit not needed for growth comes back to you. What if the company keeps it?

If the company can reinvest at its ROE, keeping profit is the best outcome. The pot grows 20% a year, from ₹1,046 to about ₹2,600 in five years, and at a steady P/B the share price grows 20% a year too. Your return arrives as price growth, which in India is also taxed more lightly than dividends.

If it cannot, the picture changes. Say the loan book can only grow 6%. Each year about 6% of the lending equity goes into new loans at 20%, and the rest of the profit sits as cash earning perhaps 5% after tax. We modelled five years of that.

The lending business is exactly as good as before, but the idle cash drags ROE down, and a lower ROE means a lower fair P/B. No dividend combined with a falling multiple is the worst pairing a shareholder can get. A well-run company with spare cash raises its dividend, buys back shares, or expands into businesses that also earn high returns. Muthoot pays about ₹30 of its roughly ₹264 of annual profit per share as dividends. If its growth slows to single digits, watch whether it pays out more or keeps piling up cash.

A one-page checklist for any lender

When you next open a bank or NBFC on a stock screener, work through these in order.

1. Ignore the wrong ratios. Skip ROCE, debt-to-equity and operating margin. Check that the ROE shown reflects the latest quarters.
2. Split ROE into ROA and leverage. A high ROE from high leverage is more fragile than one from high ROA.
3. Check loan quality. Gross and net NPA (or Stage 3), the Stage 2 trend, credit cost, and coverage judged against the collateral.
4. Check funding. CASA ratio for banks; maturity matching, commercial paper reliance and cash buffer for NBFCs.
5. Check capital. CRAR well above the minimum, and how fast growth is using it up.
6. Estimate a normal ROE. Average it across a full cycle; do not use a boom year.
7. Price it. Starting return ≈ ROE ÷ P/B, and fair P/B ≈ (ROE − g) ÷ (r − g).
8. Run the bad case. Lower ROE and a lower multiple together, and see what return the price implies then.

Try it yourself

Exercise 1. Pick any private bank you know. Find its ROE and P/B on screener.in and calculate ROE ÷ P/B. Then look up its five-year ROE history. Is the current ROE above or below its average, and how would your starting return change if ROE moved back to the average?

Exercise 2. Using the fair P/B formula, work out what P/B a bank with a 16% ROE and 10% growth deserves if you want 13% a year. (Answer: (16 − 10) ÷ (13 − 10) = 2.0 times book.) Now drop the growth to 7% and recalculate. Notice how sensitive the answer is to growth when ROE is only modestly above your required return.

Exercise 3. Open the latest investor presentation of Muthoot Finance or Manappuram Finance. Find the tonnes of gold held and the gold price for the last five quarters, and split loan growth into tonnes, price and lending per gram, the same way we did above. Then check whether Stage 2 has cured or rolled into Stage 3 since September 2026.

Key takeaways

• Expected return ≈ dividend yield + earnings growth + change in multiple. Run it backwards to see what a price already assumes.
• ROE measures the business and the price measures the deal. Your starting return is roughly ROE divided by price-to-book.
• For lenders, ignore ROCE and debt-to-equity. Deposits and borrowings are raw material, so use ROE, ROA, NIM, bad-loan ratios and capital adequacy.
• Leverage cuts both ways. At ten times leverage, 2% of bad loans wipes out a fifth of the owners' money, so loan quality comes first.
• A fair P/B depends on ROE, growth and your required return. A durable 20% ROE can justify 2 to 3 times book; a 9% ROE is worth less than book.
• For gold lenders, track tonnes and LTV, not just rupees lent. Most of Muthoot's 43% loan growth came from the gold price and higher lending per gram.
• Provisions shift profit in time. Use PPOP to see core earnings, and watch Stage 2 as the early warning it is.

This post is an educational analysis, not investment advice, and nothing in it is a recommendation to buy, sell or hold any security. Figures are approximate and as of late October 2026, drawn from company filings, Muthoot Finance's Q2 FY27 investor presentation and screener.in. Scenario returns depend on the stated assumptions and are not forecasts.