Jio Financial: a giant vault, barely earning
If you owned the whole company
Buy all of Jio Financial Services today and you pay about ₹1,58,574 crore. For that, the business earned ₹1,561 crore in the year ended March 2026. That's an earnings yield of about 1 paisa per rupee — the stock trades at a P/E of 103, meaning you are paying one hundred and three years of today's profit, upfront, in cash.
Now, before you faint, one honest complication. The book value — the owners' capital sitting inside the company — is ₹211 per share against a price of about ₹240. So you're paying only 1.14 times the money already in the vault. Screener lists that among the pros. In other words: you are not really paying 103 years of earnings for a business. You are paying roughly what's in the vault, for a vault that has so far earned almost nothing on its contents. The return on equity is 1.19% — for every ₹100 of shareholders' money, ₹1.19 of profit came back last year. A savings account sneers at that.
So this chapter is not about a moat. It is about a question: can an enormous pile of capital, a famous surname, and three years of history become a real financial business?
What does this business actually do?
Jio Financial was carved out of Reliance Industries in 2023 — the corporate equivalent of a rich father setting up a son with a very large inheritance and a shop that hasn't opened its shutters fully yet. It is a holding company and a registered NBFC (a non-banking financial company: lends like a bank, isn't one), operating through subsidiaries: Jio Finance (lending), Jio Insurance Broking (selling other people's insurance for a commission), Jio Payment Solutions, and a payments-bank joint venture.
Where does the money come from today? Mostly from the inheritance itself — investment income on a treasury of roughly ₹1,33,853 crore of equity capital and reserves — plus the first revenues of a young lending book. Sales went from ₹45 crore in March 2023 to ₹3,513 crore in March 2026, which sounds explosive until you notice the profit: ₹1,605 crore, ₹1,613 crore, ₹1,561 crore over the last three years. Revenue tripled; profit stood still, actually dipping 5% in the trailing twelve months. The shop is spending to open.
The science underneath
Charlie here, and the science today is the coldest arithmetic in finance: the arithmetic of capital and its return.
A financial business is a machine for earning a rate on a pile. Two numbers, multiplied: the size of the pile, and the rate. Jio Financial has a magnificent pile and a miserable rate. ₹1,33,853 crore of net worth earning ₹1,561 crore is 1.2%. The machine exists; it just isn't turning yet.
Why is the rate so low? Because lending profitably is not a capital problem — it is an information problem. The lender's equation is (interest charged − cost of funds − operating cost − defaults) × leverage. A newcomer suffers on three of the four terms:
- Defaults. Probability of default is learned, not bought. You discover which borrowers repay by lending to them and eating the losses of your mistakes. A seasoned rival's loan book is a graveyard of paid tuition; Jio's tuition largely lies ahead.
- Leverage. The established lender runs many rupees of loans per rupee of equity. Jio is the opposite — borrowings of ₹21,768 crore against ₹1,33,853 crore of net worth. It is, in effect, a lender running on almost no leverage, which is safe as a bank vault and about as profitable.
- Operating cost. Branches, apps, collections teams — all being built now, all charged against today's meagre revenue.
But invert the weakness and you see the bet. The parent brings the one raw material that can shortcut the information problem: distribution and data at national scale. Reliance's telecom and retail arms touch hundreds of millions of Indians. If — and it is an if — transaction and usage data can be turned into credit judgment, the newcomer skips a decade of tuition. That chain (existing customer data → cheaper customer acquisition → lower default surprises → viable spread) is the entire thesis. It is plausible. It is also, so far, unproven in this company's numbers, and we are in the proof business.
Note also the negative cash from operations — minus ₹15,439 crore in FY26. For a lender ramping up, loans pushed out the door count as cash going out; this is the sound of the machine finally starting to lend. Screener's flag that working-capital days ballooned is the same phenomenon wearing an accountant's spectacles.
The moat test
Hand a rival ₹1,58,574 crore and ten years. Could they build Jio Financial? Here the thought experiment collapses into comedy: that is exactly what Jio Financial is — a rival handed a fortune and told to take other castles. It is the attacker in everyone else's moat test.
Run the checklist honestly:
- Brand? A famous name in telecom and retail; unproven in money. Trusting a company with your phone bill and trusting it with your loan are different psychology.
- Switching costs? None yet; customers barely arrived.
- Network effects? Possible someday in payments; not demonstrated.
- Low-cost scale? The parent's distribution could become one; today the costs run ahead of the scale.
- Regulation? Licences held — NBFC registration, a payments-bank JV. Necessary tickets, shared with many rivals.
- Float? It broke insurance — as a broker, earning commission, which produces no float at all. The vault holds its own money, not other people's premiums.
The moat, today, is none. What it has instead is a war chest and a wealthy, determined parent — promoters hold 49.13% and have increased their stake by about 2% in the latest quarter, which tells you what the family thinks. But a war chest is what you use to dig a moat. It isn't one.
The numbers Warren would check
| What we check | What it means | Jio Financial |
|---|---|---|
| History available | How much proof exists | 4 years (Mar 2023–Mar 2026) |
| Sales | Revenue ramp | ₹45 cr → ₹3,513 cr |
| Net profit | The actual earnings | ₹1,561 cr (flat three years; TTM −5%) |
| Return on equity | Rate earned on the pile | 1.19% (3-yr ~1%) |
| Stock P/E | Years of profit paid upfront | 103 |
| Price to book | Price vs the vault's contents | 1.14× |
| Borrowings | Lending leverage beginning | ₹0 (FY24) → ₹21,768 cr (FY26) |
| Promoter holding | Parent's conviction | 49.13%, rising |
| Dividend payout | Cash sent home | 24% of profit in FY26 |
Ten-year growth rates simply don't exist — the columns in the data are blank, and we won't invent them. That blankness is the finding. The one-year stock price return is −26%; early shareholders paid for the dream at a higher price, and the number of shareholders still grew from about 39.8 lakh to 49.8 lakh — rarely a sign of cold-eyed analysis.
What could go wrong
Invert: what kills this business? Almost nothing — and that's the problem inverted. With 1.14× book, minimal leverage, and a parent with bottomless pockets, bankruptcy risk is negligible. The real risk is subtler: permanent mediocrity. The vault could earn 1–2% on its capital for years while rivals with real underwriting machines compound at 15–18%. For a shareholder, a decade of 1% ROE is a slow leak, not an explosion. Other honest worries:
- Lending losses ahead. The book is young; defaults show up two or three years after the loans do. The tuition bill hasn't arrived yet.
- Regulation. The RBI watches conglomerate-owned lenders closely; rules on connected lending and big-group NBFCs can tighten.
- The FII vote. Foreign institutional holding fell from 21.58% to 11.61% since September 2023 — the professionals have been quietly leaving while the public arrived.
- Conglomerate attention. The parent runs telecom, retail, and energy empires. Finance may not always be the favourite child.
What management must do to keep the castle
There is no castle yet; this memo is about building one, so it is long.
- Pick two businesses and win them; a holding company of six half-built ventures earns holding-company discounts on all six.
- Publish underwriting metrics early and honestly — default rates, cost of funds, spreads. Prove the data thesis with numbers, not adjectives.
- Grow the loan book slower than the temptation. Every reckless rupee lent now is a headline in 2029.
- Raise leverage only as fast as demonstrated credit skill — the 1% ROE must rise because the machine works, not because the gearing did.
- Keep buying the stock personally, as the promoters have — and keep reporting to minority shareholders as partners, not spectators.
- Convert distribution into earned trust: pay claims and process loans so well that the brand means money, not just data.
The verdict
Moat: none — with the best-funded moat-digging crew in India standing by. Jio Financial is not a bad business; it is barely a business at all yet — a ₹1.3-lakh-crore vault earning about 1% while it builds lending, broking, and payments operations from scratch. The honest statement is that we don't know whether the Reliance distribution machine converts into credit judgment, and neither, on the evidence of three flat years of profit, does anyone else. The price is not absurd against the vault (1.14× book) but is absurd against the earnings (103×), which is the market's way of saying it is valuing the promise, not the proof. We prefer proof. Check back when the return on equity has a digit before the decimal point that isn't a one.
Written in the style of Buffett & Munger for the Omaha Investments book project. Educational material, not investment advice. Numbers from Screener.in and live NSE data via Angel One as of the date above.