Financials

Financials — reading a holding company, not a business

Figures converted from Indian rupees at historical FX rates — see data/company.json.fx_rates for the rate table (fiscal years mapped to the nearest available period-end rate). Ratios, margins, multiples, share counts and percentages are unitless and unchanged.

Edelweiss Financial Services (EFSL) is not one company you can read off a single income statement. It is a holding company that consolidates seven very different regulated businesses — alternatives asset management (EAAA), a mutual fund, an asset reconstruction company (EARC), an NBFC and a housing-finance lender, plus life and general insurance — each with its own capital, its own regulator, and its own economics. The consolidated income statement bolts spread income, asset-management fees, ARC recovery income, insurance premiums and holding-company treasury together into one line. The job of this page is to pull them back apart, because the headline numbers and the underlying numbers tell two different stories.

The 30-second version: reported FY2026 (year to March 2026) attributable profit jumped 37%, from $47 million to $61 million — yet the seven operating businesses earned less profit than the year before ($58 million vs $66 million). The entire increase came from the holding-company "Corporate" line swinging from a $4 million loss to an $18 million profit [1]. That is the single most important fact on this page. Growth is real at the franchise level (alternatives and the mutual fund are compounding fast), but the reported earnings recovery is being flattered by holding-company gains, deferred-tax credits and a smaller equity base — and the stock trades at ~2.5x book and ~21x attributable earnings, multiples that only make sense if you believe the sum-of-the-parts value unlock (above all, the EAAA listing) is about to be realized.

---

1. The shape of the business: where profit actually comes from

Before any ratio, understand the mix. EFSL's consolidated total income of $1.21 billion in FY2026 is a blend, not a spread book. Net interest income (the classic bank/NBFC lens) is almost immaterial here — interest earned of ~$308 million barely exceeds interest expense of ~$277 million, because the lending book has been deliberately shrunk. Profit is dominated instead by fee and recovery income from asset management and the ARC, with insurance still a drag.

The cleanest way to see this is management's own earnings-distribution table, which assigns profit after tax to each business:

Source: Q4/FY2026 earnings presentation, Earnings distribution across businesses [2].

Two franchises — the ARC ($39 million) and alternatives ($30 million) — produce essentially all of the group's operating profit. The mutual fund ($9 million) is small but compounding fast. The two lenders that the market thinks of as "the NBFC" are now tiny: NBFC PAT collapsed to $1.6 million from $6.4 million, and housing finance is just $2.6 million — these are run-off books, not growth engines. And insurance lost a combined $24 million [3]. This is a fee-and-recovery story wearing an NBFC's old clothes.

Sources: total income derived from reported financials [4]; attributable PAT, customer assets ($26.7 billion) and liquidity ($723 million) from the Q4/FY2026 presentation [5][6].

---

2. The year-wise statements — and the gap between reported and operating profit

Here is the standard multi-year scorecard. Because EFSL is a financial holding company, the lines that matter are total income, finance costs, pre-tax profit, consolidated profit before and after minority interest, book value per share, and returns — not gross margin.

Source: income statement, balance sheet and EPS as reported in exchange financial filings; FY2025 figures reconciled to the FY2025 Annual Report management discussion [7]. Equity attributable to owners and book value per share derived from reported financials.

Three things jump out, and each is a teaching point:

(a) Pre-tax profit doubled, then stalled. PBT went from $47 million (FY2023) to $94 million (FY2025) and $90 million (FY2026). Most of that step-up came from the cost side — finance costs fell from $309 million to $282 million as the wholesale loan book ran off, and total income held roughly flat at ~$1.1 billion [8]. This is deleveraging-driven profit, not revenue-driven profit.

(b) The reported PAT growth is lower-quality than it looks. Watch the bridge from operating profit to reported profit:

Source: Q4/FY2026 earnings presentation, Earnings distribution across businesses [9]. Note: the FY2026 figure embeds $16 million of exceptional ESOP, labour-code and GST charges inside operating PAT.

Operating-business PAT fell $8 million year-on-year (66 → 58). The reported pre-MI PAT nonetheless rose $13 million (63 → 76). The reconciling item is the Corporate (holding-company) line, which swung from −$4 million to +$18 million — a $22 million swing that more than accounts for the entire group improvement [10]. The Corporate line is where holding-company treasury gains and the fair-value impact of stake monetizations (such as the EAAA pre-IPO placements) land. That is high-quality cash if the stake sales are real, but it is not recurring operating earnings, and an investor paying an earnings multiple should strip it out.

(c) "Tax" repeatedly flatters the number. Look at the effective tax line across years: FY2023 and FY2024 carried net tax credits (deferred-tax write-backs of $224 million and $367 million respectively), so net income exceeded pre-tax income in those years. The quarterly record is even starker — in two of the four quarters of FY2026 the group reported negative pre-tax profit yet positive net profit, because deferred-tax credits ($30 million in Q2, $20 million in Q4) more than erased the pre-tax loss. Deferred-tax timing is a legitimate accounting outcome, but a profit that depends on it is a profit you should discount.

---

3. Earnings quality: does profit become cash? (and what "cash" means for a lender)

For an operating company, free cash flow is the acid test. For a holding company of lenders, operating cash flow is dominated by changes in the loan book — when a lender shrinks its book, cash floods in (loans repaid are not reinvested), and when it grows, cash drains out. So EFSL's high "cash conversion" is partly a symptom of contraction, not a sign of a cash-printing machine.

Source: cash-flow statement as reported in exchange financial filings; the strategic wind-down of the wholesale book is described in the FY2025 Annual Report [11].

Operating cash flow ran far above net profit in FY2023–FY2025 (e.g. $348 million of OCF vs $64 million of profit in FY2024) precisely because the wholesale loan book was being liquidated — from $1.82 billion in FY2020 to $292 million by FY2025 [12]. The flip side: in FY2026 OCF collapsed to $100 million, down 56%, as the easy run-off cash dried up. Read positively, the balance sheet is now closer to its steady state; read skeptically, the cash tailwind from shrinking is largely spent, and future cash generation must come from the fee and recovery businesses doing more work. Either way, do not treat EFSL's reported FCF as classic owner-earnings — it is heavily a function of book size.

---

4. Balance-sheet resilience: a deleveraged holdco of well-capitalised subsidiaries

This is the genuinely strong part of the story, and it deserves credit. EFSL has spent five years deliberately taking risk out of the balance sheet.

• Net debt fell from $3.87 billion (FY2020) to $1.31 billion (FY2025) — a more-than-halving — funded by the Nuvama (wealth) value-unlock and serial stake sales [13].
• Corporate (holding-company) net debt — the most important leverage figure, because the subsidiaries are ring-fenced — fell ~20% over two years, from $940 million (Mar-2024) to $713 million (Mar-2026) [14].
• The operating subsidiaries are over-capitalised by Indian standards: NBFC capital adequacy 30%, housing finance 29%, ARC 80%, with life-insurance solvency at 176% and general insurance at 157% [15].
• Consolidated liquidity stands at $723 million, described as covering the next year of obligations [16].

Source: Q4/FY2026 earnings presentation, Corporate net debt declined by 20 percent over 2 years [17].

The caveat to the strength: book value has shrunk, not compounded. Equity attributable to owners fell from $816 million (FY2023) to $515 million (FY2026), and book value per share fell from $0.91 to $0.54. Much of the FY2023→FY2024 step-down reflects the deconsolidation of the wealth business (Nuvama) — that was value realised, not destroyed, and reported net worth was $1.03 billion at the FY2023 peak [18]. But since FY2024 the owners' equity base has been roughly flat at $515–572 million, which means the improving ROE optics are partly arithmetic: the same modest profit divided by a smaller, non-growing equity base. A genuinely compounding financial holds or grows book value per share; EFSL has not yet shown that.

---

5. Returns and capital allocation — improving, but off a low base

Source: ROA and attributable ROE derived from reported financials (attributable PAT ÷ owners' equity; net income ÷ total assets).

Consolidated ROA sits at just ~1.6% and attributable ROE at ~11.8% in FY2026 — respectable for a deleveraging holdco but well short of the 18–20%+ that the best Indian wealth/alternatives franchises earn. The trajectory is up, but as noted above, part of the ROE improvement is the shrinking-denominator effect.

On capital allocation, management has been consistent and credible about its deleveraging-and-unlock framework: "continue to further reduce corporate debt aided by stake sales in our underlying businesses," with EAAA's planned listing described as the next milestone [19]. Dividends are modest ($53 million paid in FY2026). The capital story is therefore not "compound retained earnings at high ROIC" — it is "monetise minority stakes in the crown-jewel businesses, pay down holdco debt, and surface value through listings." That is a perfectly valid model — but it makes EFSL a value-realization stock, where the catalysts (not the run-rate income statement) drive the return.

---

6. The crown jewels: segment economics that justify the SOTP case

The bull case lives in two compounding franchises that are largely invisible in the blended consolidated numbers.

Source: Q4/FY2026 earnings presentation, Alternative Asset Management financial performance snapshot [20].

• Alternatives (EAAA): AUM grew to $8.1 billion with fee-paying AUM of $5.0 billion, generating $30 million of PAT — a 23% two-year PAT CAGR [21][22]. This is the capital-light, high-return engine that the EAAA listing is meant to crystallize.
• Mutual fund: AUM reached $16.6 billion, up 12%, with equity AUM up 43% to $7.3 billion and PAT up 40% — a 50% two-year PAT CAGR off a small base [23][24].
• ARC (EARC): still the largest single profit pool ($39 million), with fee-paying AUM of $1.4 billion and recoveries of $670 million in FY2025 [25]. This is a maturing book — profitable, but unlikely to grow given the industry's reduced stressed-asset flow.

The drag on the other side is insurance: life and general insurance lost a combined $24 million in FY2026 (it was $33 million in FY2024, so the losses are narrowing but persistent), with $8 million of one-off charges inside the FY2026 figure [26]. Insurance ties up capital and depresses group ROE today; its value is in embedded value that the consolidated income statement does not show.

---

7. Valuation — cheap on parts, not on the consolidated income statement

At $1.30 a share, EFSL carries a market capitalisation of ~$1.23 billion. Put that against the financials:

Source: P/B and P/E derived from the latest share price and reported equity and attributable profit; ROE and ROA derived from reported financials.

A 2.5x book multiple on an ~12% ROE business is demanding — by the standard P/B-vs-ROE yardstick a sub-12% ROE financial would normally trade below book, not at 2.5x. And ~21x attributable earnings is a full multiple for profits whose growth this year came from the Corporate line and tax. On a pure consolidated read, EFSL looks expensive, not cheap.

The bull rebuttal — and it is a serious one — is that the consolidated P/B is the wrong lens because it values a 100%-consolidated insurance loss-maker and run-off lender alongside the crown jewels at the same multiple. The market is implicitly paying up for a sum-of-the-parts unlock: a listed EAAA alternatives manager comparable to pure-play peers, plus optionality on insurance embedded value and the ARC. The read-through is the peer set:

Source: market capitalisations from exchange data as reported (converted at the 27 Jun 2026 INR/USD rate); peer net-worth anchors below cited to peer filings.

Two anchors frame the gap. 360 ONE WAM — the cleanest comparator for what a standalone listed EAAA could look like (pure wealth-plus-alternatives) — commands a $5.0 billion market cap, roughly 4x EFSL's entire group value, on a far higher ROE. That premium for a clean, capital-light alternatives franchise is exactly the value the EFSL bull expects EAAA's listing to surface. On the other side, larger diversified peers carry far bigger equity bases — Aditya Birla Capital reported standalone net worth of $2.94 billion at a 4.41x debt-to-equity ratio [27], and Piramal Enterprises carried net worth of $3.15 billion [28] — both several times EFSL's $515 million owners' equity, underlining that EFSL is a small, concentrated holdco whose value is bet on a handful of franchises rather than on scale.

The crux catalyst is explicit: management has been guiding to an EAAA listing around April 2026, positioning it as a standalone, institutionalised alternatives platform [29], with the most recent FY2026 update reframing the immediate step as a pre-listing placement. The valuation EAAA achieves is the single number that validates or breaks the 2.5x book multiple.

---

8. The verdict and the metric to watch

What the financials confirm: the balance sheet is genuinely de-risked (net debt halved, subsidiaries over-capitalised, $723 million of liquidity), two franchises — alternatives and the mutual fund — are compounding profit at 20–50%, and the ARC throws off real cash. The deleveraging-and-unlock strategy has been executed credibly for five years.

What the financials contradict: the quality of the FY2026 earnings recovery. Operating-business profit fell; the headline jump was manufactured by the Corporate line and deferred tax; consolidated ROA (~1.6%) and attributable ROE (~12%) remain mediocre; book value per share has shrunk, not compounded; and operating cash flow halved as the run-off tailwind faded. At 2.5x book and ~21x attributable earnings, the market is already paying for a value-unlock that has not yet been priced by an arm's-length transaction.

This is a sum-of-the-parts, catalyst-driven holding company, not a steady earnings compounder — and it should be underwritten as one.

The first financial metric to watch is operating-business PAT excluding the Corporate line and exceptional items. It fell from $66 million to $58 million (or $74 million pre-exceptional) in FY2026 — if that core operating profit pool reaccelerates in FY2027, the 2.5x book multiple is defensible; if it keeps drifting while the headline is propped up by holdco gains and tax, the quality gap will eventually close the valuation gap. The clinching event within that is the realised valuation of the EAAA listing versus its carrying value.