Business

Know the Business — Bajaj Finserv Limited

Bottom line. Bajaj Finserv is a holding company — not an operating business. Its market cap is essentially a portfolio of one listed crown jewel (Bajaj Finance, 51.32% owned by BFS direct per Q4 FY26 PPT page 4), one listed mortgage spin-out (Bajaj Housing Finance, ~45% effective via BFL's 86.70%), two unlisted insurance franchises that turned 100% Bajaj-group-owned post the March 2026 Allianz buyback (BAGIC and BALIC; 77.33% held by listed BFS, the balance by promoter holding entities), and a small basket of capital-light option-value businesses (AMC, broking, health, fintech marketplace). The single most important shift in a decade just happened — the 24-year Allianz JV ended in March 2026 — and the market still values BFS like an Allianz-encumbered holdco. The right way to value this company is sum-of-the-parts, not P/E; the consolidated ROE of 13.2% is a blend, not a signal.

1. How This Business Actually Works

BFS itself owns ~138 windmills in Maharashtra. Everything else on the income statement is the consolidated result of five different financial businesses, with three different regulators (RBI for the NBFCs, IRDAI for the insurers, SEBI for the AMC/broking), three different profit engines (spread, combined-ratio gap + float, fee bps), and three different sensitivities to interest rates.

The fundamental mechanic of the holdco — and the one investors most often miss — is that the consolidated revenue line of ₹1,50,504 crore in FY2026 is not a single revenue stream. About ₹1,00,000 crore of that is interest income on BFL+BHFL's loan books (which carries ₹28,232 crore of interest expense against it), another ~₹35,000 crore is gross premium plus investment income from the insurance subsidiaries, and a few hundred crore is fee/marketplace revenue. Looking at consolidated operating margin (38%) tells you almost nothing about how any of the underlying businesses are performing.

Notice the cycles don't line up. BFL's ROE held above 19% even when BAGIC dropped to 11% (FY24, motor-TP pricing war). BALIC's VNB margin compressed in FY24 while BFL's ROE was at a peak. This is the value of being a holdco — and also the reason consolidated P/E is meaningless. Each line is a different business with a different cycle.

2. The Playing Field

There is no clean peer for BFS because there is only one other listed Indian diversified financial holdco — Aditya Birla Capital (ABCAPITAL). The honest peer comparison therefore unbundles BFS by pillar and benchmarks each subsidiary against its pure-play listed analogue: CHOLAFIN/SHRIRAMFIN for the NBFC, ICICIGI for general insurance, SBILIFE for life insurance, and HDFCBANK/ICICIBANK as the scale-and-cross-sell yardstick.

What this peer set reveals — three things, in order of importance.

First, BFS is priced like a holdco discount to its own subsidiary: at 3.55× book and 27.8× earnings, it trades at a lower multiple than its NBFC child Bajaj Finance (5.62× book, 29.7× P/E) despite owning ~52% of it plus three other meaningful franchises. The implied SOTP discount is in the 10–15% range and has been roughly stable for a decade — narrow enough that buying BFS is not a "cheap proxy for BFL" but expensive enough that buying BFL outright is the cleaner trade if you only want lending.

Second, "good" in each pillar is a different number. The best NBFC peer comparison gives BFL 19% ROE on 4-5% ROA — and BFL is at that frontier (FY26 print 18.1%). The best general-insurance peer (ICICIGI) earns 17.8% ROE on the strength of float, and BAGIC's FY26 ROE annualised at 200% solvency is estimated at ~18.5% per BFS PPT page 68, with a materially better combined ratio than the broader industry average. The best life-insurance peer (SBILIFE) trades at 76× P/E, 9.8× P/B because Indian GAAP under-counts life insurer profit — the entire SBILIFE valuation thesis is that consolidated earnings are an accounting artefact. BALIC's reported PAT is therefore deeply misleading; only VNB margin (19.2%) and embedded value tell the real story.

Third, the closest direct peer — Aditya Birla Capital — is the cautionary tale. ABCAPITAL has the same structural mix (NBFC + housing + life + general + AMC) but earns 11.7% ROE versus BFS' 13.2%, and trades at 2.7× book vs BFS' 3.6×. The difference between the two holdcos is not the structure; it's that BFS owns the #1 NBFC in India and ABCAPITAL owns a mid-pack one. The quality of the largest subsidiary drives the entire holdco's rating.

3. Is This Business Cyclical?

Yes, but not uniformly — and the cycles partially offset each other. That offset is the actual reason this holdco structure was kept together. The lending engine (BFL+BHFL) is the most cycle-exposed; insurance is regulation-cycle exposed; AMC is market-cycle exposed.

The chart tells the most important story about cyclicality: consolidated PAT has compounded at ~22% over 8 years through a credit cycle, a pandemic, a motor-TP pricing war, and an insurance-regulation shock. The holdco structure absorbs these — when one pillar takes the hit, another carries the quarter. Q4 FY26 is a textbook case: BAGIC's combined ratio elevated to 113.6% on government-health timing and a deliberate crop pullback, but BALIC's VNB grew 29% and BFL's PAT crossed ₹5,464 crore. Consolidated PAT was flat — but the underlying mix shifted, not deteriorated. This is what diversified financial-services holdcos are supposed to do; very few in India actually deliver it.

4. The Metrics That Actually Matter

P/E and EV/EBITDA are useless here. The five metrics below are the ones that drive sub-valuations and therefore drive the SOTP that drives the stock.

Operating-metric scorecard by pillar (higher score = stronger; 100 = best-in-class).

BFL and BHFL screen as best-in-class on every relevant lender metric. BAGIC dominates general insurance on COR and solvency. BALIC is improving on VNB margin but persistency is the soft spot. AMC is the venture line — growing fast but still scaling toward breakeven.

Beyond these five, three professional-investor metrics deserve special mention:

• VNB margin is to a life insurer what ROIC is to an industrial. It is the single number that captures pricing power, product mix, persistency assumption, and distribution cost in one figure. BALIC's jump from 14.5% (FY25) to 19.2% (FY26) is the most consequential operating-metric change in the entire BFS group this year — it implies a step-change in embedded-value compounding.
• Combined ratio gap to industry, not absolute COR, is the moat metric for general insurance. BAGIC's 102.8% COR sounds unremarkable until you see top-5 private peers averaged 112.2% in FY25 (BFS Q4 FY26 PPT page 21) and the broader industry including PSU insurers ran ~109% in FY26. The structural gap is what generates a 7.2% market share for an insurer that does not own a banking channel.
• For BFL, watch credit cost as a leading indicator, not GNPA as a lagging one. FY24's credit cost step-up (1.5% → 2.1%) predicted the FY25 PAT moderation; the Q4 FY26 print of falling Stage-2/3 assets (-₹430 crore) is the leading signal that the cycle has turned back up.

5. What Is This Business Worth?

The only honest valuation lens for BFS is sum-of-the-parts. Consolidated P/E, P/B, ROE — all are blends of unlike parts. The four meaningful pieces are: (a) BFS' 51.32% stake in listed Bajaj Finance, (b) BFS' indirect ~45% effective stake in listed Bajaj Housing Finance (via BFL's 86.70%), (c) the 77.33% direct stake in BAGIC (100% Bajaj-group post Mar-26 Allianz exit, balance held by promoter entities), and (d) the 77.33% direct stake in BALIC on the same basis. AMC + Health + Direct + Securities + windmills together are less than 5% of group value and should be treated as option upside, not anchor value.

Three judgement calls govern whether the stock is cheap or expensive.

First, what multiple to put on BAGIC and BALIC now that Allianz is out. Before the buyback, the JV friction (Allianz had veto rights on dividends, capital, IPO) justified a ~20% private-business discount. After the March 2026 buyback, BFS controls 100% of the dividend pull and the IPO timing. The discount should compress — and that mathematical compression alone, applied to BAGIC and BALIC combined, is worth roughly 10-15% of BFS' current market cap. Whether the market re-rates this is the single biggest near-term catalyst question.

Second, whether BFL goes the universal-bank route. The RBI's Scale-Based Regulation makes BFL an "Upper Layer" NBFC, with bank-like compliance and capital requirements. The market periodically debates whether BFL applies for a banking licence; a conversion would unlock deposit funding (cheaper than bond-market) but require regulatory dilution and a smaller dividend stream. The current consensus is "no banking licence in this cycle" — but if that changes, BFL's multiple re-rates and so does BFS'.

Third, persistency at BALIC. Q4 FY26 explicitly flagged 13-month persistency "dips against certain cohorts in line with the market." The "Bajaj Life 2.0" thesis (VNB margin rising to 19%+) is now mostly priced in; the offsetting risk that persistency drops toward 79% would compress embedded value enough to negate two years of margin gains. This is the most-likely surprise that would break the BALIC contribution to SOTP.

This section does not give a target price. Anyone presenting a precise SOTP for BFS is anchoring on assumptions — discount rates for two unlisted insurance subsidiaries, terminal margins, and a holdco discount — that have wide error bars. What it does identify is which inputs matter: subsidiary quality, not corporate cost-cutting; insurance margin expansion, not lending volume; the Allianz-exit re-rate, not next-quarter PAT.

6. What I'd Tell a Young Analyst

Three things the market is most likely to be wrong about, today.

The first is the Allianz-exit re-rate has not yet happened in the multiple. The buyback closed in March 2026; the JV friction it removes — Allianz veto over dividends, capital, IPO — is now structurally gone. BFS spent a generation as "a Bajaj-Allianz JV holdco"; it is now a "Bajaj-only holdco with a listed NBFC subsidiary." That repositioning hasn't fully filtered into the multiple.

The second is BALIC is no longer the weak child. Five years ago BALIC was the bottom of the private-insurance pack — sub-15% VNB margin, sub-30% 61-month persistency, sub-scale. FY26 prints (19.2% VNB margin, 54%+ 61-month persistency, 8.3% private share) put it in the second tier with a credible path to the first tier. SBILIFE and HDFC Life trade at 9-10× P/B because life insurance is a 20-year compound; BALIC's marked-to-similar-trajectory is not a stretch.

The third is BFL is the engine, not the entirety. About half of BFS' value is BFL; the other half is BHFL + BAGIC + BALIC + emerging. Investors who buy BFS as "a way to own BFL at a discount" are misreading the structure — they're buying three insurance and HFC franchises bundled in, and they should look at the fundamentals of those too.

What would change the underwriting: (a) BFL credit cost above 2.5% on a sustained basis, with stage-2 assets building rather than running off — a genuine cycle turn; (b) BAGIC like-for-like combined ratio above 108% for two consecutive quarters on competitive-pricing rather than timing or bulk-line reasons — a moat-erosion signal; (c) BALIC 13-month persistency below 78% on a rolling 12-month basis — an embedded-value re-rate trigger. None is signalled in the current prints. All are observable in the BFS investor presentation and IRDAI monthly data, every quarter.