CDSL — half of a regulated duopoly, riding (and waiting out) the market's mood

The Pulse

CDSL is one of only two companies licensed to hold Indians’ shares in electronic form — a regulated duopoly with NSDL, and the one that dominates the retail end of the market with roughly 80% of the country’s 22-plus-crore demat accounts. It is the plumbing of the stock market: when you buy a share, CDSL is where it sits. The economics are lovely — capital-light, ~32% return on capital, more than half of profit paid out as dividend — but FY26 was a reminder that the plumbing’s volume ebbs and flows with the market’s mood. Consolidated profit actually fell ~13% to ₹455 crore as new-account growth cooled, IPOs thinned, the KYC subsidiary’s earnings halved, and the regulator cut KYC fees. The moat is genuine and the long runway (India is still barely invested) is real. The sticking point is the price: even after falling from ₹1,814 to ₹1,190, the stock trades at ~54x earnings on a year when profits went backwards.

The Business

A depository is the electronic vault of the securities market. When shares, bonds or mutual-fund units are “dematerialised,” they live as records inside CDSL (or NSDL), and every transfer, pledge or corporate action runs through that ledger. CDSL earns money several ways: an annual fee from every company whose securities it holds (a sticky annuity), a small charge on each transaction, fees from IPOs and corporate actions (cyclical, tied to market activity), KYC services through its CVL subsidiary, and a growing line in demat of unlisted company shares and data services. The blend is part annuity, part market-linked — which means CDSL compounds beautifully in a bull market and visibly stalls when activity cools, as it did in FY26 (combined market turnover down, IPO pipeline thin, account additions slowing from their post-COVID frenzy).

What makes CDSL special is structural: it is half of a regulated duopoly. India will not licence a third depository casually, the two incumbents enjoy network effects (every broker and registrar plugs into them), and switching is essentially unheard of. Within that duopoly CDSL has won the retail land-grab decisively — about 80% of all demat accounts and 85-90% of incremental new accounts, ten times more accounts than it had in 2019 — while NSDL retains the institutional and issuer-side lead. The promoter is the BSE stock exchange at 15% (capped low by the rules for market infrastructure), and the float is overwhelmingly retail — some 15 lakh small shareholders. The crucial limiter on the moat is that the same regulator that protects it also controls its pricing: SEBI must pre-approve any fee change, charges have been frozen for years, an issuer-fee hike has been “pending” for roughly a decade, and in FY26 SEBI actually cut KYC fees sharply. A moat the regulator both guards and caps.

How Management Thinks

The MD, Nehal Vora, runs the calls with a distinctive and somewhat polarising philosophy: CDSL is “infrastructure,” it builds “roads ahead of the traffic,” and margins are “a byproduct” to be judged over years, not quarters. He means it — he has explicitly rejected a shareholder’s plea to run the company as a profit-maximiser and restore margins, and instead is deliberately front-loading spending on technology and capacity ahead of demand. That long-termism is defensible for a piece of market infrastructure that must never fail. But it shades into a real transparency gap: technology spend has ballooned from ~7% to ~14% of revenue, quadrupling in three years while volumes grew far less, and management has refused every analyst request to break it down — capped by Vora’s deflection that you should “eat the fruit, why count the trees.” For a business whose margins are visibly compressing, declining to explain the single biggest cost increase is the weak point in an otherwise principled posture.

On capital allocation the approach is reasonable but unspectacular. The ~54% dividend payout fits a business that can’t reinvest most of its cash. The capacity-ahead-of-demand investment is genuinely prudent for critical infrastructure. But there has been no buyback even as the stock fell by a third, and the “margins don’t matter” stance, while principled, arguably leaves shareholder value on the table when paired with opaque, surging costs. The credibility on the core franchise is high — the duopoly position and account-share gains are real and delivered; management was also candid that prior-year profit comparisons were distorted by subsidiary dividends and accounting reclassifications. The credibility gap is specifically around cost discipline and what the tech spend is buying.

Where It’s Going

The long-term runway is the genuinely attractive part: only about 10% of Indians have a demat account, so structural account growth has years to run, and CDSL keeps taking the lion’s share of new openings. New revenue lines add optionality — demat of unlisted companies (where a level-playing-field change will let CDSL issue ISINs and erode NSDL’s old exclusivity), data services (gated on data-protection rules), an insurance repository, and a new KYC presence at the GIFT City financial hub. The annual issuer-charge fee hike, frozen for a decade, is the single biggest latent earnings lever if SEBI ever approves it — management sounded its most hopeful yet on this.

But the near-term and the tensions are all about cyclicality and regulation. FY26 showed how quickly the market-linked revenue (IPOs, transactions, KYC) softens when sentiment cools — about a year and a half of capital-market sluggishness, with investor attention drifting to commodities. The KYC subsidiary, CVL, was the visible casualty, its profit roughly halving on lower volumes and SEBI’s fee cuts. And that regulatory dependence cuts both ways: SEBI controls pricing, so CDSL cannot simply raise fees to offset a slow market, and a fee cut (as on KYC) lands straight on profit. The franchise is durable; the earnings are cyclical and the regulator holds the pricing pen.

The Four Checks

1. Quality & moat (gate) — 7/10. A genuinely strong, durable moat — a regulated duopoly with near-unobtainable licences, network effects, no meaningful switching, and ~80% of retail demat accounts. That is bona fide niche dominance. It is held back from the top tier by the defining limit on a regulated MII: the same regulator that protects the franchise also controls its pricing, so CDSL cannot fully monetise its position and can have fees cut at will — as FY26 demonstrated.

2. Returns on incremental capital & runway — 6/10. Capital-light with ~32% returns and a long structural runway (only ~10% demat penetration, with CDSL winning most new accounts) — the kind of capital-free growth the framework rewards. Tempered by genuine cyclicality (earnings swing with market activity), regulator-capped pricing, and currently-compressed incremental returns as heavy, opaque tech spend and the weak KYC arm drag on profitability.

3. Capital allocation for the stage — 6/10. The dividend (~54% payout) is appropriate, and investing in capacity ahead of demand is the right instinct for critical infrastructure. But the surging, unexplained technology spend, the explicit refusal to optimise margins, and the absence of any buyback while the stock fell a third are real quibbles. Sensible custodianship; not sharp value-maximisation.

4. Price — 3/10. Demanding. At ₹1,190 the stock trades at ~54x earnings and ~13x book — on a year in which profit actually fell. Even granting the duopoly quality and the long penetration runway, paying 54x for declining trailing earnings prices in a strong market-cycle recovery and years of uninterrupted growth. The quality is real; the multiple leaves no margin of safety.

Engine score: 19/30 (moat 7 + reinvestment 6 + allocation 6). Price 3.

Sources

• Concalls read: Q1 FY26 (call 28 Jul 2025), Q2 FY26 (3 Nov 2025), Q3 FY26 (2 Feb 2026), Q4/FY26 (4 May 2026) — cleaned BSE transcripts, the backbone of this digest (demat-account share, revenue-line splits, the KYC/CVL drag, the tech-spend debate, regulatory detail).
• Annual reports: FY22, FY24, FY25 (FY23 was unavailable; the fetch substituted FY22) — all three extracts were heavily trimmed (MD/CEO letters and strategy sections survived mostly as headers); the usable signal was the three-segment revenue tables (Depository / KYC / Repository). The qualitative read leans on the concalls.
• Snapshot: screener.in (consolidated, logged-out) fetched 2026-06-11 22:38 IST.
• Gaps flagged: FY23 annual report missing (FY22 used instead); trimmed ARs; some calls carried minor CFO figure discrepancies (noted in the digests); no dividend/buyback discussion on the calls; logged-out snapshot. Promoter BSE ~15%.
• Research dumps: vault/Sources/Earnings/Central Depository Services (India) Ltd/.