MSE had the licence for ten years. It never had the money

On 11 September 2026, Metropolitan Stock Exchange of India put out a press release. Its Electronic Bond Platform had facilitated a tokenised corporate bond issuance for IIFL Finance under SEBI's new Demat 2.0 pilot, with Trust Investment Advisors as arranger. SEBI Chairman Tuhin Kanta Pandey and RBI Governor Sanjay Malhotra had launched the initiative at Global Fintech Fest.

If you own MSE shares, this probably felt like the news you've been waiting a decade for.

Before you get excited, let's do the boring work. What is this business, how does it make money, and — the question nobody in the unlisted-share WhatsApp groups is asking — if MSE has held this licence since 2016, why is its entire operating revenue still ₹3.4 crore?

The answer turns out to be more interesting than the press release.

Say a company wants ₹100 crore. A bank loan is expensive, equity dilutes the founder. So it issues bonds — privately, to a handful of institutions.

Before 2016, this happened over the phone. The CFO would call four fund managers. One got 11.2%, another got 11.6%, and the market had no idea what the company's real cost of borrowing was. Favouritism, mispricing, zero transparency.

So SEBI built the Electronic Book Provider — a screen where these private deals have to happen in the open.

Here's the part most people get backwards: the EBP rule applies to private placements, not public issues. Public issues already have a prospectus and full disclosure. It was the private market that needed fixing.

Since SEBI's May 2025 circular, an EBP is mandatory when a private placement of debt securities, NCRPS or municipal bonds is ₹20 crore or more — down from ₹50 crore earlier. That counts single issues, shelf issues across tranches, and cumulative issuance in a financial year.

This is not a niche. Well over 90% of Indian corporate bond issuance is private placement. Public bond issues are the exception.

How it works, end to end:

1. Issuer registers on the EBP, signs an agreement, maps its arranger

2. Issue notification goes up at least two working days before bidding — three for a first-time issuer

3. Placement memorandum and term sheet get uploaded

4. Bidding opens. QIBs bid directly; arrangers bid for themselves or for clients; non-QIBs only if the issuer specifically authorises them

5. The system sorts bids by yield — cheapest money first. The issuer sets a cut-off. Same-price bids get proportionate allotment

6. Settlement through a clearing corporation or escrow, T+1 or T+2

7. The bond lists on the exchange

8. The EBP must publish the full result — 36 prescribed line items — by end of day, or 1 PM the next day

That last point is useful to you as an investor. Cut-off yields, investor counts and issue sizes sit publicly on NSE, BSE and MSE websites, free.

Where MSE earns: onboarding and annual fees, a book-running fee scaled to issue size, listing fees, clearing through its 95.85%-owned MCCIL, and — much later — transaction fees if those bonds trade on its RFQ platform.

Almost nothing about Part 1. That's the thing to internalise.

On 10 September 2026, SEBI and RBI launched Demat 2.0 — a regulatory sandbox pilot for tokenised corporate bonds. Three issuances have happened, totalling ₹1,025 crore:

Note the date on REC. It issued three days before the launch ceremony. The system had been running quietly.

What stayed identical: the bidding process, the participants, the ₹20 crore threshold, the two-day notification, the ISIN, the investor's demat account, the bond's legal character. SEBI has been emphatic — tokenisation does not create a new asset class, does not create a separate market segment, and does not make a bond safer.

What changed, all of it downstream of allotment:

• The bond is born as a native digital token on a private, permissioned ledger owned by the depositories

• The cash leg runs on RBI's wholesale CBDC, linked via the Unified Market Interface

• Both legs settle atomically — simultaneously, or not at all. Issuers get money the same day instead of T+2

• Coupons land in bondholders' CBDC wallets on the due date; every authorised institution sees the same record at once

• Smart contracts will eventually automate coupons and redemptions

What hasn't happened: secondary trading (phase two), retail access (phase three, no date announced), or graduation out of the sandbox. Twenty-three investors participated across all three issues. That is the entire universe so far.

And the detail that matters most for you as a shareholder:

The ledger belongs to the depositories. The cash rail belongs to RBI. The exchange contributes the bidding storefront.

Demat 2.0 handed MSE a better back office. It handed NSE and BSE the same one.

Yes, in the plumbing sense. A ₹100 crore issue routed through MSE generates a registration fee, a book-running fee, a listing fee, and clearing revenue through MCCIL. Small per deal. Recurring across the bond's life. Pure platform economics — MSE takes no credit risk, it rents out a screen and a settlement pipe.

Scale matters more than any single deal. Per ICRA, FY2025 saw roughly 4,650 debt issuances from about 1,250 unique issuers, averaging ₹234 crore. That's the pool. Every single one of those above ₹20 crore had to land on somebody's EBP.

Almost none of them landed on MSE's.

MSE has been a registered EBP since the framework began. BSE and NSE both went live on 1 July 2016.

Ten years. FY26 operating revenue: ₹3.4 crore.

Within two months of launching in 2016, NSE had already done 33 issues worth ₹25,785 crore and was publicly claiming about 75% market share. Today it holds roughly 95% of the debt RFQ platform market.

So what went wrong?

It was never the licence. MSE always had permission. Permission was never the bottleneck. This is the single most important sentence in this article, because it reframes everything that follows.

Bond issuance runs on relationships. An issuer picks an EBP because its arranger — Trust, ICICI Securities, A.K. Capital — is comfortable there, because the ops team answers the phone at 9 PM when a bid file breaks, because the last five issues went smoothly. That network took NSE a decade to build. The network is the moat. The software is a commodity.

Nobody gets fired for choosing NSE. A treasurer routing ₹500 crore has no upside from saving ₹2 lakh in fees and enormous downside if something goes wrong.

And MSE had no money to fight with. This is the part everyone skips. A dying exchange cuts its sales and relationship teams first. You cannot hire six senior bond-market people, park them in Mumbai, and have them spend three years drinking coffee with arrangers when your annual revenue is ₹3.4 crore and you're burning ₹25 crore a year. The bond desk was never going to get funded when the core business was collapsing.

MSE didn't lose the bond market on technology or regulation. It lost on payroll.

Now hold that diagnosis next to what's happened on MSE's cap table.

January 2025. At its 158th board meeting on 22 January, MSE resolves to allot 119,00,00,000 equity shares of Re 1 each on a private placement basis. Four allottees, and the split is exactly equal — 29,75,00,000 shares each to Billionbrains Garage Ventures (Groww's parent), Rainmatter Investments (the Zerodha founders' fund, the only one classified as a Fund rather than a Corporate), Securocrop Securities India and Share India Securities. At ₹2 per share — Re 1 face value plus Re 1 premium — that's ₹59.5 crore apiece, ₹238 crore total. Share India, being listed, confirmed its number in a stock exchange filing: ₹59.5 crore for 4.958%, pegging MSE at ₹1,200 crore.

August 2025. A second round, and a much larger one. Board approval on 8 July, shareholders sign off on 7 August, and the private placement offer letter goes out on 10 August for ₹10,00,00,00,000 — one thousand crore — circulated to 29 offerees.

Read that list and the strategy becomes obvious. Peak XV Partners is there. So is Jainam Broking. But so are Monarch Networth, Marwadi Chandarana, KIFS International, Findoc, Excel Stock Broking, Achintya Securities, StockGro, Mansi Share & Stock Broking — and Securocrop and Share India again, doubling down from round one. Zerodha and Groww do not reappear.

This isn't a venture round with a few brokers attached. It's a broker consortium with a venture fund attached. MSE didn't sell equity to investors who wanted returns. It sold equity to people who control order flow.

Total across both rounds: ₹1,238 crore.

Now do the comparison that matters.

MSE now has roughly 365 times its annual operating revenue sitting in the bank. At the FY26 burn rate, that's decades of runway.

Suddenly the constraint from Part 4 evaporates. Hiring six senior bond-market professionals costs maybe ₹10-15 crore a year — about 1% of the war chest. Building a relationship desk that calls every one of those 1,250 unique issuers is now a rounding error, not a bet-the-company decision.

You can undercut NSE on fees for three straight years and not feel it. You can eat losses on a bond desk while it builds a book, because the equity relaunch is what the money was actually raised for and the bond desk is riding along for free.

This is the real story, and it isn't the one in the press release. The tokenisation news matters far less than the balance sheet news. Demat 2.0 gave MSE a shiny back-end that NSE and BSE got simultaneously. The ₹1,238 crore gave MSE something neither of them handed over: the ability to finally fund the thing it was never able to fund.

Two ₹59.5 crore cheques are not large for either firm. They didn't write them for the returns.

They wrote them because they are the distribution.

Zerodha has roughly 6.5 million active investors. Groww counts around 13 million. Between them they sit on a substantial share of India's active retail brokerage accounts. Add Share India and Securocrop, and MSE's register now includes people who control order flow rather than just capital.

Why does that matter for the bond business specifically?

Phase three of Demat 2.0 is retail access. SEBI has said it, though no date is attached. And tokenisation makes fractionalisation technically trivial — when a bond is a digital token, splitting it into small tickets is easy, which is precisely how a ₹1 crore institutional lot becomes something a retail investor can buy.

Now ask: when retail corporate bonds finally arrive at scale, who actually reaches the retail investor?

Not the exchange. The broker. And two of India's largest brokers are already on MSE's shareholder register.

There's a second, quieter benefit. When an arranger evaluates whether to route an issue through MSE, the question in their head is "will this exchange still exist in three years?" Ten years of losses made that a fair question. Names like Rainmatter, Groww's parent and Peak XV on the cap table change the answer. Credibility is an input to relationship-building, and for the first time MSE has some.

Go back to the August 2025 offer list. Twenty-nine names. Most of them are brokers you'd expect.

Number fourteen is Trust Investment Advisors Private Limited, Bandra Kurla Complex, Mumbai.

That name should stop you, because you have already read it in this article.

Trust Investment Advisors is the arranger on the IIFL tokenised bond issue — the deal in MSE's press release, the one that started this whole conversation. MSE offered equity to one of India's larger debt arrangers in August 2025. In September 2026, that same arranger routed MSE its first tokenised corporate bond.

This is the entire thesis of Part 4, rendered as a single line item in a regulatory filing.

MSE didn't win that deal on fees, or on settlement speed, or because its DLT stack was better than NSE's — it wasn't, they got the same one. It won because somebody in the bond market had a reason to pick up the phone.

That's what the money was for. Not a data centre, not marketing, not even the equity relaunch everyone is watching. It was for buying reasons to pick up the phone.

And this is the most encouraging signal in the entire story — far more so than the tokenisation, which is shared infrastructure nobody owns exclusively. It is also, read honestly, the smallest possible version of that signal: one arranger, one deal, ₹25 crore, one investor. The mechanism is proven to work exactly once.

Now the other side, because a shareholder deserves both.

1. Money hires people. It does not buy ten years of trust. A treasurer who has routed every issue through NSE since 2018 doesn't switch because MSE hired a good VP. That takes years of flawless execution on issues nobody will ever write a press release about.

2. The bond desk may never be the priority. Read MSE's own language on the fundraise — deepen liquidity in the equity cash segment, then derivatives. That is where the ₹1,240 crore is pointed. Zerodha and Groww invested because they want a third venue for equity and F&O order flow, not because they're excited about corporate bond book-building. The bond business could easily stay an afterthought inside its own owner's strategy.

3. Demat 2.0 is still a sandbox pilot. Not production. Twenty-three investors. No secondary market. No retail. No date for either. Building a business case on phase three is building on a roadmap, not a rail.

4. The dilution is real. Share count went from 481 crore to 600 crore to 1,099 crore between FY24 and FY26. The capital is genuinely transformative for the company. Your per-share claim on it is a different question entirely.

And NSE is not standing still. It got the same tokenisation infrastructure on the same day, it has 95% of the debt RFQ market, and it can match any fee MSE offers out of petty cash.

Stop reading press releases. Start counting.

Every EBP must publish issue-level results publicly. Open NSE's EBP portal and MSE's side by side. Over the next two or three quarters:

• How many issues land on MSE versus NSE and BSE? This is the only number that matters. Everything else is narrative

• Does MSE hire visibly into debt capital markets? Senior bond-market hires showing up on LinkedIn would be the earliest real signal that the money is being pointed here at all

• Does the pilot exit the sandbox, and does phase two arrive? Secondary trading is where recurring transaction revenue lives. One-time issuance fees are the smaller half

• Does the operating revenue line move in the quarterlies? Not total revenue. FY26's headline rose mainly on ₹55.7 crore of other income from the fundraise. Operating revenue is the honest number, and it fell

Here's the fair summary. For ten years MSE had the licence and no money, which is why nothing happened. It now has the licence and a great deal of money, which is the first time both conditions have been true simultaneously.

That is a genuinely better position than it has ever been in. It is not the same thing as a turnaround.

A ₹1,025 crore pilot sounds impressive. MSE's slice was ₹25 crore — 2.4%, one investor. Six months from now, if MSE is running twenty issues a quarter, something real has started. If it's still doing the occasional ₹25 crore deal, then the money bought a data centre and a press release.

This is an explainer, not investment advice. MSE shares trade in the unlisted market, where pricing is opaque, spreads are wide and SEBI has itself cautioned about unregulated platforms. Fundraise details are taken from MSE's own filings — the board resolution of 22 January 2025 and the Form PAS-5 dated 13 August 2025. Note that the board resolution records face value only (₹29.75 crore per allottee); the ₹59.5 crore per-investor figure derives from the ₹2 issue price confirmed in Share India's stock exchange filing. Verify current figures from MSE's filings and SEBI circulars before acting on anything here.