IPO Underwriting in India: What Underwriters Commit To and When They Have to Buy Unsold Shares

Every mainboard IPO prospectus has a section called “Underwriting”, and most retail investors scroll straight past it. That’s a pity, because IPO underwriting in India answers a question people rarely think to ask: if this issue doesn’t sell, who is on the hook to buy the leftover shares? The honest answer, for most book-built IPOs, is “nobody, really”, and once you see why, you’ll read IPO documents a little differently.

We’ll go through what underwriting means under SEBI’s rules, the difference between soft and hard underwriting, why SME IPOs work on a stricter rule, and then run a few worked examples. The issue sizes and share counts in the examples are made up to keep the arithmetic clean. The rules are not.

What IPO underwriting in India actually means

Start with the definition, because it’s narrower than most people assume. SEBI’s Merchant Bankers and Stock Brokers Regulations define underwriting as an agreement to subscribe to, or procure subscription for, securities that remain unsubscribed. SEBI quoted that definition in its April 2023 board memorandum on ICDR amendments. Two words in there matter: “subscribe” and “procure”. An underwriter doesn’t have to put its own money in. It can meet its obligation by finding other buyers.

The rulebook that governs this is the SEBI (Issue of Capital and Disclosure Requirements) Regulations, 2018, usually shortened to ICDR. Regulation 40 covers underwriting in an IPO. Its current text is reproduced in the “existing provisions” column of SEBI’s October 14, 2026 consultation paper on SME IPOs, and for a book-built issue it says, in plain terms:

• The issue shall be underwritten by the lead managers and syndicate members.
• For a company listing under the alternative eligibility route in regulation 6(2), at least 75% of the net offer has to be allotted to qualified institutional buyers, and that QIB portion cannot be underwritten.
• Before filing the prospectus, the issuer signs an agreement with the lead managers and syndicate members stating how many shares they will subscribe to on account of rejection of bids, at a price not below the issue price.
• If the issuer wants cover for under-subscription, it has to sign that agreement before filing the red herring prospectus and disclose the maximum number of shares the underwriters will take up.
• If syndicate members fail to meet their obligation, the lead managers have to fulfil it.
• Lead managers and syndicate members can’t subscribe to the issue in any other way except to meet their underwriting obligation.

Read the third and fourth points again. They describe two very different promises, and the market has names for both.

Soft underwriting: covering rejected bids, not weak demand

In a book-built IPO, the price is set only after bidding closes. SEBI’s 2023 memorandum explains that the underwriters (the book-running lead managers and syndicate members) compulsorily underwrite the issue after pricing is finalised but before the prospectus is filed. By that point, they already know exactly what demand looked like.

So what are they actually covering? Rejections. A bid can look valid on the last day and still fail later, for technical, procedural or legal reasons. Law firm Trilegal’s quarterly capital markets roundup for April-June 2023 describes soft underwriting as filling the deficiency in valid ASBA bids that arises solely from technical or procedural rejections, such as funds not being available. It also notes that soft underwriting is triggered only after the minimum subscription is received.

Soft underwriting insures the issuer against payment risk. It does not insure anyone against the risk that investors simply don’t want the stock.

That’s the key point. SEBI’s memorandum put it plainly: under the book-building route, underwriting essentially covers “payment risk” rather than “subscription risk”. Chambers and Partners, in an article on SEBI’s revised underwriting framework, adds that soft underwriting is the prevalent form in Indian public offerings, including book-built IPOs.

Hard underwriting: the promise to buy what investors don’t

Hard underwriting is the version most people imagine when they hear the word. The underwriter agrees, in advance, to take up shares if demand falls short. SEBI’s memorandum gives a simple illustration: an issuer offers 1,00,000 shares, gets demand for only 60,000, and the underwriters subscribe to the remaining 40,000, or whatever quantity the agreement specifies.

For book-built issues, hard underwriting is optional. Before 2023, the rules didn’t clearly separate the two types, and SEBI flagged a problem with that. In theory an issuer could sign a hard underwriting deal after the issue closed, to rescue a weak book, and investors who bid on the strength of the red herring prospectus would never have known it was coming. Since they can’t withdraw a bid after closing, that didn’t seem fair.

The fix was approved by SEBI’s board on March 29, 2023, and, as Trilegal notes, the amendment was notified on 23 May 2023. According to Chambers, the main changes were:

• ICDR now expressly recognises two types of underwriting, one for demand shortfall and one for technical rejections, and the issuer chooses which arrangement it wants.
• An agreement covering demand shortfall must be signed before the red herring prospectus is filed.
• An agreement covering technical rejections continues to be signed after pricing and before the final prospectus.
• Underwriters can meet the obligation by procuring subscribers, not only by buying themselves.
• The underwriting price can’t be lower than the issue price.

Has it caught on? Trilegal wrote that only a handful of book-built deals had been hard underwritten before the amendment. Chambers pointed out why: at the red herring stage the underwriter has no idea how large its devolvement might be, so the risk is bigger and is likely to be priced into its fees.

The 90% line that decides whether an IPO survives

Underwriting only makes sense alongside minimum subscription. The October 2026 consultation paper notes that ICDR regulation 45 prescribes a 90% minimum subscription for a mainboard IPO, with a refund of application money if that level isn’t reached. Regulation 40 also says that where an issue is required to be underwritten, the obligation should at least cover the minimum subscription.

SEBI’s 2023 memorandum spells out what happens without hard underwriting: if demand is inadequate and nobody has made a prior financial commitment, the issue is treated as failed and the issuer refunds all the application money. That’s the real consequence for you as a bidder. You don’t get stuck with shares in a half-empty IPO. You get your money unblocked.

Worked examples (hypothetical numbers)

Example 1: a soft-underwritten issue with rejections

Company A offers 50,00,000 shares through a book-built IPO. Total bids at close come to 70,00,000 shares, so the issue is 1.4 times subscribed overall. After closing, the registrar finds that bids for 80,000 shares fail because the money wasn’t available in the ASBA accounts.

Does anyone need to buy anything? Probably not. Demand is still comfortably above the shares on offer, so valid bids simply fill the place of rejected ones. Soft underwriting matters only when rejections push valid demand below what is needed, which is rare in an oversubscribed IPO.

Example 2: a weak book with no hard underwriting

Company B makes a fresh issue of 50,00,000 shares at ₹300, with no offer for sale. Bids come in for 40,00,000 shares, which is 80% of the offer. Its red herring prospectus had no hard underwriting agreement. Since 80% is below the 90% minimum subscription, the issue fails and the blocked money is released. Soft underwriters don’t step in here, because their commitment was about rejected bids, not missing ones.

Example 3: the same weak book, hard underwritten

Now suppose Company B’s red herring prospectus disclosed a hard underwriting agreement for a maximum of 8,00,000 shares at the issue price. Demand of 40,00,000 shares plus the underwriters’ 8,00,000 gives 48,00,000 shares, which is 96% of the offer. That clears the 90% line.

The underwriters’ bill is 8,00,000 shares at ₹300, or ₹24 crore. Change the demand to 35,00,000 shares and the arithmetic flips: 35,00,000 plus the 8,00,000 cap is 43,00,000, only 86% of the offer, so the issue still fails. A hard underwriting cap is a cushion, not a guarantee of success.

Example 4: an SME IPO

SME issues run on a stricter rule today. Under ICDR regulation 260, as summarised in SEBI’s October 2026 paper, an SME IPO must be 100% underwritten, not just up to the minimum subscription, and the lead managers must underwrite at least 15% of the issue size on their own account.

Take a hypothetical ₹20 crore SME issue. All ₹20 crore must be covered by underwriting commitments, and at least ₹3 crore of that (15%) sits on the lead manager’s own books. If the issue is undersubscribed and other underwriters or nominated investors don’t pay up, the lead manager has to make good the shortfall.

Why the SME rule may change soon

That 100% rule costs money. SEBI’s October 14, 2026 consultation paper says underwriting fees made up roughly 49.52% of total issue expenses for SME IPOs on NSE in FY 2024-25 and 38.41% in FY 2025-26. On BSE, the figures were 32.20% and 33.99%. The paper also points out that full underwriting can mean paying for cover even when subscription turns out to be more than adequate.

The working group’s recommendation is to drop both the 100% requirement and the lead manager’s 15% own-account obligation, apply the mainboard underwriting framework instead, and add a separate minimum subscription rule for SME issues aligned with regulation 45. Comments are open until November 4, 2026. It’s a proposal, not a rule yet. We covered the wider package, including the market-making changes, in our post on the SEBI SME IPO proposal.

For an SME bidder, full underwriting has quietly been part of the safety story. If that goes, minimum subscription becomes the main line of defence.

What to look for in the offer document

1. Find the underwriting section. In a book-built mainboard IPO, check whether the red herring prospectus mentions a hard underwriting agreement. If it only says an agreement will be signed after pricing, assume soft underwriting.
2. Check the cap. If there is hard underwriting, note the maximum number of shares the underwriters will take up and compare it with the offer size.
3. Don’t confuse anchors with underwriters. Anchor investors bid before the issue opens, but they are investors, not guarantors.
4. For SME issues, read the underwriting table. Under the current rule, it should show the lead manager’s own-account share and any nominated investors.
5. Remember the downside is a refund. In a mainboard IPO that misses minimum subscription, your money comes back. The bigger risk is usually in an issue that scrapes through and then lists weakly.

Underwriting sounds like a safety net, and in a narrow sense it is. But in most Indian book-built IPOs, the net is spread under the paperwork, not under demand. If you want to understand how demand is actually built and allocated, our guides to the book building process and the wider IPO investing in India guide are good next reads.

Disclaimer: This article is for educational purposes only and is not investment advice. Company A, Company B and all issue sizes, prices and share counts in the examples are hypothetical. Investing in IPOs and stocks involves risk of loss. Please read the offer document and consult a SEBI-registered investment adviser before making any investment decision.