An IPO, and the question that decides everything

What an IPO is, and what it is not

An initial public offering — an IPO — is the first sale of a company's shares to the general public, after which those shares trade on an exchange. What decides where your money goes is not the company or the price. It is whose shares are being sold — newly created ones, or somebody's existing ones.

Two separate things happen at the same moment, and they get treated as one. The company becomes listed, which means its shares acquire a market where they can be bought and sold every trading day. And somebody receives the money that the buyers pay in. Those are different events, and the second one has two possible recipients.

There is also a difference in the transaction itself that survives the day. Buying in the offering means buying from the issuer or from a named selling shareholder, at a price arrived at through a process the seller designed. Buying the same share a month later means buying from another investor, at a price two strangers agreed on. The share is identical; the terms on which you obtain it are not, and almost everything that makes an offering unusual comes from that difference.

Fresh issue or offer for sale — the split that decides everything

An offering can raise new money for the company, transfer existing shares from current owners to new ones, or do both at once. Most Indian offerings do both, and the offer document states the split.

A fresh issue creates shares that did not exist before. The company issues them, receives the proceeds, and the total number of shares outstanding rises. Every existing shareholder who did not buy more now owns a smaller percentage of a company that has more cash in it than it did yesterday.

An offer for sale creates nothing. Shares that already exist move from named selling shareholders to the buyers, and the money follows the same path in reverse. The company's share count is unchanged, its bank balance is unchanged, and the sellers' stake in it has fallen. The company is the venue, not the recipient.

Here is the mistake this section exists to prevent, and it is a common one. The offer document carries a section headed “objects of the issue”, listing what the money will be used for — a plant, working capital, repaying borrowings. That section describes the fresh portion only. If most of the offering is an offer for sale, the projects listed are being funded by the minority slice, and the headline offering size is not the number paying for them. Read the two figures together or the section is misleading in a way nobody has to lie to achieve.

None of which makes an offer for sale a defect. An early backer who put money in eight years ago has to be able to get it out, a founder is entitled to sell a slice, and a listing needs shares in public hands to function at all. It simply changes what you are taking a view on. Fund a plan and your risk is whether the plan works; buy a seller's stock and your risk is entirely the price you paid for it.

Book building, and who really sets the price

In a book-built offering the company and its merchant bankers publish a price band before bidding opens — a floor and a cap. Investors bid within it, and once bidding closes the final price is set from the resulting book of demand. Retail bidders are usually offered the option of bidding “at cut-off”, which means accepting whatever price is discovered rather than naming one.

That is genuine price discovery, and it is worth being precise about its boundaries. Discovery happens inside a band the seller chose, before a single public bid existed. The market gets to say where within the band the price lands, and whether the offering fills at all. It does not get to say the band was set too high. A process can be competitive and still start from a number one side of the table picked.

The offer document contains the issuer's own defence of that number, in a section on the basis for the issue price. It typically sets the company's ratios beside a group of listed peers. The peer group is selected by the issuer. That is not a scandal — somebody has to choose it, and the choice is disclosed — but the useful reading is the reverse one: look for the obvious comparable that is missing from the list, and ask what including it would have done to the comparison.

A portion of the offering is normally placed with large institutions before bidding opens to everybody else, at a price fixed then, and their names and quantities are published. This anchor book is widely read as validation.

What it establishes is narrower. Certain institutions were willing to take stock at that price on that day, having negotiated ahead of everybody else, and their shares cannot be sold for a period afterwards. Whether they intend to hold beyond it is not disclosed and is not knowable. Appetite and value are different findings, and the anchor book reports the first — which is usually a response to a broader flow picture, set out in FII and DII flows.

Why oversubscription hands it to a draw

The offering is divided into reserved portions before it opens — a share for qualified institutional buyers, a share for large non-institutional bidders, a share for retail individual bidders, and sometimes a reservation for employees. The proportions are fixed by regulation, and they are not the same for every company; which set applies depends on the route the company qualified under.

The mechanism that matters is that each portion is settled inside itself. Institutional bids do not compete with retail bids for the same shares. So the number that governs whether you receive an allotment is your own category's subscription multiple, and the headline figure — “the offering was subscribed forty times” — is an aggregate of books that were never pooled.

The arithmetic from there is simple division. Suppose the retail portion receives bids for twelve times the shares reserved for it, a figure chosen here to work the sum rather than drawn from any particular offering. Then across the whole retail book there is roughly one lot available for every twelve lots bid for. Where most applications are for the minimum lot, that is roughly one application in twelve receiving shares.

Now the part the standard explanation runs together, and it is worth pulling apart because the two halves have different standing. The first half is arithmetic. Shares are whole units and the minimum bid is a lot, so scaling every retail bid down by the same factor would hand most applicants a fraction of a lot — a quantity that cannot be delivered because it does not exist. Indivisibility forces a subset: some applicants receive a whole lot and the rest receive nothing, because there is no arrangement in which everybody gets a little.

The second half is a choice. Once only a subset can be served, something still has to decide who is in it, and the candidates are all workable — the order the applications arrived in, the size of the bid, the price bid. Indian book-built issues settle it by a random draw over the minimum bid lot, so that the largest possible number of applicants receive one lot each. That is a rule the allotment basis follows, not something the arithmetic dictated. The arithmetic makes rationing unavoidable; the draw is the method chosen to do the rationing, and running the two together is how people conclude that a bigger application must shorten the odds.

Which produces a consequence worth holding on to. In a heavily oversubscribed retail book the unit of success is the application, not the size of it, because everyone in the draw is competing for the same one lot. And there is a ceiling on what a retail application can be worth: cross it and the application leaves the retail category and competes in a differently rationed bucket, against bidders whose applications are much larger. Bidding harder does not scale the way instinct says it does, and the conventional description of this — that allotment is a lottery — is accurate rather than dismissive.

Applying is not free, either, and the cost is easy to miss because it is not a fee. The application blocks the money in your bank account: it stays yours, it is not debited, and it is also not available to you until the allotment is settled or the block released. On a heavily oversubscribed offering you are handing over the use of that money for the window in exchange for a probability you can calculate from the subscription figures. That is a real price, and a small one, but it is not zero.

The seller picks the day and writes the document

Everything above is mechanics. This section is the framing the mechanics add up to, and it is the reason an offering deserves a different kind of attention from an ordinary purchase on the exchange.

Three structural facts, each uncontroversial on its own.

The timing is the seller's choice. No company is compelled to sell shares in a particular month. It comes to market when it and its bankers judge the moment right — after a strong run of results, into a receptive market. Nothing in that is improper, and it does not require a single number to be wrong. It means only that the financial periods you are shown were selected, in the plain statistical sense, and selection has a direction.

The disclosure is the seller's document. The offer document is drafted by the company and its advisers and filed with the regulator, and it is genuinely the most complete account of a business that a retail buyer will ever get before buying — more complete than anything you will find on an established listed company on an ordinary Tuesday. It is also organised by the party selling. What goes in is largely prescribed; the order, the framing and the emphasis are not.

There is no market record at all. For any listed share you can look at how the price behaved through a bad quarter, a sector downturn, a management change. Here that series does not exist yet. The first observation is created on listing day, and every judgement before it rests on documents rather than on behaviour.

Put those together and you get the honest framing: an offering is the one purchase where the seller chooses the timing, controls the disclosure, and faces a buyer with no price history to check it against. That is a standing tilt in the structure, not an accusation against any company. Plenty of businesses have listed and gone on to compound for decades. The tilt does not tell you which ones — and how any particular offering turns out is not something anyone can tell you in advance, which is why nothing in this article attempts to.

The shares that are not for sale yet

On listing day, only a part of the company's shares can actually be traded. The promoters' holdings, the shares held by investors who came in before the offering, and the anchor allocation are each locked for a period, and the periods are not the same for each group. Every one of them is set out in the offer document with its expiry.

The mechanism follows directly. The listing price forms on a restricted supply, because most of the share count is not permitted to reach the market yet. As each lock-in expires, a further block of shares becomes sellable. Whether those holders choose to sell is unknown and unknowable. That they become able to is scheduled, published, and readable before you apply.

Give the reading error a name and it stops happening: the float illusion — treating the first weeks of trading as the market's verdict on the company, when it is the market's verdict on whichever fraction of the shares was allowed to reach it. A price set by a small free float is set by fewer holders, which is a narrower sample of what owners would accept than the same price a year later.

Two honest boundaries on that. It implies nothing about direction — a supply schedule is not a forecast, and no arrangement of it becomes one. And a lock-in expiry is a date on which selling becomes possible, not one on which it happens. The claim here is only that the quantity available to trade is a known variable that changes on known dates, and that a price formed while it is small was formed on less information than it will be formed on later.

What to read in the offer document, in order

The document runs to hundreds of pages and almost nobody reads it end to end. Six passages carry most of what a reader can actually act on, and they are not the six that get quoted.

The offering structure — the fresh issue and offer for sale split, and the names of the selling shareholders with the quantity each is selling. A founder releasing a slice and an eight-year-old fund exiting in full are different facts. Neither disqualifies anything; both are worth knowing before the rest of the document colours your reading.

The objects of the issue — how the fresh money divides between repaying borrowings and building something, and how much sits under general corporate purposes, which is by definition the portion with no stated use.

Related-party transactions — where the company buys from, sells to, lends to or rents from entities its promoters control. Ordinary in a family-founded business, and the place where economics can leave a listed company without anything appearing to have gone wrong.

Contingent liabilities and litigation — obligations that are disclosed but do not sit in the balance sheet totals, precisely because whether they crystallise is undecided.

The basis for the issue price — the peer set, read for what is absent from it, as above.

The risk factors — and here is the specific mistake. Readers skim this section as legal boilerplate because much of it is: every document warns about monsoons, regulation and competition. The ones that are not boilerplate read differently, and they are recognisable. A single customer accounting for a large share of revenue. A licence or lease due for renewal. One plant, one supplier, one contract. Those paragraphs are there because they had to be, and they are the cheapest research available to anyone applying.

What deserves the least weight is the pair of numbers everybody watches: the live subscription multiple, and the unofficial premium quoted before listing in the grey market. The second one is worth understanding for what it is — an off-exchange indication of a price, with no exchange record of what actually changed hands, quoted by people who may hold the position it prices. The first is a real number about demand, but demand for an offering and the quality of a business are different questions, and only one of them is answered in the document you are holding.

One practical point on the other side of listing: a sale made shortly after allotment is a short-term holding for tax, and the treatment turns entirely on the calendar — the mechanics are in capital gains tax explained.

What happens once it is an ordinary share

The moment listing is done, almost everything distinctive about an offering stops applying. The shares are ordinary listed shares. The seller no longer controls the timing or the document. Quarterly results arrive, a price series starts accumulating, and the company begins to be assessed the way every other listed company is.

Which is precisely the difficulty a new listing presents to any tool built on history. FNOTrader's Stocks app screens and backtests across roughly 2,390 stocks and 17 NSE sector and size indices, on ratio and price history that a company listed last month does not yet have — the ranking and factor work described in the stock screener guide needs years of observations to say anything. That gap is not a limitation to be worked around. It is the same fact this article started with, stated in the language of data: at the moment of the offering, the evidence available is a document, and the evidence that accumulates afterwards has not been produced yet.

Anyone comparing a single company against owning a diversified basket of them will find that trade-off set out in equity funds explained, which is a different decision from this one and worth keeping separate.

Common questions

Continue reading

More in Investing Basics · App: Stocks · Definitions: glossary · Free tools: calculators · All: every article