Follow-On Offering: Raising Again After Listing

A follow-on offering is a listed company issuing fresh shares after its first sale, usually placed with a small number of large buyers at a price below the level the shares already trade at. Nobody on the existing register receives an entitlement, so every holder's proportion falls whether they act or not. Having a published price already is what makes it quick.

A company that has been listed for a few years is in an odd position when it needs money. The company is not a stranger any more. There is a level printed against its name every trading day, a set of accounts going back years, and a list of holders that anybody can download. All of that was expensive to build, and the interesting thing is what it can now be spent on.

Being known already buys speed. Once a level exists in public, a company selling fresh shares no longer has to persuade anybody what the shares are worth, only agree a number close to the one already printed. That single fact is the whole of why a follow-on offering can be arranged in days rather than months, and it is also why the exercise cannot be copied by a company selling shares for the first time.

One hypothetical placement is worked all the way through on Sarvani Coatings Limited, an invented paints and coatings maker used throughout these notes. The definition of a share, what a holder of one actually has, and how a profit ladder is built are covered separately. The arithmetic of the placement itself follows below, and in particular the part almost nobody works out until it has already happened: exactly what a placement does to the shareholding pattern.

What is a follow-on offering, and why can a listed company do it quickly?

Strip it to the bones. The company creates new shares that did not exist before, sells them for cash, and the cash goes onto its own balance sheet. Those three steps are the whole transaction. A follow-on offering is not a holder selling to another holder, and no existing share changes hands. The number of shares in issue goes up, and the company has more money than it did.

Two things about that make it different from any other way of raising money. First, the money never has to come back. There is no maturity, no coupon, no covenant and nothing to refinance in five years. Second, the people who put the money in are buying a claim on the same profits that everybody already on the register was expecting to divide among themselves.

Now the speed. Consider selling a second-hand car. A seller who has never sold one and has no idea what it is worth spends weeks on it: photographs, listings, conversations, three people who come and look and do not buy. If the neighbour sold the identical model last Tuesday and everybody in the street knows what it went for, a price can be agreed over a cup of tea in twenty minutes. Nothing about the car changed. The difference is that a reference number exists.

A listed company selling fresh shares is the neighbour with the reference number, and a company selling shares for the first time is the person with no idea. The published level does not tell either party what the shares are worth in any deep sense. The published level does something narrower and more useful: it gives both sides a number to argue around, so the negotiation is about the gap rather than about the whole quantity.

How is this different from the first time the shares were sold?

The first sale has one job that no later sale has: it has to find out what buyers will pay. Everything about how a first offering is run follows from that. The long document exists because buyers who have never valued this business need the whole of it in front of them. The road show exists because the company has to be met. The price range exists because nobody yet knows where inside it the answer sits, and the book has to be collected before anybody can say.

A follow-on offering has none of that job to do. The buyers have been able to read this company's results for years. Analysts have been publishing on it. There is a level printed every day that thousands of separate decisions have already agreed on. So the exercise collapses to a single question. What discount to that level does a buyer require in order to take a large parcel in one go?

Because the question is smaller, everything attached to it is smaller: the document is shorter, the time is shorter, and the risk that the company goes through the whole exercise and fails to sell anything is far lower. None of that makes a follow-on offering better than a first sale. A first sale and a follow-on offering are not alternatives. A company can only run the second because it already ran the first.

The full side-by-side treatment of the two, including what the disclosure requirement does in each case and how the risk sits differently on the company, is set out under initial public offering (IPO) versus follow-on offering. The single mechanical consequence matters here: because a price already exists, the transaction can be aimed at a handful of buyers instead of at everybody, and that choice creates all of the arithmetic below.

Who actually buys, and why is the price below the quoted level?

In most placements of this kind the shares go to institutions: funds, insurers, and other buyers large enough to take crores of rupees of stock in a single decision. Dealing with four buyers is faster than dealing with four lakh of them, so the company is not obliged to spread the shares across the whole market and typically does not want to.

Now the price. Sarvani Coatings' shares stand at an illustrative Rs 486/- on the date this guide was built. The hypothetical placement is at Rs 460/-. The placement price is Rs 26/- a share below the quoted level, or 5.35 per cent of it, and across 1,20,00,000 shares that is Rs 31,20,00,000/- of difference against the quoted level.

Why would a company accept less than the printed number? Because the printed number is the price of a few shares, not of a blockA parcel of shares large enough that it cannot be bought or sold through ordinary market dealing without moving the level at which the shares trade against whoever is doing it. this size. Sarvani Coatings trades an illustrative Rs 42 crore of value on an average day. A buyer who wanted 1,20,00,000 shares from the market would be trying to absorb about thirteen days of the entire market's turnover, and the level would move against them long before they finished. The discount is what makes it possible to do in one afternoon what could not be done in a fortnight.

Consider a wholesaler and a corner shop. A shop sells one kilogram of rice at the price on the shelf. A buyer who wants two tonnes delivered tomorrow does not pay the shelf price two thousand times over, and no seller expects it. The shelf price was never a price for two tonnes. A quoted level is a shelf price and a placement is a wholesale transaction. Comparing the two directly compares two different things.

Who is diluted, and by exactly how much?

Here is the hypothetical in full, fixed so that every treatment in these notes that reaches for it lands on the same figures. Sarvani Coatings Limited has 24,00,00,000 shares in issue. The company places 1,20,00,000 new shares with institutions at Rs 460/- each, raising Rs 5,52,00,00,000/-, or Rs 552 crore. Nothing else about the company changes on the day. The count becomes 25,20,00,000.

Nobody on the existing register is offered anything. There is no entitlement, no application form and no decision to take. A holder of 1,000 shares had 1,000 shares before the announcement and has 1,000 shares after it. The change is underneath them: the total those 1,000 shares are measured against is 5.0 per cent larger than it was.

Work the earnings first. Sarvani Coatings reported profit after tax of Rs 278 crore in its most recent published year, giving earnings per share of Rs 11.58/-. The placement does not change that profit by a single rupee on the day it settles. Divide the same Rs 278 crore by 25,20,00,000 shares and earnings per share is Rs 11.03/-.

Because it makes the arithmetic doable in the head straight from an announcement, the identity is worth holding on to. New shares as a proportion of the enlarged total is the proportional fall in earnings per share, on the assumption that profit does not move. The assumption is worth stating every time. Profit is what eventually stops holding still.

The fall in earnings per share is immediate and certain, and the recovery is neither. The Rs 552 crore is real money sitting on the balance sheet from the day of allotmentThe moment the company formally assigns the new shares to the buyers who applied for them. Before allotment the shares are applied for; after it they are in issue and count towards the total., and it will earn something. When it earns enough that profit rises by more than 5.0 per cent, earnings per share is back where it started. One year or six depends on the use the money is put to, and the announcement of a placement names no use.

Do all the per-share figures move the same way?

No, and this is where a reader who has learned the word dilution and stopped there goes wrong. Dilution is not a force that pushes every per-share figure downwards. Dilution is a change in one number, the denominator, and what happens to any particular per-share figure depends on what the placement did to that figure's numerator as well.

Take book value per shareWhat the balance sheet says stands behind each share: the equity total spread evenly across every share in issue. Book value per share is an accounting measure, and not a price anybody has ever paid.. Sarvani Coatings' net worthEverything shareholders have put into the company plus everything the company has kept back out of past profits, after every liability has been deducted. Net worth is the equity total on the balance sheet. at the end of its most recent published year was Rs 1,486 crore, which on 24,00,00,000 shares is Rs 61.92/- a share. The placement puts Rs 552 crore of fresh cash in, so net worth becomes Rs 2,038 crore. Divide that by 25,20,00,000 shares and book value per share is Rs 80.87/-.

It went up. Book value per share rose 30.6 per cent in the same transaction that took earnings per share down by 4.76 per cent. Earnings per share fell because the new money earned nothing yet, and book value per share rose because the new shares were sold for far more than the book value each existing share already carried. Two per-share figures, one transaction, opposite directions, and both are correct.

The balance sheet moves too. Sarvani Coatings was already sitting on Rs 312 crore of cash and investments while owing Rs 240 crore, so it carried net cash of Rs 72 crore before any of this happened. Add Rs 552 crore and it holds net cash of Rs 624 crore. The complete sweep across all the per-share figures a corporate action can move, including several not needed here, is set out separately under How Corporate Actions Affect Shares and Per-Share Metrics.

One caveat on all of the above. A real placement costs something to run, and those costs come out of what the company keeps. Rs 552 crore is therefore the gross amount, and the net amount landing on the balance sheet is a little lower.

What does the same company look like if it runs a rights issue instead?

The comparison with a rights issue is what makes the whole thing click, and the two hypotheticals were fixed so that it works. In the rights issue worked under Buyback vs Rights Issue, each twenty shares already held carry an entitlement to one more at Rs 350/-. The entitlement also creates 1,20,00,000 new shares. The count again goes to 25,20,00,000. Earnings per share again lands at Rs 11.03/-.

Identical count. Identical earnings per share. And yet the two transactions do completely different things to the people already on the register. One difference does it: in a rights issue every existing holder is offered new shares in proportion to what they hold, and in a placement nobody is.

Follow one holder through both. Under the rights issue, a holder of twenty shares is offered one new share at Rs 350/-. If they take it they hold twenty one shares out of 25,20,00,000, exactly the same proportion they held before. The theoretical ex-rights priceWhere a share lands the instant a rights issue takes effect, assuming nothing else moves. The figure comes out of averaging the holding a person already had with the newly issued share, each at what was paid for it. of Rs 479.52/- means the value they gave up on the old shares comes back on the new one. If they decline, they are diluted, but the dilution was their decision.

Under the placement the same holder is offered nothing. The holder still has twenty shares, now out of 25,20,00,000 rather than out of 24,00,00,000, and there was no moment at which anything else could have been chosen. A rights issue makes dilution optional and a placement makes it automatic, and for a holder without a large cheque book that is the single most important difference between the two.

Why would a board choose one route over the other?

Set the ranking aside. There is no better route here, only a set of trade-offs, and which route to prefer turns on circumstances that no general treatment can see.

A placement buys speed and certainty. A handful of buyers can be approached, priced and closed inside a very short window. Because those buyers commit before the offer is public, the company knows the money is coming. And because they are large, they can absorb an amount that would take an ordinary offer weeks to gather.

A rights issue buys fairness of treatment. Every holder is offered the same proportion on the same terms, so the change in the register is a choice each of them makes rather than something done to them. A rights issue is also usually slower, and there is a real possibility that part of it is not taken up.

Now the honest part, the bit most treatments leave out. The speed and certainty are not free and they are not paid for by the company. Look at the two hypotheticals again. Rs 460/- is Rs 110/- more per share than Rs 350/-, so the placement raises Rs 132 crore more than the rights issue for exactly the same number of new shares. The additional money comes from selling nearer the quoted level, and the payment for it is that the existing register no longer gets first refusal. The promoter proportion ends at 49.9 per cent in one case and 52.4 per cent in the other.

The rights issue would also have required something of the promoter group. Taking up its full entitlement of 52.4 per cent of a Rs 420 crore issue means finding Rs 220.08 crore in cash. A promoter group without that money has an entitlement it cannot use. Several perfectly ordinary reasons of that kind send a board to a placement instead, and a reason is not a verdict.

What does the size of the discount tell a reader?

A reader wants a verdict at this point, and the published facts do not supply one.

The fact is simple enough. Sarvani Coatings placed at Rs 460/- against an illustrative quoted level of Rs 486/-, a discount of 5.35 per cent. Across the whole placement the gap is Rs 31,20,00,000/- less than the same shares would have fetched at the quoted level. The arithmetic is worth noticing.

The meaning is a different question, and at least three explanations fit the same number. The parcel might simply have been large relative to what the market absorbs in a day, and a larger parcel takes a wider discount. The market on that particular day might have been unsettled, and a buyer commits to a fixed price while everything around them is moving. Or the buyer might have judged the business worth less than the quoted level. Every one of those explanations produces the same published discount, and the announcement contains nothing that separates them.

So the discipline is the same one these notes apply to every observable with more than one sufficient cause. Record the discount next to the amount raised and the speed at which it was raised. Write down what would separate the explanations. Then stop. The step from a number to a judgement about the people running a company is exactly where research quietly turns into invention.

What does this do to the shape of the shareholding pattern?

The shareholding pattern is the part worth the most attention, and almost nobody works it out in advance.

Sarvani Coatings' published pattern carries four lines: 52.4 per cent against promoter and promoter group, then 18.2 against foreign portfolio investors, 14.6 against domestic institutions and 14.8 against retail and others. The four lines are proportions, so convert them to shares before doing anything else. On 24,00,00,000 shares, the promoter groupOne line in the published pattern, covering whoever is classified as founding or controlling a company. The classification follows a rule made by the regulator and is not a label the company picks for itself. holds 12,57,60,000 shares, the foreign portfolio investors hold 4,36,80,000, the domestic institutions hold 3,50,40,000, and retail and others hold 3,55,20,000. The four counts add to 24,00,00,000 exactly.

Now the placement. Not one of those four numbers changes. Nobody sold anything. A fifth line appears holding 1,20,00,000 shares, and the total becomes 25,20,00,000. Divide each unchanged holding by the new total.

The placement leaves every share count in the second column unchanged except the fifth, and the fifth is the one it creates. The percentages are the same counts divided by 24,00,00,000 and then by 25,20,00,000, rounded to one decimal place, and both columns add to a hundred at that precision.

Read down the last two columns and notice that every existing line fell, including the ones nobody was thinking about. Retail holders went from 14.8 per cent of the company to 14.1 per cent without a single retail holder doing anything at all.

Two figures on that table are worth stating in words. Free floatWhatever sits outside the promoter group's line in the published pattern, which is the portion that can genuinely change hands. The figure comes from the pattern. A share count on its own cannot supply it., meaning everything not held by the promoter group, was 11,42,40,000 shares or 47.6 per cent, and becomes 12,62,40,000 shares or 50.1 per cent. Institutions taken together, if the placement buyers are institutions, go from 32.8 per cent to exactly 36.0 per cent. A placement is one of the few corporate actions that changes the shape of a register rather than only its size. Comparing the pattern before and after therefore shows things the share count alone never will.

How far can a promoter group fall without selling anything?

Now the figure that makes people sit up. The promoter group held 52.4 per cent. After the placement it holds 49.9 per cent. The promoter group sold nothing, agreed to nothing and did nothing, and it is on the other side of half the company.

Sit with the arithmetic rather than the drama. 12,57,60,000 shares is more than half of 24,00,00,000 and less than half of 25,20,00,000. The two comparisons are the whole of it. Half of 25,15,20,000 is exactly 12,57,60,000, so the crossing happens the moment the total passes 25,15,20,000 shares. The crossing point arrives at 1,15,20,000 new shares, or Rs 529.92 crore raised at Rs 460/-.

And that is the honest qualification. The hypothetical placement of 1,20,00,000 shares is only 4.2 per cent larger than the placement that just crosses, so the promoter group ends up 0.095 of a percentage point below half rather than comfortably below it. Anybody presenting this as a dramatic loss of position has read a rounding boundary as an event. The arithmetic demonstrates something narrower and more useful. A proportion has a denominator and somebody else controls it, so a proportion can move without anybody transacting.

Dilution has the shape of an everyday arrangement. Four people share a house and split the rent four ways. A fifth moves in. Nobody moved out, nobody gave anything up, and everybody's share of the rent has changed. The rent split is dilution. Shares feel like possessions in a way that a quarter of the rent does not, and that is the only reason the same arithmetic feels strange in a company.

What does the announcement itself actually contain?

Everything worked above comes off six lines that a placement announcement carries as a matter of course. The useful skill is knowing which line answers which question, and knowing which questions none of the six lines touches at all.

What is settled elsewhere, and what falls outside a placement?

Several questions get asked in the same breath as a follow-on offering, and each is answered in its own place so that it is done once and properly.