Payout Policy: The Decision, the Ratio and the Trade-Off

Payout policy is the standing decision about what a company returns to shareholders and what it keeps. The dividend payout ratio measures it: dividend per share over earnings per share. Sankalp Industrial Systems Limited, invented, paid Rs 2.60 regular and Rs 1.00 special against earnings of Rs 6.90, a 52.17 per cent payout. Every rupee kept instead is assumed to earn 18.00 per cent against a 12.00 per cent capital charge.

The shape of this decision is older than any company, and it is recognised faster at a kitchen table than at a board table, faster in a house than in an annual report. A household earns Rs 90,000 in a month and spends Rs 62,000 on the things a month demands. Rs 28,000 is left over. The only interesting question in household finance is what happens to that Rs 28,000: does it come out and get enjoyed, or does it go back into the shop that produced it, the tempered glass on the shopfront, the second delivery cycle, the extra freezer?

Notice what is not in dispute. The Rs 28,000 exists. Nobody is arguing about how much was earned. The argument is entirely about where the leftover goes, there are exactly two destinations, and every rupee sent to one is a rupee not sent to the other. A payout decision is a division of one leftover pool between two destinations, and that is the whole of its structure at any size. A company working with Rs 1,38,00,00,000 of profit is doing arithmetic identical in shape to the household doing it on Rs 28,000, and it feels different only because the number has more commas in it.

A company adds a second question the household usually skips. The household asks whether it wants the money now. A company has to ask something harder and more answerable: what will the rupee earn if it stays, and what was it already required to earn? The return a kept rupee earns and the return it was required to earn are the spine of this guide, and at Sankalp Industrial Systems Limited they are 18.00 per cent and 12.00 per cent. The gap between them is the thing to hold on to. Everything else here is that gap, made visible.

What is the Payout Decision, and who actually takes it?

Payout policyThe standing decision about how much of a company's profit is returned to shareholders rather than kept. is the third of the three decisions a company takes with money, and it is taken last for a reason that is structural rather than administrative. The first decision is what to invest in. The second is how to fund it. Only once both of those have been settled is there a figure left over, and the payout decision is a decision about that figure. The payout decision is a decision about a residual, and a residual cannot be decided before the things that create it.

Taking the payout decision last matters more than it sounds. If the dividend is decided first and the investment programme discovered afterwards, no payout decision has been made at all. Whatever the dividend consumed is no longer available to the programme, so an investment decision has been made by accident. The payout decision is taken last not out of politeness but because it is arithmetically the remainder, and a remainder decided first stops being a remainder.

Who takes it is a matter of company law and of a company's own constitution rather than of finance. A board proposes, shareholders approve where approval is required, and the whole apparatus of who may declare what, on what basis and with what disclosure is set by the authorities named in the sources below. Company law and listing conditions change. The arithmetic underneath them does not, and that arithmetic is the subject here.

For Sankalp Industrial Systems Limited, invented, the Year 0 residual is Rs 1,38,00,00,000 of profit attributable to owners on 20,00,00,000 shares. Every rupee of that either walks out of the company or stays inside it. There is no third state. A rupee can walk out as a dividend, it can walk out as cash paid to shareholders whose shares the company buys back, or it can stay. In Year 0 the invented company used one of the two exits and left the other unused: Rs 72,00,00,000 went out as dividend, no shares were bought back in that year, and Rs 66,00,00,000 stayed.

What does the Dividend Payout Ratio divide by what?

The dividend payout ratioDividend per share divided by earnings per share, for the same period. is dividend per share divided by earnings per share, for the same period. The definition is that short, and the two words carrying all the danger in it are the same period. A dividend declared for one year against earnings reported for another is not a ratio; it is two numbers standing next to each other.

The same thing can be computed from the rupee totals instead: everything distributed, divided by profit attributable to owners. When the share count has not moved between the two figures, the two routes give the identical answer, and doing both once shows that they are the same statement. For Sankalp Industrial Systems Limited, invented, in Year 0: Rs 2.60 of regular dividend over Rs 6.90 of earnings per share is 37.68 per cent, and Rs 52,00,00,000 over Rs 1,38,00,00,000 is 37.68 per cent as well.

Then a second ratio arrives, and it is a different measure rather than a refinement of the first. The total payout ratioEverything returned to shareholders in a period, dividends and buybacks together, divided by profit. counts everything that went back to shareholders in the period, whichever exit it used, over the same profit. Where a company pays a dividend and buys back no shares, the two ratios are the same number. Where it does both, they part company, and the dividend payout ratio on its own is describing one of two exits and calling it the whole.

In Year 0 the invented company paid a regular dividend of Rs 2.60 a share and added a special dividendA distribution declared once, outside the regular rhythm, and carrying no commitment to repeat. of Rs 1.00 a share. Rs 3.60 in total, Rs 72,00,00,000 in rupees, against Rs 6.90 of earnings per share: 52.17 per cent. The 37.68 per cent and the 52.17 per cent are both correct arithmetic on the same year and they describe two different things, and which one a note prints changes the picture of the company by fourteen and a half points.

The ratio column, read downwards, comes before anything else. The five regular figures are 37.50, 38.46, 39.29, 38.10 and 37.68 per cent. The highest is 39.29 and the lowest is 37.50, so the entire five year spread of the regular payout is 1.79 points. A 1.79 point spread is about as steady as a published series ever gets. Then the last row jumps to 52.17 per cent, and every point of that jump is the Rs 1.00 special dividend and nothing else. The share count moved once as well, from 20,75,00,000 to 20,00,00,000 at the end of year minus 2, and the smaller share count is why earnings per share climbs faster than profit does across the middle of the table.

Payout vs Retention: is that two numbers or one?

It is one. The retention ratioThe share of profit kept, being one less the payout ratio. is one less the payout ratio, so a company that has stated one has stated the other whether it meant to or not. Sankalp Industrial Systems Limited, invented, paid out 52.17 per cent of Year 0 profit and therefore retained 47.83 per cent of it, and the second figure was never a separate decision. The retention ratio is the first figure read from the other end.

The identity sounds like a triviality. It is not, and the reason is what happens to attention. A payout ratio is published, quoted and compared. A retention ratio usually is not, so the rupees it describes travel through a report without ever being named. A number nobody states is a number nobody asks a question about, so naming the retention ratio out loud is the quickest way to make the kept rupees visible. Rs 66,00,00,000 stayed inside this invented company in Year 0, and unless somebody writes that figure down, the only number in the conversation is the Rs 72,00,00,000 that left.

What does a retained rupee actually have to earn before it is worth keeping?

Here is where most first readings of a payout ratio go wrong, and it is worth slowing down. Ask somebody what a retained rupee is worth to a shareholder and they will reach for the return the company makes on it. The return on its own is missing a subtraction. The rupee was not free. The rupee belonged to shareholders and lenders, who could have had it, and they price the use of it. So a retained rupee has to clear a charge before it has done anything at all for anybody.

Take the street version first. Somebody borrows Rs 1,00,000 at 12 per cent a year to buy a second sewing machine, and the machine brings in Rs 12,000 a year. The machine works. The workshop is busier. The Rs 12,000 goes straight out again as interest, so the household is exactly where it started. Only earnings above the 12 per cent are earnings the household gets to feel. Retained profit behaves the same way, except that nobody posts an invoice for the charge, so it stays invisible.

For Sankalp Industrial Systems Limited, invented, the two figures are locked in the forecast this guide borrows them from. The return on new capitalWhat a rupee of newly invested capital is assumed to earn, here 18.00 per cent. is 18.00 per cent, and it is an assumption of that forecast rather than an observed fact about the company. The cost of capitalThe blended return the company's funders require, here 12.00 per cent. is 12.00 per cent, and how that figure is built is settled elsewhere and used here as a single number. The difference between them, 6.00 points, is the spreadThe gap between what capital earns and what it costs, here 6.00 points., and the spread is the entire economic content of a retention decision.

The cost of capital is precisely what a retained rupee was already required to earn, so a rupee earning exactly the cost of capital has added nothing. Draw the line of value against return and it crosses zero at 12.00 per cent, not at zero per cent. The crossing at 12.00 per cent is why a company reporting a healthy looking return on its projects can still be doing nothing for anybody, and why the payout question cannot be settled by looking at the return alone.

What did the payout decision earn and cost at this invented company in Year 0?

The two halves now come together on one year, line by line. Everything below belongs to Sankalp Industrial Systems Limited, invented, in Year 0, and every figure comes from the same locked record.

Now run the retained rupees through the spread, and watch the subtraction that people skip. At the assumed 18.00 per cent, Rs 66,00,00,000 of retained profit earns Rs 11,88,00,000 a year. Against that stands the capital charge at 12.00 per cent, being Rs 7,92,00,000 a year. Those same rupees were required to earn that much before anybody could call anything a gain. Rs 3,96,00,000 a year is what is left, and that is the honest figure for what Year 0 retention is assumed to be worth.

The whole payout decision at this invented company is a decision about how many rupees stay inside a 6.00 point spread, and the 18.00 per cent that creates the spread is an assumption of the forecast rather than an observed fact. Those two things belong in the same breath; separated, a choice becomes a certainty. The day this company runs out of projects earning 18.00 per cent, the spread is zero and retention stops adding anything, and nothing in the record says when that day arrives.

Does paying a dividend change what the company is worth?

Under a stated set of conditions, no, and the argument that says so is worth knowing precisely rather than vaguely. The argument belongs to Miller and Modigliani, Dividend Policy, Growth and the Valuation of Shares, Journal of Business, 1961. Their reasoning is simple once it is seen. If the investment programme is fixed, then a rupee paid out is a rupee that has to be raised again from somewhere, and a shareholder who wanted cash could have made their own by selling a sliver of what they hold. Moving cash from one pocket to another creates nothing, and the payout is irrelevant to the value of the company.

The conclusion is only as strong as the conditions underneath it, and the conditions are the useful part. Miller and Modigliani assume perfect capital markets: no taxes that treat a dividend differently from a sale, no transaction costs and no cost of issuing new shares, and everybody knowing the same things at the same time. Miller and Modigliani also assume no conflict between managers and shareholders about what the cash inside the company is for. And, most importantly of all, they assume the investment programme is fixed and does not change because of what is paid out.

Every one of those conditions that fails on a particular company is a reason that company's payout does matter, so the assumptions are not a weakness in the argument but the instructions for using it. Run them one at a time against Sankalp Industrial Systems Limited, invented, and the picture becomes concrete rather than abstract.

How Payout Policy Affects Firm Value and Financial Flexibility: what moves and what does not?

Two separate things are happening and they are easy to run together. The first is the value effect, and on the arithmetic above it is small and conditional: rupees kept inside a 6.00 point spread produce Rs 3,96,00,000 a year on the Year 0 retention, and that figure survives only as long as the 18.00 per cent assumption does. The second is the flexibility effect, and it is not conditional at all. The flexibility effect is a claim on cash that has not arrived yet.

Financial flexibilityThe capacity to fund an opportunity or absorb a shock without asking anyone for money. is the capacity to fund something or absorb something without going and asking anybody for money. Every rupee committed to a recurring dividend is a rupee of that capacity spent in advance. The regular dividend at Sankalp Industrial Systems Limited, invented, has risen in each of the last five years, from Rs 1.80 a share to Rs 2.60, and the Rs 2.60 level costs Rs 52,00,00,000 a year. Nobody signed anything. The company will be read against the Rs 2.60 next year and the year after, so the level is a claim all the same.

Set that against what the same record says about cash. The company holds Rs 1,20,00,00,000 of cash and equivalents, of which Rs 40,00,00,000 is described as the operating cash the business needs to run and Rs 80,00,00,000 as excess. The excess divided by the annual regular dividend is about 1.54 years. The buffer this invented company has, measured in years of its own regular dividend, is about a year and a half, and that is a statement of arithmetic rather than a judgement about whether the buffer is the right size.

The same record also shows why the two effects pull against each other. The forecast puts Rs 1,00,00,00,000 of net new capital to work every year, of which Rs 25,00,00,000 comes from new borrowing on the schedule. Year 0 retention was Rs 66,00,00,000. Put those two figures side by side, and Rs 66,00,00,000 plus Rs 25,00,00,000 is Rs 91,00,00,000 against a programme of Rs 1,00,00,00,000. The comparison sets an accounting figure against a cash programme rather than stating a funding identity, and how cash actually gets to a project is settled elsewhere. But the direction it points in is real: the payout, the borrowing and the programme are three numbers in one arithmetic, and moving one of them moves at least one of the others.

What about the other door out, the buyback?

A dividend is one route by which cash reaches shareholders. A buyback is the other: the company buys its own shares from whoever will sell, and the cash leaves in exchange for shares that are then extinguished. Sankalp Industrial Systems Limited, invented, did this once in the five year record. At the end of year minus 2 it bought 75,00,000 shares at Rs 80.00, being Rs 60,00,00,000, and the share count fell from 20,75,00,000 to 20,00,00,000.

A buyback is a payout, so a dividend payout ratio in a year that contains one is describing part of what left and calling it the whole. That is the reason the total payout ratio exists as a separate measure. How a buyback is executed, what it does to earnings per share, and the difference between the effect a company would print and the effect an honest calculation produces are covered separately. The structural point is what matters here: two doors, one pool, and any ratio that counts one door has understated the payout.

How to Analyse a Company’s Payout Policy: what are the six checks, in order?

Three of the six checks produce a wrong answer if they run before the first one, so the order matters more than any single check. Work down the list on whatever the company itself has published, and do not divide anything until step one is done.

When has retention run out of good options?

The honest answer is that the record on this invented company does not say, and the more useful answer is what a record would have to show before anybody could say it. The moment the return on new capital falls to the cost of capital, the spread is zero and a kept rupee earns exactly what it was already required to earn, so retention stops adding anything. Nothing about that moment announces itself. There is no line in any statement that reads the projects have run out.

Four things would have to be visible together instead. The return on new capital would have to be falling towards the charge rather than sitting above it. Cash would have to be piling up without a stated use, the open question left by the fifth condition on the irrelevance list, with Rs 80,00,00,000 sitting in a line described as excess. The reinvestment the company actually does would have to be shrinking rather than holding at Rs 1,00,00,00,000 a year. And the reasons given for keeping the money would have to be getting vaguer year on year.

The record locks the arithmetic and not the conclusion, so none of those four is evidence on its own and none of them is a finding about Sankalp Industrial Systems Limited. What matters is where to look and what the spread has to do before the answer changes. Anyone who reads a payout record and comes away with a verdict has added something to it that was not there.

What does a lender, an analyst or a household actually do with this?

Three different readers use the same two ratios for three different purposes, and it is worth seeing all three because the arithmetic is identical and the question is not.

A lender reads the payout as a competing claim. Cash that leaves as dividend is cash that is no longer available to service debt, so a lender looks at the regular dividend of Rs 52,00,00,000 as a standing outflow that sits ahead of nothing but has to be found each year, and at the Rs 20,00,00,000 special as something that happened once. Loan documentation frequently constrains distributions for exactly this reason, and the terms of any such constraint are a matter for the documents themselves.

An analyst reads the payout as a statement about the opportunity set. A company keeping a lot is implicitly saying it has somewhere to put the money; a company keeping little is implicitly saying it does not, or that it has chosen not to. The ratio itself checks neither statement. Check five in the list above exists for that reason: name what the kept rupees are assumed to earn, and the implicit claim becomes an explicit one that can be examined.

A household reads it as the difference between cash arriving and value accumulating somewhere it cannot spend. The two are genuinely different, and the difference is the whole reason payout policy is a subject at all. The three readers reach for the same ratio because it is one of the few numbers a company publishes that is simultaneously a statement about cash, a statement about opportunities and a statement about what the board expects to be able to keep doing.