Economic Profit: Profit After the Cost of the Capital Used

Economic profit is operating profit after tax less a charge for the capital that produced it. Sankalp Industrial Systems Limited, invented, earned Rs 1,80,00,00,000 on Rs 12,00,00,00,000 of invested capital, and at its locked 12.00 per cent cost of capital the charge is Rs 1,44,00,00,000, so economic profit is Rs 36,00,00,000. Both routes to that figure agree exactly.

The awkward part of this idea is much easier to feel in a tailoring shop than in a set of consolidated accounts. Start on a lane behind a wholesale market. A man runs the shop. He has two machines, a rack of cloth, a cutting table and one helper. At the end of the year he counts up: takings less rent, less the helper's wages, less thread and electricity and the loan instalment on the second machine. Rs 4,00,000 is left. He tells his brother the year went well. The shop made a profit.

His brother asks one question, and it is the only question that matters. How much money is sitting inside that shop? The two machines, the cloth on the rack, the cutting table, the bills the market traders have not yet settled, the deposit with the landlord. Add it up honestly and it comes to Rs 20,00,000. Every rupee of it came out of savings or out of profits the tailor left in the business instead of taking home. The Rs 20,00,000 is not free. The same money could have sat somewhere else. Somebody, somewhere, is being paid nothing for it.

So the brother asks the second question. If putting Rs 20,00,000 into this shop is worth doing, the shop has to pay for the use of that Rs 20,00,000 before anybody calls the year a good one. Suppose the fair price for money of that kind, in that trade, at that risk, is twelve paise on the rupee a year. Then the shop owes Rs 2,40,000 for the capital it used, before any of the Rs 4,00,000 counts as a gain. The leftover is Rs 1,60,000. The profit that survives after the capital inside the business has been paid for is economic profit.

Nothing about that reasoning is unusual, and none of it is an accounting rule. Anybody makes that comparison instinctively when weighing a shop against leaving the money alone. The odd thing is that a company's income statement does not make it. The statement charges the business for the money it borrowed and charges nothing at all for the money its shareholders left in. So the tailor's brother is doing arithmetic that a printed profit and loss account does not do, and a company the size of Sankalp Industrial Systems Limited needs somebody to do it on purpose.

What is economic profit, and how is it different from the profit on an income statement?

Economic profitOperating profit after tax less a charge for the capital used to earn it. is a profit figure with one extra deduction in it. The measure starts from the profit the operating business threw off, subtracts a charge for the capital that was tied up producing it, and what remains is the answer. There is no second step and no adjustment table. The measure is a subtraction, and everything difficult about it lives in choosing what goes on either side of the minus sign.

Sankalp Industrial Systems Limited, invented, is a listed maker of industrial valves, precision castings and the aftermarket parts and service that go with them. Take its base year, called Year 0 throughout, where earnings before interest and tax were Rs 2,40,00,00,000. The company's own assumed effective tax rate of 25.0 per cent is an assumption of this worked example rather than a published rate. Applying it leaves operating profit after taxEarnings before interest and tax, taxed as though the company carried no debt at all. of Rs 1,80,00,00,000.

The capital tied up in the operating business is Rs 12,00,00,00,000. The Rs 12,00,00,00,000 is invested capitalNet working capital plus net fixed assets: the capital sitting inside the operating business., being net working capital of Rs 1,80,00,00,000 plus net fixed assets of Rs 10,20,00,00,000. What invested capital contains, and why, is settled elsewhere. The rate the company's capital costs is 12.00 per cent, and the 12.00 per cent is the company's own locked weighted average cost of capitalThe single blended rate a company's capital costs, taken here as a given figure rather than derived., taken as a given. How a cost of capital is built, and where each of its inputs comes from, is covered separately.

The capital multiplied by the rate gives the capital chargeInvested capital multiplied by the cost of capital: what a year of using that capital costs.: Rs 12,00,00,00,000 at 12.00 per cent is Rs 1,44,00,00,000. Subtract it. Rs 1,80,00,00,000 less Rs 1,44,00,00,000 is Rs 36,00,00,000, and that is Sankalp's economic profit for Year 0. The company's reported profit did not change, its cash did not change, and nothing about the business moved. One deduction was added that the income statement never makes.

Economic profit is denominated in rupees, not in percentage points. The rupee denomination is the reason the measure exists alongside a return on capital rather than instead of one. A percentage answers how well the capital did. A rupee figure answers how much was produced. The two questions are different, they can move in opposite directions, and improving the first can destroy the second.

What is a capital charge, and why do the owners get charged too?

Read an income statement from the top and watch who gets paid. Customers pay revenue in. Suppliers are paid for materials. Employees are paid wages. The landlord is paid rent. Then, near the bottom, the lenders are paid interest. Sankalp's Year 0 interest bill is Rs 48,00,00,000 on gross debt of Rs 6,00,00,00,000, and it is deducted in full, as an expense, exactly like wages.

Now look for the line where the shareholders are paid for the money they left in the business. There is no such line. The shareholders of Sankalp Industrial Systems Limited have Rs 9,00,00,00,000 of book equity sitting in the company, and the income statement charges the business nothing whatsoever for the use of it. An income statement treats borrowed money as expensive and shareholders' money as free, and the capital charge exists to close exactly that gap.

Nobody handed the tailor a template, so he did not make that mistake. He knew perfectly well that the Rs 20,00,000 in his shop had a price whether or not a bank was involved. The template is printed and the shareholders never send an invoice, so companies make the mistake constantly. Silence is not the same as free.

Two details in that drawing are worth pausing on. First, the charge covers all the capital at one rate, lenders and shareholders together. The charge is not an equity charge bolted on top of the interest bill. The single blended rate matters enormously for the numerator, and the commonest failure in practice comes from getting it wrong. Second, the invested capital of Rs 12,00,00,00,000 is measured from the asset side, as net working capital plus net fixed assets. Invested capital is not built by adding up the debt and the book equity, and the two totals on this company are not the same number. The capital charged here is the capital visible in the operating business, stated as such.

What are the two routes to economic profit, and do they agree?

An answer that arrives twice by different arithmetic is understood in a way that a single formula never delivers. There are two ways to compute economic profit, and both are run below.

The first is the residual routeProfit less the capital charge: the leftover after the capital has been paid for., and it is the one already run above: the profit, then the charge, then the subtraction. Rs 1,80,00,00,000 less Rs 1,44,00,00,000 is Rs 36,00,00,000. The route is called residual because what is left is a residue: the part of the profit nobody had a prior claim on.

The second route starts from percentages. Sankalp's return on invested capitalOperating profit after tax divided by invested capital, expressed as a percentage. divides Rs 1,80,00,00,000 by Rs 12,00,00,00,000 and comes to exactly 15.00 per cent. The capital costs 12.00 per cent. The difference between what the capital earns and what the capital costs is the spreadThe return on invested capital less the cost of capital, measured in percentage points., and here it is exactly 3.00 points. Apply that spread to the capital it was earned on: Rs 12,00,00,00,000 multiplied by 3.00 points is Rs 36,00,00,000.

Two calculations, four inputs between them, one answer to the rupee, and the agreement is not a coincidence but an identity. The two routes are the same equation rearranged. Profit less capital times rate is the same as capital times return minus capital times rate. Both come to capital times the difference between the return and the rate. Running both takes thirty seconds and it catches a real class of error: if they disagree, one of the four inputs has been taken from a different definition than the others, and that surfaces before publication rather than afterwards.

Which profit figure and which capital figure actually go into it?

Everything difficult about economic profit is here. The subtraction is trivial; the two figures being subtracted are not, and almost every wrong answer in practice comes from picking one of them carelessly.

Start with the numerator. The numerator has to be operating profit after tax: earnings before interest and tax with a tax charge applied as though the company carried no debt at all. Sankalp's is Rs 2,40,00,00,000 taxed at the company's own assumed effective rate of 25.0 per cent, giving Rs 1,80,00,00,000. The reason it has to be that figure and not a post-interest one is structural. The capital charge is already charging for every rupee of capital in the business, borrowed and owned alike. If the numerator has already had interest taken out of it, the lenders have been charged for twice, once as an expense and once inside the charge.

So profit after taxProfit after interest and tax: a different figure, built for a different question, and not the numerator here. is the wrong numerator. Sankalp's Year 0 profit after tax, before the minority interest comes out, is Rs 1,44,00,00,000: earnings before interest and tax of Rs 2,40,00,00,000, less interest of Rs 48,00,00,000, less tax of Rs 48,00,00,000 at the assumed rate. Profit after tax is a perfectly real figure that answers a perfectly real question. The question it answers is not this one.

Now the denominator. The denominator has to be the capital the operating business actually uses, and that is invested capital of Rs 12,00,00,00,000. Invested capital is not total assets, and it must not include anything that produces none of the profit in the numerator. Sankalp holds Rs 1,20,00,00,000 of cash and Rs 1,00,00,00,000 of non-operating assetsAssets producing none of the operating profit, so they do not belong in the capital being charged for., being a surplus land parcel at Rs 45,00,00,000 and a 26.0 per cent holding in Aruna Tooling Private Limited, invented, carried at Rs 55,00,00,000. Neither the land nor the associate holding produced a single rupee of the earnings before interest and tax above. Charging the operating business for them would be charging it for assets it does not use.

The rule that governs both sides is one sentence long: the numerator and the denominator must belong to the same claimants. Operating profit after tax belongs to everybody who financed the operating business, so the capital charged has to be everybody's capital in that same business. Writing both figures out in full before the subtraction, and asking who has already been paid out of the top line, is the single habit that catches nearly every version of this error.

What happens to economic profit when the return equals the cost of capital?

Economic profit becomes exactly zero, and the interesting part is what the company looks like at that moment.

Hold Sankalp's invested capital at Rs 12,00,00,00,000 and its cost of capital at 12.00 per cent, so the capital charge stays at Rs 1,44,00,00,000 whatever else happens. Now imagine the return on that capital is not 15.00 per cent but 12.00. Operating profit after tax would be Rs 1,44,00,00,000. Rs 1,44,00,00,000 is a large, respectable-looking figure. The figure would appear on an income statement as a profit, the company would pay tax on it, and nobody reading the accounts would find anything odd about it. Economic profit at that point is nil. The business earned exactly what the money inside it costs and produced nothing beyond that.

Below that point the measure goes negative while the profit stays positive. At a 9.00 per cent return the company earns Rs 1,08,00,00,000 of operating profit after tax and an economic profit of minus Rs 36,00,00,000. Both statements are true at the same time, and only one of them appears anywhere in the reported accounts. A company can be profitable and be producing nothing, and that is the single gap this measure exists to close.

The relationship between the return and the answer is a straight line, and it is worth seeing as a line rather than as a table. Economic profit is invested capital multiplied by the spread, so with the capital held still it moves in a straight line with the return, crosses zero at exactly one place, and that place is the cost of capital. Nothing else on the line decides the sign.

What does the same measure give five years later?

Sankalp's forecast puts Rs 1,00,00,00,000 of net new capital into the business in each of five years. How much growth costs and what new capital is assumed to earn is settled elsewhere; take the figures as given. By Year 5 invested capital is Rs 17,00,00,00,000 and operating profit after tax is Rs 2,70,00,00,000.

Run the same subtraction on the same convention used at Year 0. The convention matches a year's profit against the capital base of that year. The capital charge is Rs 17,00,00,00,000 at 12.00 per cent, being Rs 2,04,00,00,000. Economic profit is Rs 2,70,00,00,000 less Rs 2,04,00,00,000, and the difference is Rs 66,00,00,000. Economic profit has risen from Rs 36,00,00,000, and the first question to ask about any such rise is where it came from.

Here it splits exactly in two, with no remainder. The original Rs 12,00,00,00,000 of capital is still earning 15.00 per cent against a 12.00 per cent charge, so it still contributes a 3.00 point spread, being Rs 36,00,00,000. The Rs 5,00,00,00,000 added over the five years is assumed to earn 18.00 per cent against the same 12.00 per cent charge, a spread of 6.00 points, contributing Rs 30,00,00,000. Adding them gives Rs 66,00,00,000 to the rupee. The entire rise of Rs 30,00,00,000 is the new capital's own wider spread, and it rests on an assumption rather than on anything anybody has measured.

The 18.00 per cent is the most load-bearing figure in the whole forecast and the easiest to print without noticing, so the assumption is worth stating plainly. The forecast assumes capital put into the ground from Year 1 onwards earns 18.00 per cent while the capital already there earns 15.00. No evidence for that is recorded anywhere in the case. The assumption is narrower than it looks. The operating profit margin in this forecast is a flat 15.00 per cent of revenue on old capital and new alike, so the 18.00 per cent is not a claim about margins at all. Rs 1,00,00,00,000 of new capital is assumed to buy Rs 1,20,00,00,000 of revenue, or 1.20 turns. The existing Rs 12,00,00,00,000 supports Rs 12,00,00,00,000 of revenue, or 1.00 turn. At the same margin, 1.20 turns gives 18.00 per cent and 1.00 turn gives 15.00. The whole of the assumption is that new capacity spins faster than old capacity, and stating it that way makes it something a reader can actually argue with.

The return on invested capital at Year 5 follows from the same split: Rs 2,70,00,00,000 over Rs 17,00,00,00,000 is 15.88 per cent when rounded for printing, up from 15.00 per cent, and the spread widens to 3.88 points on the same basis. Both figures are computed on the unrounded values and printed to two decimals.

Which year's capital does the charge run on, and does it matter?

The choice matters by a knowable amount, and the honest treatment is to state which convention was used rather than to hope nobody asks.

There are two defensible ways to charge a forecast year. One is to charge the capital the year opened with, on the reasoning that a year's profit was earned by the capital that was already there when it started. The other is to match each year's profit to that year's own capital base, and the Year 0 and Year 5 figures above did exactly that. Neither is wrong. The two conventions give different answers, and the difference on this company is exactly one year's charge on one year's addition.

Run both across the forecast. Under the opening capital conventionCharging a year for the capital it started with, rather than for what it ended with., Year 1 is Rs 1,98,00,00,000 less Rs 12,00,00,00,000 at 12.00 per cent, or Rs 1,98,00,00,000 less Rs 1,44,00,00,000, being Rs 54,00,00,000. Under the same-year convention it is Rs 1,98,00,00,000 less Rs 13,00,00,00,000 at 12.00 per cent, being Rs 42,00,00,000.

Three things fall out of that table and all three are exact. Profit rises Rs 18,00,00,000 a year while the charge rises Rs 12,00,00,000, being 12.00 per cent on the Rs 1,00,00,00,000 added, so both columns rise by Rs 6,00,00,000 a year. The two columns sit exactly Rs 12,00,00,000 apart in every year, and that gap is the same one year's charge. And Year 5 reads Rs 78,00,00,000 on the opening convention and Rs 66,00,00,000 on the same-year one, so the Rs 66,00,00,000 quoted in the section above is the same-year figure and is labelled as such wherever it appears.

A convention is not an opinion and it is not a rounding. Two people using different ones will differ by a predictable amount, and the only defect is failing to state which one was taken. Year 0 was itself computed on the same-year convention, so every comparison of Year 0 with Year 5 above uses that one, and the opening convention is set out alongside it so the difference is visible rather than lurking.

Can a company grow its profit every year and create no economic profit?

Yes, and it is not an edge case. The measure is a spread multiplied by a base. If the spread is nil, no amount of base produces anything. A company that reinvests heavily at a return exactly equal to its cost of capital grows its profit every single year, reports rising earnings, and produces economic profit of nil in every one of those years.

Take Sankalp's own capital and run the thought experiment. Suppose the Rs 1,00,00,00,000 a year of new capital earned 12.00 per cent instead of an assumed 18.00. Operating profit after tax would rise by Rs 12,00,00,000 every year rather than Rs 18,00,00,000, so profit would still be growing, and every rupee of that growth would be swallowed exactly by the charge on the capital that bought it. Economic profit would sit still. Growth would be visible in the accounts and absent from this measure.

The household version is a shop that opens a second branch by pledging the house. The second branch turns over money, adds to the takings, and the interest and the risk on the pledged house eat precisely what the branch brings in. The takings line grows. Nothing has been produced. Growth is only worth having when the capital funding it earns more than that capital costs. The spread comes before the size of anything.

Koller, Goedhart and Wessels put growth, return on invested capital and value into a single expression, and that expression is the reason economic profit is built the way it is: the measure will not let growth count as an achievement without first charging for what the growth consumed. Turning a stream of such readings into a valuation is a separate exercise, and it needs a discounting step that a single year's subtraction does not contain.

What does economic profit do that a percentage return cannot?

Economic profit refuses to be improved by shrinking, and that is very nearly the whole of the difference.

A return on invested capital is a ratio. Ratios can be lifted by removing the worst part of the denominator. Removing it is a real and legitimate management action, and it makes the return on invested capital the easiest number in corporate finance to flatter. Suppose an invented business runs two divisions, one earning 20.00 per cent on its capital and one earning 13.00 per cent, both against a 12.00 per cent charge. Sell the second. Removing the weaker half leaves the better ratio, so the reported return on invested capital rises immediately. But the second division was earning a 1.00 point spread on real capital, and those rupees of economic profit have now left with it. The percentage went up and the rupee measure went down.

None of that is an argument for keeping the division. The sale might be exactly the right one for a dozen reasons the measure does not capture. The argument is for never reading the percentage alone. A percentage cannot tell whether the business got better or simply got smaller. Read together, the two measures answer the two halves of the question: the return says how well the capital is doing and the rupee figure says how much of it there is to do well with.

The other half of the same point is that the rupee measure depends on the size of the base, so two businesses with identical returns are not in identical positions. Sankalp's 3.00 point spread on Rs 12,00,00,00,000 gives Rs 36,00,00,000. The same 3.00 point spread on a business carrying Rs 2,00,00,00,000 of invested capital gives Rs 6,00,00,000. The return on invested capital is exactly 15.00 per cent in both, and one produces six times what the other produces. Neither fact is more true than the other; they are answers to different questions.

Where does this calculation go wrong in practice?

Almost never through carelessness. The calculation goes wrong when somebody careful picks a figure that is real, published and correctly labelled, and puts it on the wrong side of the subtraction. On this company the commonest version produces an answer so tidy that it looks like a discovery.

What can economic profit not show?

Quite a lot, and a measure this tidy earns its keep only if the limits travel with it.

Economic profit is not cash. Economic profit for Year 0 is Rs 36,00,00,000; Sankalp's Year 1 free cash flow to the firm is Rs 98,00,00,000. The two figures are different objects built a different way from different lines, and neither is a version of the other. A company can produce a positive economic profit in a year in which cash left the building, and the reverse happens just as often. Where the two part company is settled elsewhere.

Economic profit is not a value. A value discounts a stream of readings at a rate over time, and a single subtraction does neither. A single year's economic profit is a reading for one year on one capital base under one convention. Turning a stream of such readings into a value is a separate exercise with its own machinery, covered separately.

Economic profit moves entirely with the rate it was handed. The 12.00 per cent used throughout is the company's own locked figure, taken as a given, and building a cost of capital is covered separately. Move it to 13.00 per cent and the charge becomes Rs 1,56,00,00,000 and the answer falls to Rs 24,00,00,000, a third of it gone on one point of rate. Any economic profit figure is only as firm as the cost of capital somebody chose for it. The rate belongs beside the answer and never buried.

The measure also says nothing about why. The measure records that Sankalp produced Rs 36,00,00,000 in Year 0 and does not say whether that came from pricing, from the aftermarket business, from a supplier contract or from an accounting policy. And it is a single-period reading on a business that runs on multi-year assets: a company that has just commissioned a large plant carries the full capital in the denominator and only part of the profit that plant will eventually produce, so its economic profit looks worse than the underlying business is. None of that is a defect in the arithmetic. All of it is a reason to read one year's figure as one year's figure.