Revenue Growth vs Reinvestment: What Each Rupee of Growth Costs

Revenue growth is an output. Reinvestment is the input that bought it. Sankalp Industrial Systems Limited, an invented manufacturer, puts Rs 1,00,00,00,000 of net new capital in every year of its forecast and gets Rs 1,20,00,00,000 of new revenue and Rs 18,00,00,000 of new operating profit after tax. The new capital therefore earns an assumed 18.00 per cent, being a 15.00 per cent margin multiplied by 1.20 turns.

A small printing press near a wholesale market printed more invoice books this year than last. The extra books are the first fact. The second fact is that the press bought a second machine in order to print them, and the machine was paid for months before the extra books were sold. Both facts describe the same twelve months. Only one of them appears on the line that everybody quotes.

Revenue growth comes first below, defined on its own terms, then reinvestment on its own terms, and only then the two set against each other. Most of the trouble here starts with a rough sense of both quietly merged into one idea about a company doing well, and pulling them apart again is the work.

What exactly is revenue growth, before anything is compared to it?

Revenue growth is a measurement taken on a single line of a single statement across a single period. The analyst finds the top lineThe revenue figure that sits at the head of a profit and loss account, counted before a single cost has been taken off it. for this year, finds it for last year, and states the difference either as rupees or as a percentage of where it started. Nothing else enters the calculation. No balance sheet is opened. No cash flow statement is consulted. Revenue growth is the cheapest measurement in finance to take, and that cheapness is exactly why it travels so far.

Revenue growth is a flow measured across a period, and it carries no information whatever about what the period cost. A business that added revenue by winning a contract, a business that added it by raising prices, and a business that added it by buying a competitor all produce the same shaped number. So does a business that added it by building a factory. The line cannot separate them and does not try to.

Sankalp Industrial Systems Limited makes industrial valves, precision castings and the parts and service that go with them. Its five year forecast opens on revenue of Rs 12,00,00,00,000 in the base year and then puts another Rs 1,20,00,00,000 on that line across each of the five years that follow, finishing at Rs 18,00,00,00,000. Because the rupees added stay identical while the base underneath them keeps rising, the growth rate falls in every year of the run, from 10.00 per cent down to 7.14 per cent, and the compound annual rate across the whole path is 8.45 per cent. The reason identical rupees added each year show a shrinking percentage is worked through in full separately, and that result is taken as settled here.

The rupee line matters more than the rate. Sankalp adds Rs 1,20,00,00,000 of revenue in Year 1 and again in Year 5, and those are the same rupees. The rate is a description of the rupees, and the rupees are the event.

And what exactly is reinvestment?

Reinvestment is a different kind of measurement altogether. Reinvestment is not read off one line. The figure is assembled from three lines, two of which sit on statements that a revenue growth calculation never opens.

The figure used throughout is net new invested capital, and it is built as capital expenditureSpending on plant, machinery and buildings, which lands on the balance sheet as an asset instead of inside the year's costs. less depreciationThe yearly write-down of an asset that was paid for once, so reported profit drops while no money leaves the business. plus the movement in net working capitalMoney locked inside the trading cycle: stock on the shelf and customer balances still due, less whatever suppliers are still owed.. Depreciation comes out because the capital expenditure that merely replaces a worn machine is not adding capacity, it is standing still. The working capital movement goes in because a business selling more has to carry more stock and wait on more customer balances before the money comes back. How that three line build is put together, and why each line belongs in it, is set out separately.

For Sankalp the answer is unusually clean, and it was built that way on purpose. Here is Year 1, three lines and one total.

The four years that follow run the same three lines to the same total. Across all five forecast years Sankalp sinks the identical Rs 1,00,00,00,000 of net new capital, and it is that flatness which makes the exchange visible at all.

Notice what has just happened to the reader's workload. Reading the revenue growth needed one line. Reading the reinvestment needed three, drawn from two statements, with a rule about which capital expenditure counts and which does not. The asymmetry in effort is a large part of why the revenue line gets quoted alone.

Which one is the input and which one is the output?

Reinvestment is the input. Revenue growth is the output. The capital goes into the ground first and the revenue arrives afterwards, and that order is not a convention: it is the physical sequence of what actually happens in a business. The printing press bought the second machine in March and printed the extra invoice books from June.

Held in that order, a second difference falls out of it. Reinvestment accumulates and revenue does not. Sankalp starts with invested capital of Rs 12,00,00,00,000 and adds Rs 1,00,00,00,000 to it in each of five years, so by the end of Year 5 the invested capital stands at Rs 17,00,00,00,000. Nothing was taken back out. The revenue behaves in the opposite way: the Rs 1,20,00,00,000 added in Year 1 is earned, collected and spent, and Year 2 has to earn it again before it can add anything on top.

Reinvestment is a stock that piles up while revenue growth is a flow that resets, and that structural difference is precisely why a forecast can move one of them without the other objecting. A spreadsheet holds the revenue row and the capital expenditure row in different places. Changing one does not touch the other, no total breaks, and nothing on the sheet knows that the two are supposed to be related at all.

What is the exchange rate between a rupee of capital and a rupee of revenue?

Now the two sides can be put together, and the figure that joins them is an exchange rate. Sankalp hands over Rs 1,00,00,00,000 of net new capital and receives Rs 1,20,00,00,000 of new revenue. The second of those over the first gives the rate: each rupee of new capital carries Rs 1.20 of new sales at this company, in every year of this forecast.

Inverted, the ratio reads as a price rather than as a yield. One divided by 1.20 is about 0.83, so roughly 83 paise of capital stands behind each rupee of sales added. Both readings describe one trade. A reader asking what the money bought reaches for the first reading. A reader asking what the sales cost reaches for the second. The pair is easy to swap by accident, so name the direction in the same sentence as the figure.

The exchange rate is not a rule of the industry, not a property of manufacturing, and not something Sankalp has demonstrated. The rate is what this forecast has been built to assume. A different assumption about the second machine would produce a different rate, and a reader is entitled to ask what evidence sits behind the one used here. For this forecast the honest answer is that no such evidence exists.

How does a rupee of capital become a rupee of profit?

The exchange rate stops at revenue, and revenue is not the thing anybody eventually values. So the chain runs one step further, and all three steps come from the same locked lines.

Step one is the exchange rate just established, where each rupee of new capital carries Rs 1.20 of new sales. Step two applies a margin to that revenue. Sankalp's operating profit after taxTrading profit taxed as though the business carried no borrowings at all, which keeps the financing side out of the figure entirely. rises by Rs 18,00,00,000 in each forecast year against Rs 1,20,00,00,000 of new revenue. The incremental margin is therefore 18 over 120, being 15.00 per cent. Applying that to the Rs 1.20 gives Rs 0.18. Step three is a division: Rs 0.18 of profit sitting on the Rs 1.00 of capital that produced it is the 18.00 per cent this forecast expects new invested capital to earn.

The three steps multiply out exactly, and the multiplication is worth doing on paper once. One rupee, times 1.20, times 0.15, is 0.18 rupees. Divide that by the one rupee started with and the answer is 0.18, the 18.00 per cent. There is no fourth step and no residual.

The exactness carries a consequence most readers miss. The return on new capital is not a separate assumption sitting alongside the revenue assumption and the margin assumption. The return is the product of the other two. Because the third figure is what the first two produce, two of the three cannot be held still while the third moves. Any forecast that appears to move only one of the three has in fact moved two.

Is 18.00 per cent against 15.00 per cent an assumption about margins?

It is not. The argument here is worth slowing down over. Any return on capital splits into two parts that multiply together: a margin, meaning how much profit each rupee of sales leaves behind, and a turnover, meaning how many rupees of sales each rupee of capital carries. Running that split down both sides of Sankalp, every bit of the difference turns out to sit in a single place.

On the capital already in the ground: revenue of Rs 12,00,00,00,000 against invested capital of Rs 12,00,00,00,000 is a capital turnover of exactly 1.00 times, and operating profit after tax of Rs 1,80,00,00,000 on that revenue is a margin of 15.00 per cent. Multiplied together, they make the return on the existing base 15.00 per cent. On the new capital: the margin is the same 15.00 per cent, and Rs 1,20,00,00,000 of revenue against Rs 1,00,00,00,000 of capital is 1.20 times. Multiplied together, they make 18.00 per cent.

The margin is identical on both sides, so the entire gap between 18.00 per cent and 15.00 per cent is an assumption that the new capital spins twenty per cent faster than the old. Not that products will sell for more. Not that costs will fall. Not that the tax charge will change. Purely that a rupee of new plant will carry Rs 1.20 of sales where an existing rupee carries Rs 1.00.

Saying it that way changes what a reader can do with it. A vague expectation that new investment will earn more than old investment is not something anybody can argue with. A specific claim that the new line will turn twenty per cent faster is something a reader can take to a factory manager, and it points at exactly what would have to be true: newer equipment running at higher utilisation, or the same equipment producing a mix that needs less capital per rupee sold. Neither condition is evidenced anywhere in the case. The 18.00 per cent is therefore what it has been all along: a choice made inside the forecast, not a measurement taken off the company.

Both margins in that table were worked out from the rupee lines rather than read off a label, and it is worth saying why that discipline matters. Take the base year. Earnings before interest and tax of Rs 2,40,00,00,000 on revenue of Rs 12,00,00,00,000 is a margin of 20.0 per cent. Taxing that at the company's own assumed effective rate of 25.0 per cent, a rate no statute sets, leaves Rs 1,80,00,00,000, and Rs 1,80,00,00,000 on Rs 12,00,00,00,000 is the 15.00 per cent used throughout. A descriptive label attached to a forecast can go stale while the rupee lines underneath it stay correct, and a copied label outlives the figure it once described. Compute the margin; never inherit it.

One caution about the tidiness of that first row. Sankalp's invested capital and its revenue both stand at Rs 12,00,00,00,000 in the base year, and that equality is what makes the existing turnover come out at exactly 1.00. The equality is a coincidence of the figures chosen for this case rather than anything normal, and a unit turnover is not a benchmark to carry away. Capital intensity varies enormously between a valve manufacturer, a software business and a retailer, and the split above is worth running on each of them separately.

When do the two come apart?

Everything so far has treated revenue growth and reinvestment as two ends of one transaction. Usually they are. But four specific situations break the link, and each one means something different about what the business did.

The first is revenue bought rather than built. A company that acquires another company gets its revenue line on the first day and pays a purchase price rather than a capital expenditure figure. The growth is real, the reinvestment measured as capital expenditure and working capital is untouched, and the two lines have nothing to say to each other. Companies that acquire regularly therefore report organic growthSales the business wins under its own name, as distinct from sales that arrive attached to a company somebody purchased. as a separate figure.

The second is a price rise. An auto rickshaw operator who raises the fare carries the same passengers in the same vehicle and collects more. No capital is needed. In a company the same event is a price increase pushed through the order book: revenue rises, the capital base does not move, and the return on new capital rises because there is profit in the numerator and almost nothing in the denominator. A price rise is a genuinely different event from adding a production line, and forecasting both with the same reinvestment assumption quietly treats them as the same.

The third runs the other way. Capital can be spent that produces no revenue at all: an effluent treatment upgrade, a fire safety rebuild, a mandated emissions control. The money leaves, the asset base rises, and the top line does not move. Reinvestment measured mechanically will show a company investing heavily and growing slowly. The pattern reads as a poor return, and it is in fact a different kind of spending altogether.

The fourth is negative reinvestment. If a business collects from its customers faster or holds less stock, working capital falls and cash comes back out of the trading cycle. receivablesAmounts already billed to customers and not yet collected, parked on the balance sheet until the payment lands. shrinking is capital returning, and in the three line build it appears as a negative movement that reduces net new invested capital. A company can therefore add revenue and consume less capital in the same year, and the exchange rate for that year is not a rate at all.

In all four cases the revenue line looks exactly the same and the capital behind it looks completely different. The second figure has to be printed for that reason alone.

Why does growing faster leave less cash this year?

Here is where the input and the output meet in the bank account. The cash a business generates before it pays anybody who financed it is its operating profit after tax less the capital it put back in. Sankalp's reinvestment does not move, so the subtraction is unusually simple: the free cash flow to the firm is just that year's profit less Rs 1,00,00,00,000.

Year by year the profit runs Rs 1,98,00,00,000, Rs 2,16,00,00,000, Rs 2,34,00,00,000, Rs 2,52,00,00,000 and Rs 2,70,00,00,000. Take the same Rs 1,00,00,00,000 off each and the cash runs Rs 98,00,00,000, Rs 1,16,00,00,000, Rs 1,34,00,00,000, Rs 1,52,00,00,000 and Rs 1,70,00,00,000. The cash climbs by the identical Rs 18,00,00,000 each year, the same step the profit takes, and it does so precisely because the reinvestment never moves.

Growth is paid for out of the same year's profit, so a company growing harder hands over less cash now in exchange for more later. The printing press that bought a second machine had a worse cash year than the one that did not, and a better one three years afterwards. Nothing about that sequence is a defect.

The point is worth saying carefully. The wrong reading of it is common and expensive. Low cash generation at a fast growing company is not by itself a weakness, and high cash generation at a slow growing one is not by itself a strength. Both are statements about where the company is in the trade, not about how good it is. The legitimate question is a different one: what is the capital buying, and is the return on it the return the forecast claims. Whether Sankalp should grow faster or slower is not something these figures settle, and no honest reading of them could.

What should be printed beside a revenue growth number, every time?

The capital that produced it. One extra column carries the whole answer.

A growth rate on its own is compatible with a business creating value and a business destroying it, and the two are indistinguishable from the line itself. Put the capital beside it and they separate immediately. Consider two manufacturers that each added Rs 1,20,00,00,000 of revenue this year. One used Rs 1,00,00,00,000 of new capital to do it, the other used Rs 3,00,00,00,000. At the same 15.00 per cent incremental margin the first earned 18.00 per cent on its new capital and the second earned 6.00 per cent. Their revenue growth is identical, the price they paid for it differs by three times, and nothing on the top line separates them.

The revenue line is never enough on its own, and the fix is not a better growth measure but a second number printed next to the one already in hand.