Skip to content

← Accounting

Is It Worth It?

Chapter 25 · Looking Ahead

In this chapter 6 sections
  1. Two ideas and one loan
  2. The cash, year by year
  3. How soon is it back?
  4. A rupee later is worth less
  5. Which rate?
  6. Why the tests disagree

In February 1999 two finance professors at Duke University sent a long questionnaire to the finance chiefs of the 500 largest American companies and of some 4,440 firms in an association of finance executives, and 392 of them filled it in. One question asked how they decided whether to spend money on something new: a factory, a machine, a shop. Three in four said they always or almost always worked out a figure called net present value, the method every finance textbook recommends. But more than half also used a much older and simpler test, how many years until the money comes back, and among small companies it was nearly as popular as the textbooks’ method. The professors were surprised, they wrote, because textbooks had been complaining about that simpler test for decades.

Both tests are in this chapter, and so is the question of whether a sensible business should use both. They answer the same question, the one every owner asks before spending a lot of money now in the hope of more later: is it worth it?

Two ideas and one loan

By the end of the third year’s festival season, the mixing corner at Yasmin’s shop had found its limit. In the busiest weeks painters stood waiting while the helper weighed out tinters by hand, and some went to a bigger dealer rather than wait. Gaurav, the house painter who is Yasmin’s partner, had seen what bigger dealers use: a tinting machine that measures every color to a fraction of a milliliter from a row of canisters, and a shaker that mixes a tin in a few minutes. With the shaker, the voltage stabilizer it needs and the fitting, the two would cost the shop Rs. 500,000.

Yasmin had a different idea. A market town an hour up the road had grown fast and had no paint dealer yet. A room was to let there on an eight-year lease. Fitting it out with racks, a counter, a sign and a second scooter would also cost about Rs. 500,000, and Sujan, the wholesaler, would stock it on credit.

The shop couldn’t do both. The bank, which still charges it 12 percent on its first loan, would lend another Rs. 500,000 at the same rate, but not a million. So the partners had two good ideas and money for one.

A choice like this is what finance calls a capital investment decision, and the work of making it is capital budgeting: the budgeting of the last chapter, applied to things that last for years. It differs from the everyday choices of earlier chapters in four ways. The sums are large, often more than a year’s profit. The effects last a long time: the machine or the shop will shape the business for eight years. It is hard to undo, since a used machine or a shop’s fittings sell for a fraction of what they cost. And it rests on guesses about years that haven’t happened. A wrong price on one order costs a little and is soon forgotten. A wrong machine is a mistake the shop lives with.

Choices of this kind come in several sorts, and the sort changes how they are judged. Some are independent: a billing computer for the counter could be bought whatever happens to the machine, so each is judged on its own. Some are mutually exclusive, like the machine and the second shop: taking one means giving up the other, so they must be compared. Some are dependent, or contingent, because one can’t be done without another. The second shop needs a second scooter, and the machine needs its stabilizer, so each of those is counted as part of its project, never judged alone.

They can also be sorted by what they do for the business. The second shop is an expansion: more of what the shop already does, in a new place. If Yasmin started selling tiles and bathroom fittings, a line she has never sold, that would be diversification, spreading the business across different things so that one bad season hurts less. When the delivery scooter of the first year wears out, buying another is a replacement. The machine is a little of two: it replaces the corner’s hand mixing with a better way, which some books call modernization, and it lets the corner mix more tins, which is expansion.

Whatever the sort, the steps are much the same. Someone has an idea. The idea is turned into numbers: what it will cost, and how much cash it will bring in each year of its life. The numbers are judged by one test or several. One project is chosen, or none. It is carried out. And later, once the money has started to come in, someone compares what happened with what was forecast, which is the step most businesses forget and the one that makes the next forecast better.

The cash, year by year

The numbers that matter here are cash, not profit. A capital decision is about money: how much leaves the shop now and how much comes back, and when. So the machine’s price is counted once, in full, on the day it is paid. Its depreciation, the cost spread over its years, is left out, since it is only a way of dividing up the price already counted; counting both would count the machine twice. Tax is left out too, to keep the sums plain. A real appraisal takes each year’s tax off its cash, and since depreciation lowers the tax, that is where it comes back in.

Gaurav’s guess was that with the machine the corner would mix and sell about 60 more tins a month, the year round: no queue in the busy weeks, and in every month the painters who now go to bigger dealers for an exact shade, one the machine can match again years later. Each extra tin sells for Rs. 850, and the cash it costs is its base at Rs. 590, what Sujan has charged since कात्तिक, its tinters and its sticks and labels, Rs. 670 in all, leaving Rs. 180 of cash. The helper’s time isn’t counted, as it was when the last two chapters worked out a tin’s cost: she is paid by the month whether she mixes or not, so her minutes cost the shop no extra cash, and with the machine a tin takes her a few minutes instead of twelve. The machine also wastes less tinter, about Rs. 10 a tin, on all 260 tins a month. Against that, its electricity and servicing cost about Rs. 10,800 a year. So each year brings 60 × 12 × 180 = Rs. 129,600 from the extra tins and Rs. 31,200 of tinter saved, less Rs. 10,800: Rs. 150,000 a year, for the eight years the machine should last, and nothing at the end.

The second shop’s cash would come in a different shape. At first it would have the town to itself and sell fast. Then, as the town grew, other dealers would surely come, and its share would shrink. Yasmin’s guess was Rs. 250,000 of cash in its first year, after paying Sujan, the rent and a shopkeeper, then 220,000, 170,000, 130,000, 90,000, 60,000, 40,000 and 30,000 in the last year of the lease. Every one of these figures, for both projects, is a guess. That is not a weakness of the method. It is what deciding about the future is like.

How soon is it back?

The oldest test is the plainest. Add up the cash as it comes in, year by year, and see how long it takes to reach what was spent. That is the payback period.

The second shop is quick. Its first two years bring Rs. 470,000, and the last Rs. 30,000 of the cost comes back early in the third year: a payback of about two years and two months. The machine’s Rs. 150,000 a year takes three years to bring back Rs. 450,000 and a third of the fourth to bring back the rest: three years and four months. By payback, the shop wins.

There is a lot to like in that answer. Anyone can work it out, and anyone can follow it. It says how long the money is tied up and at risk, and for a small business that matters: if a project’s cash doesn’t come early, a shop with little to spare may not survive to see the later years, or may have nothing left to try the next idea. Some finance writers argue that this is why small firms are right to use it. The Duke survey looked for signs of that and found none: payback was no more popular with firms that had borrowed heavily or had poor credit. It was most popular with older bosses who had been in the job longest and had no business degree, which suggested to the professors that a lack of training in finance, more than good reason, kept it going.

But look at what it leaves out. Payback stops counting the moment the money is back. After that, the machine still has four years and eight months to run, and over its whole life it brings Rs. 700,000 more than it cost. The shop’s whole life brings Rs. 490,000 more. Payback can’t see that difference at all. It also treats a rupee in the third year exactly like a rupee in the first. And it gives no line between good and bad: a business must choose its own limit, three years or five, and nothing in the method says which.

A rupee later is worth less

The second thing payback ignores is the most important idea in this chapter. A rupee today is worth more than a rupee in a year, because a rupee today can be put to work. At 12 percent a year, Rs. 100 now grows to Rs. 112 by next year. Turn that round, and Rs. 100 promised a year from now is worth only what would grow into Rs. 100 in a year: 100 ÷ 1.12, about Rs. 89.29 today. Two years away it is worth 100 ÷ 1.12², about Rs. 79.72, and eight years away about Rs. 40. Money grows by interest on interest, which Money, for the Curious, Chapter 10 follows year by year, and bringing later money back to today is the same sum run backward. Finance calls it discounting, बट्टाकरण, and calls this whole idea the time value of money.

So the machine’s Rs. 150,000 in its eighth year is worth only about Rs. 60,600 today. Net present value, NPV for short, does that for every year of a project. Each year’s cash is multiplied by that year’s discount factor, 1 ÷ 1.12 for the first year, 1 ÷ 1.12² for the second and so on, which brings it back to what it is worth today. The present values are added up, and the cost paid today is taken off.

Year Discount factor at 12% Machine: cash Today’s value Second shop: cash Today’s value
1 0.8929 150,000 133,935 250,000 223,225
2 0.7972 150,000 119,580 220,000 175,384
3 0.7118 150,000 106,770 170,000 121,006
4 0.6355 150,000 95,325 130,000 82,615
5 0.5674 150,000 85,110 90,000 51,066
6 0.5066 150,000 75,990 60,000 30,396
7 0.4523 150,000 67,845 40,000 18,092
8 0.4039 150,000 60,585 30,000 12,117
All eight years 1,200,000 745,140 990,000 713,901
Less the cost today 500,000 500,000
Net present value 245,140 213,901

NPV=C11.12+C21.122+⋯+C81.128−500,000 \text{NPV} = \frac{C_1}{1.12} + \frac{C_2}{1.12^2} + \dots + \frac{C_8}{1.12^8} - 500{,}000

The rule is simple. A project whose NPV is above zero brings back more than it cost, in today’s money, even after paying 12 percent a year for the use of the money; below zero, it doesn’t, and the shop would do better to leave the money where it is. The NPV itself is how much better off the project should leave the shop, measured in today’s rupees. Both projects pass: the machine by about Rs. 245,000, the shop by about Rs. 214,000. Either would be worth doing on its own. Since only one can be done, the rule for a choice like this is to take the one with the higher NPV, the one that leaves the business better off. That is the machine.

Which rate?

Everything above depends on the 12 percent. Where does it come from?

It is meant to be what the money could earn elsewhere, or what it costs, whichever is higher. The safest place for spare money in Nepal is a bank deposit. In mid-October 2026 Nepal’s commercial banks were paying 3.07 percent a year on all their deposits, on average, and 4.74 percent on fixed deposits. That is the least any project should beat: if a machine can’t earn more than a fixed deposit, the money might as well sit in the bank. The same banks were charging 6.31 percent on their loans, on average.

But the shop isn’t spending spare money. It is borrowing, at 12 percent, nearly twice that average. An average mixes every kind of loan and borrower, and Nepal Rastra Bank’s rules have each loan’s rate built from the bank’s base rate plus a premium for that kind of loan, as Money’s Chapter 14 explains; a small shop with little to pledge pays more than a big, safe borrower. Rates also move with time: in early 2023 the banks’ loans averaged about 13 percent, and a real shop borrowing in late 2026 would probably be offered less than Yasmin’s made-up 12. What matters is the principle. For Yasmin’s shop, the money costs what its own bank charges it, and the same 12 percent is what it would save by paying back part of its loan instead. So 12 percent is the right rate for this shop, and a project that can’t beat it would lose money even if every guess came true.

There is also the guessing to think of. A deposit pays what it promises; a project’s cash is a forecast, and the second shop’s depends on how fast other dealers arrive. Many businesses use a higher rate for riskier projects, so that a doubtful forecast has to clear a higher bar. The same point holds from the lender’s side: anyone valuing a Nepali company’s bond must say what return they would require, and why.

Try it

The Books: the two projects at any rate

Where a line meets zero, that project only just pays its way. The marks under the rates are the banks’ average deposit and loan rates and the shop’s own 12 percent. A made-up shop and figures.

NPV, Rs. lakh−1012345670510152025303.076.3112Rate, % a yearmachinesecond shop
Machine’s NPV
Rs. 245,146
Second shop’s NPV
Rs. 213,894

At this rate the machine is worth more today. Its money keeps coming for eight years, and later rupees still count for a good deal.

The rate changes the answer, not only the numbers. At the deposit rates, the machine’s lead is large, since later rupees are worth nearly as much as early ones and the machine has more of them. As the rate rises, later rupees count for less and less, and the shop’s early cash holds its value better than the machine’s long tail. The two lines cross at about 16 percent. A business that wanted 20 percent from its money would choose the shop.

Why the tests disagree

So payback said the shop, and NPV at 12 percent said the machine. They disagree because they look at different things. Payback looks only at the early years, and counts every rupee in them the same. NPV looks at every year, and counts each one for what it is worth today. A project with quick money that fades, like the second shop, looks best by payback. A project with steady money that goes on, like the machine, looks better by NPV, as long as the rate isn’t so high that its later years shrink to little.

There are other tests, and they lean one way or the other. The internal rate of return, IRR, is the rate at which a project’s NPV is exactly zero, where its line meets zero on the chart: about 25 percent for the machine, about 29 for the shop. It says how hard the money is working, and it is popular because a percentage is easy to compare with a loan’s rate. But a higher IRR doesn’t always mean more money, and here it favors the shop, which leaves the business less well off at 12 percent. Between two projects that can’t both be done, finance books agree that the NPV is the one to follow. The accounting rate of return, ARR, divides a project’s average yearly profit, after depreciation, by the average money tied up in it: Rs. 87,500 a year on an average of Rs. 250,000 for the machine, 35 percent. It is easy to work out from the accounts, but it uses profit, not cash, and ignores when the money comes. And payback can be worked out with discounted cash instead, which takes the time value into account but still stops counting at the moment the money is back. In the 1999 survey fewer than a third of companies said they used it.

The machine’s lead is thin, too. If it brought about three fewer extra tins a month than Gaurav guessed, the second shop would come out ahead even at 12 percent. And none of the tests can see everything. The second shop would need someone to run it. Yasmin can’t be at two counters, and Gaurav would have to give up the painting that brings the shop its painters, or they would hire a stranger to take the town’s cash. The machine would keep both partners where they are and give the painters exact shades they could match again a year later. Against it, a machine can break down, and its parts come from abroad. These are the things Cambridge’s course calls non-financial factors, and in a small business they often decide.

Here, the numbers and the rest pointed the same way. The partners borrowed the Rs. 500,000, and the machine arrived in माघ of the third year. A year later they would do the last step, and lay what it actually brought beside Gaurav’s 60 tins a month.

Deciding whether something is worth it comes down to a few plain steps. Count the cash, not the profit, year by year, for the whole of the project’s life. Bring every year’s cash back to today at a rate that says what the money costs or could earn elsewhere, and take off what is spent now. A project worth more than it costs is worth doing; of two that can’t both be done, the one worth more is the better choice. Payback says how soon the money is back, which matters to a small business, but it can’t see what comes after. Every one of these figures rests on guesses, and on the rate, and the rate is a choice that needs a reason.

Found a mistake? Please tell us.