Skip to content

← Money

Spreading It

Chapter 25 · Owning a Share of Things

In this chapter 6 sections
  1. Twenty companies are not twenty coins
  2. Kinds of business
  3. A fund that holds many for you
  4. Lending instead of owning
  5. Short papers, and where they all trade
  6. Try this

Take the coin bet from Chapter 21 once more: heads you win Rs. 1,500, tails you lose Rs. 1,000. It is a good bet, and most people still refuse it, because one toss can take the whole Rs. 1,000. Now change one thing. Instead of one toss for Rs. 1,000, make twenty tosses for Rs. 50 each: heads wins Rs. 75, tails loses Rs. 50. The odds haven’t changed, and neither has the average. Played either way, the bet leaves you about Rs. 250 ahead.

What has changed is everything else. With one toss, you lose money half the time, and when you lose, you lose it all. With twenty, you come out behind only if seven or fewer of the twenty land heads, which happens about one time in eight. Eight heads leaves you exactly where you started; nine or more puts you ahead. And to lose the whole Rs. 1,000, all twenty would have to land tails, a chance of about one in a million. Same money, same odds, same average. Splitting it simply makes the bad ends rare.

That is the whole idea of this chapter, and it is older than any stock exchange. People say it as “don’t put all your eggs in one basket”. What they mean is that a single unlucky thing shouldn’t be able to sink everything you have.

Twenty companies are not twenty coins

Shares are not coins, though, and the difference matters. Coins don’t know about each other. One coin landing tails tells you nothing about the next. Companies do know about each other, in a way: they share the same country, the same interest rates, the same customers with the same amount of money in their pockets. When a whole economy has a bad year, most companies have a bad year together. So a company’s ups and downs come in two parts. One part is its own luck: a manager who chooses well or badly, a factory that burns down, a product people love. The other part is the market’s luck, which every company shares.

The American economist Harry Markowitz worked out in 1952 what this means, and it won him a share of the Nobel Prize in economics in 1990. Spreading money over many companies cancels out their own luck, because one company’s fire is another’s good year. But it can’t cancel the market’s luck, because that falls on all of them at once. What decides how risky a basket is, then, isn’t how risky each company is on its own. It is how much the companies rise and fall together. Spreading can make the risk much smaller, but never nothing.

The widget below plays this out with made-up numbers. A hundred savers each put Rs. 1 lakh into one company. A hundred more put Rs. 1 lakh each into twenty companies, Rs. 5,000 in each, and leave it there. Every year the whole market does one thing, the same for every company, and each company adds its own luck on top. Now and then a company fails and is worth nothing from then on.

Try it

One company or twenty

Each dot is a made-up saver who put in Rs. 1 lakh; the box holds those left with nothing. The returns and failures are made up, not any market’s history, and one company and twenty have the same average.

Rs. 1 lakh put inOne company eachTwenty companies each0.51251020nonewhat each saver has, in lakh rupees
Middle saver, one company
Rs. 1.59 lakh
Middle saver, twenty
Rs. 2.81 lakh

After 30 years, 45 of the hundred one-company savers have less than the Rs. 1 lakh they put in, and 35 of them have nothing left. Of the hundred twenty-company savers, one has less, and none has nothing.

After thirty years, the one-company savers are scattered across the whole scale. Two have done wonderfully, with more than Rs. 20 lakh each, and nine in all have more than Rs. 10 lakh. But 45 of the hundred have less than they put in, and 35 have nothing at all, because their one company failed somewhere along the way. The twenty-company savers sit in a tight bunch. Only one of them has less than they put in, and none has nothing: for that, all twenty of a saver’s companies would have had to fail. Their middle saver ends with more than the one-company middle saver too, since a lone saver’s middle is dragged down by all the companies that died.

Year 20 tells a different story. The market in this made-up history has just had three bad years in a row, and 38 of the twenty-company savers are below what they put in. Spreading didn’t save them from that. It saved them from the one company that fails. It can’t save anyone from the whole market falling, though once the market climbs again, three years later, nearly all of them are back above. Real companies show why the one company matters so much: in one broad American index, four in ten lost most of their value and never came back, while nearly all the index’s gains came from a small share of big winners (Chapter 21). A person with one company is mostly betting on which kind theirs will be. A person with twenty has, very likely, a couple of the winners and a few of the losers, and ends up with something close to the average.

Kinds of business

Twenty companies can still be one basket, if they are all the same kind of company. Twenty banks rise and fall with the same interest rates and the same rules from the central bank. Twenty hydropower companies share the same rivers and the same rains. What hurts one usually hurts the rest. Real spreading means companies whose fortunes don’t move together: a bank, a power company, an insurer, a maker of things, a hotel.

That is harder in Nepal than in most places. In mid-October 2026, 305 companies were listed on NEPSE (Chapter 23). Of those, 132 were banks, other financial institutions and insurers, and 112 were hydropower companies. By value, the banks and insurers alone made up more than half the whole market, and hydropower another sixth. A person who picks twenty shares at random from that list is mostly buying banks and rivers.

The same thinking stretches past shares. A family that keeps everything in one plot of land, or in one cooperative, or in one business it runs itself, has one basket. Spreading means putting money in things that fail for different reasons: some in a bank, some in shares, some lent, perhaps some in land or gold, which Chapter 26 takes up. Each can go wrong, but they are unlikely to all go wrong in the same year.

A fund that holds many for you

Buying twenty companies yourself, in sensible amounts, takes a fair sum of money and a lot of watching. A mutual fund does it for you. It is pooling again, the old idea behind a ढुकुटी or an insurance company: many people put their money together, and the pool buys what none of them could buy alone. A fund manager, licensed by SEBON, buys shares, debentures and deposits with the pooled money. Each saver holds units of the fund, and owns a slice of everything it holds, in proportion to the units. In Nepal the word is सामूहिक लगानी कोष, a collective investment fund.

Nepal’s rules make a fund spread by law. It may not put more than a tenth of its money into the papers of any one company, or hold more than a tenth of any one company’s shares. It may not lend money, and it may borrow only from a bank, at most a fifth of what it holds and for no more than six months, to pay savers who want their money back or to pay out its earnings. Its holdings are kept by a separate depository, and a fund supervisor watches over the manager, each paid a fee of its own.

What a unit is worth has a name: the net asset value, or NAV. It is everything the fund owns, less everything it owes, divided by the number of units. Suppose a made-up fund owns shares and deposits worth Rs. 12.3 crore, owes Rs. 30 lakh in fees and bills, and has 1 crore units. Each unit is worth Rs. 12 crore divided by 1 crore, or Rs. 12. If it first sold its units at Rs. 10, it has grown. If its shares fall by a tenth tomorrow, so does most of the NAV. A fund is not a deposit: nothing promises the money back, and when the market falls, the fund falls with it, as the basket did in year 20.

The pooling costs something, and the cost comes every year, good or bad. Nepal’s rules let a fund manager take up to 1.5% of the fund’s value a year, and the fund’s supervisor and its depository up to 0.2% each, nearly 2% in all. That sounds small. But it compounds, the way interest does in Chapter 10. Rs. 1 lakh growing at 8% a year becomes about Rs. 10 lakh in thirty years. Take 1.9% of its value away each year and it becomes about Rs. 5.7 lakh instead. A fund has to pick its shares well enough to earn back its own fees before the saver gains anything from the picking.

Funds come in two shapes. An open-ended fund has no end date and no fixed number of units. Anyone can buy new units from the fund, or sell units back to it, at that day’s NAV, which the fund must publish every day. The fund grows when savers buy and shrinks when they sell, and it may charge a small fee on selling back. A closed-ended fund sells a fixed number of units once, in a public issue like a company’s IPO, and then runs for a fixed term. Its units are listed on NEPSE and traded there like shares, so their price is whatever buyers will pay, which needn’t be the NAV. Around the world, closed-ended funds often trade below what their holdings are worth, and sometimes above. When the term ends, the fund sells everything, pays its bills, and shares out what is left among the units.

Open-ended funds make one more thing easy. A saver can put in the same sum every month, a plan Nepal’s funds call a SIP, a systematic investment plan. In months when the NAV is low the sum buys more units, and in months when it is high, fewer. Nobody has to guess the right month to buy.

Nepal’s funds are growing quickly. In the year to mid-July 2025, SEBON registered ten new funds, four open-ended and six closed-ended, worth about Rs. 10 billion together, against six the year before, and let funds already running sell Rs. 5.5 billion more in units. In the three months to mid-October 2026, mutual funds made up nearly half of all the money SEBON approved companies to raise from the public.

Lending instead of owning

A share is a piece of a company. A debenture is a loan to it. Chapter 21 split everything people put money into between those two families, owning and lending, and this is the lending side. A company that wants to borrow from many people at once sells debentures, ऋणपत्र: papers that promise to pay a fixed interest every year or half-year and to repay the whole sum on a fixed date. Governments do the same, and their papers are usually called bonds. In Nepal a debenture is usually named by how long it runs and what it pays, so “a 7-year, 9% debenture” tells you most of what it is.

The yearly interest is called the coupon. The name is a leftover from paper bonds, which came with small tickets printed along the edge, one for each payment. On the day each payment was due, the owner cut off a ticket and handed it in for the money. The date the whole sum comes back is the maturity, and the sum is the face value.

A lender is paid before an owner. A debenture holder gets the interest whether the company had a good year or a bad one, and if it fails, lenders are paid back before the owners get anything, since the owners stand last in the queue (Chapter 22). So debentures swing much less than shares, and usually earn less over the years. They aren’t safe, though. A company can fail so badly that even its lenders lose. That is why a debenture carries a grade from a credit rating agency, AAA for the safest down to D for one already unpaid (Chapter 16). A company with a low grade has to offer a higher coupon to find anyone willing to lend.

The surprise about bonds is what happens to their price. A bond’s coupon is fixed, but the rate everyone else pays is not, and a bond can be sold to someone else before it matures. Take a made-up debenture with a face value of Rs. 1,000, paying 10%, or Rs. 100 a year, for five years. While new loans also pay 10%, it is worth Rs. 1,000. Now suppose rates rise and new debentures pay 12%. Nobody will pay Rs. 1,000 for a paper that earns Rs. 100 when Rs. 1,000 can buy one that earns Rs. 120. What a buyer will pay is the backward sum of Chapter 10, done for each payment: each Rs. 100 and the final Rs. 1,000 brought back to today at 12%. That comes to about Rs. 928. If rates fall to 8% instead, the same paper is worth about Rs. 1,080. When rates rise, bond prices fall, and when rates fall, they rise. If the coupon is paid in two halves, Rs. 50 every six months, the sum is done in half-years at half the rate, and the price hardly changes, about Rs. 926 at 12%.

The two ends of that idea are worth seeing. A zero-coupon bond pays no interest at all along the way, only the face value at the end, so it is sold for less than its face. Chapter 10’s Rs. 1 lakh promised in ten years, worth about Rs. 61,400 today at 5%, is exactly such a bond. At the other end, a perpetual bond pays its coupon forever and never repays at all, which makes it Chapter 10’s perpetuity: Rs. 100 a year at 10% is worth Rs. 1,000, and at 12%, only about Rs. 833. The longer a bond runs, the further its price moves when rates change. A person who holds a bond to maturity gets the face value back whatever happened to its price in between. One who has to sell early takes whatever price the day gives.

Governments borrow too, and Nepal’s borrows a great deal. For short loans the government sells treasury bills, which run for 28, 91, 182 or 364 days. A treasury bill pays no coupon. It is sold at auction for less than its face and repaid at face, a short zero-coupon bond. In mid-October 2026 the average rate on 91-day bills was just 1.01% a year. For longer loans the government sells development bonds, विकास ऋणपत्र; in 2026 it offered them for as little as three years and as long as fifteen. By mid-July 2026 it owed about Rs. 1,000 billion on development bonds, and about three-quarters of it to the commercial banks. Smaller savings bonds are sold to ordinary people too, including some for Nepalis working abroad. The government is the borrower least likely to fail to pay in its own rupees, which is why its rates are usually the lowest.

Short papers, and where they all trade

Beside treasury bills sit a family of other short papers, each a promise of money within a year: the promissory note, the bill of exchange, the certificate of deposit and the letter of credit that banks deal in (Chapter 11), and two more. Commercial paper is a large company’s own short IOU, usually for a few months, with nothing behind it but the company’s good name. A banker’s acceptance is a bill of exchange that a bank has signed to say it will pay, so it trades as a promise of the bank rather than of the buyer who ordered the goods.

Papers like these, a year or less, trade in what is called the money market: banks and big companies lending to each other for days or months, to cover the gaps between money coming in and going out. Shares, debentures and bonds of many years trade in the capital market, where money is raised to build things that take years to pay for. A third kind, the commodity market, trades goods themselves: grain, metals, oil, often as promises to deliver them later at a price fixed now.

There are also two ways a trade can happen. On an exchange such as NEPSE, everyone’s orders meet in one place, under one set of rules, and every price is public. Over the counter, two parties deal with each other directly, and the price is whatever they agree, often not published. The trouble with dealing directly is that a buyer may have to wait for a seller to turn up. A market maker fills that gap. It is a firm that stands ready, all day, to buy at one price and sell at a slightly higher one, so anyone can trade at once. It earns the small difference between the two, and takes the risk of holding whatever nobody else wants that day. In Nepal, some banks and financial institutions are licensed as market makers in the government’s papers.

All of this, the funds, the debentures, the bills and the bonds, is spreading in one form or another. A fund spreads money over companies. Lending instead of owning spreads it between the two families, so that a fall in shares doesn’t take everything. Splitting money this way never makes it safe: the whole market can still fall, a fund still charges every year, and a bond can still lose value before it matures. What it does is make sure no single company, and no single kind of risk, can take everything at once. Over many years, that is most of what keeps a saver in the game. Crowds push prices too high and too low (Chapter 24), and nobody knows in advance which way they will run next; spreading is the quiet answer to not knowing.

Try this

  • Toss a coin twenty times and count the heads. Would the Rs. 50 bets have left you ahead? Do it three times.
  • Look up NEPSE’s list of listed companies by sector. If you picked one company from each of five different sectors, which five would they be?
  • Find any closed-ended fund’s last traded price on NEPSE and its latest weekly NAV, which every fund must publish. Is the price above or below the NAV?
  • Work out a made-up fund’s NAV: it owns Rs. 5 crore, owes Rs. 20 lakh, and has 40 lakh units.

Found a mistake? Please tell us.