Skip to content

← Money

Too Good to Be True

Chapter 29 · Traps

In this chapter 6 sections
  1. Where a return comes from
  2. Fifty percent in forty-five days
  3. A pyramid, level by level
  4. Network marketing here
  5. Seven warning signs
  6. Try this

A few years before 2010, a company was offering Nepali families a deal. Pay Rs. 15,000, and if you died in an accident, your family would get Rs. 5 lakh. Pay Rs. 15,750 instead, and you would have free health checks and treatment for life, at hospitals it said it ran in 19 districts. And there was more: bring in other people, and you would earn a commission on what they paid. By 2010 more than 368,000 people, in Nepal and abroad, had paid it something like Rs. 3.8 billion between them.

The company was called Unity Life International, and in April 2025 Nepal’s Supreme Court sentenced its leaders to prison, most of them for two and a half years. Its members had been told about accident cover, hospitals and commissions. What they hadn’t been told was where the money for all of it would come from. That question is the whole of this chapter. Asked early, it could have saved many of them.

Where a return comes from

Every rupee anyone promises to pay you has to come from somewhere, and there are only two places it can come from. Either the money you hand over earns something, or someone else’s money pays you.

The first kind is ordinary. A bank lends your deposit to people who pay it interest, and passes part of that interest back to you. In mid-October 2026 Nepal’s commercial banks charged borrowers 6.31% a year on average and paid depositors 3.07%, or 4.74% on fixed deposits, as Chapter 11 shows; the gap pays the staff, the buildings and the bad loans. A company pays its owners a dividend from the profit it makes selling cement or electricity. A field pays its owner in rice, a flat in rent. In every one of these, you can name who is paying: a borrower, a customer, a tenant. And each payer can only pay so much, which is why honest returns are modest and why, as Chapter 21 explains, the higher ones come with a real chance of losing.

The second kind has no earner. If a scheme promises far more than anything real could earn, and nobody can say who the borrower, customer or tenant is, then the only person left to pay you is the next person who joins. That is the test, and it fits in one question: who is paying this, and from what?

It works because it doesn’t need you to understand the scheme’s story. Stories are easy to tell. A list of who pays is hard to fake, because the numbers have to add up.

Fifty percent in forty-five days

The best-known scheme of the second kind ran in Boston, in the United States, a hundred years ago, and the most careful short account of it is a judgment of the US Supreme Court, written in 1924 when the money was being shared out among the losers.

In December 1919, the court says, Charles Ponzi began borrowing money with $150 of his own. He gave each lender a written promise: lend him $100 now and get $150 back in ninety days. Then he paid the promises early, at forty-five days, so in practice it was 50% in forty-five days. His story was that he bought international postal coupons cheaply in some countries and sold them in others at double the price. It wasn’t true. The court’s words are plain: he made no investments of any kind, and all the money he ever had came from the people who lent it.

He also paid his agents 10% of what they brought in. So every $100 that came through the door cost him $160 within forty-five days, and the only place he could find $160 was in the next people’s $100. The court noticed what that meant. Ponzi owed more than he had from the start, and owed more the better his business did, because every new lender was a new debt bigger than the money that came with it.

Put that into a sum. The lenders’ 50% alone makes a debt that grows by half every forty-five days, which over a year is about 27 times, as Thinking Clearly works it out. Add the agents’ 10%, and every $100 becomes $160, so what he had to find grew 1.6 times every forty-five days, about 45 times in a year. To keep paying, the new money coming in had to grow just as fast. Compare it with a bank: 5% a year, spread over forty-five days, is about 0.6%. Ponzi was promising about eighty times that, every six weeks.

For a while the sum worked, because people kept coming. By the start of July 1920 he was taking in about a million dollars a week. In eight months he took in $9.58 million and wrote promises for $14.37 million. Then an official investigation began, he stopped taking new money on 26 July, and on 2 August Boston’s most widely read newspaper declared him hopelessly insolvent, owing far more than he could ever pay. Crowds filled the street outside his bank for eight days, trying to get their money out first. When the account ran dry, about five hundred people were left holding checks from him that couldn’t be paid. On 9 August his account was overdrawn, and the petition to make him bankrupt followed.

The court’s own case adds one more true detail. Some lenders had got their money back in the last frantic days, before the others. The court ruled that they had to hand it back, so that everyone owed would share what was left alike. In a scheme like this there is no money of your own inside it to rescue, only other people’s.

Notice what didn’t cause the collapse. It wasn’t the investigation, or the newspaper. Those only chose the date. A scheme that pays out more than it earns, and earns nothing, has to stop when new money stops growing fast enough, and new money always stops growing in the end. Why so many sensible people believed him anyway is the subject of Thinking Clearly, for the Curious, in its chapter on money traps. The sum is the part here: whoever joins last pays for everyone before them.

A pyramid, level by level

Ponzi’s lenders didn’t have to recruit anyone. In a pyramid scheme they do. You pay to join, and you earn when the people you bring in pay to join, and often a share of what their recruits pay too. The money still comes from the next people in. The difference is that the scheme makes every member find them.

The US Securities and Exchange Commission, which regulates investments in the United States, draws it with a simple rule: each member must bring in six new members. Start with one person and keep going. Here is the table, set beside Nepal’s 29,164,578 people counted in the 2021 census.

Level New members it needs Members so far Share of Nepal
1 6 7 under 0.01%
2 36 43 under 0.01%
3 216 259 under 0.01%
4 1,296 1,555 under 0.01%
5 7,776 9,331 0.032%
6 46,656 55,987 0.19%
7 279,936 335,923 1%
8 1,679,616 2,015,539 7%
9 10,077,696 12,093,235 41%
10 60,466,176 more than Nepal has

For six levels nothing looks wrong. The scheme has fewer members than a town, and everyone in it can point to people who joined after them. Then, in three more levels, it goes from a town to four Nepalis in ten. Level 10 needs more than twice the whole country. And that count includes every baby and every great-grandmother, so a real scheme would run out well before.

There is a second thing in the table, and it matters more. Look at any level and compare its new members with everyone so far. At level 5, the 7,776 newest are five in every six of the 9,331 members. At level 8, the same. However big the scheme gets, about five members in six are the newest ones, who have paid and are still waiting for their six. That counts a member as waiting until all six recruits have joined, which is how the widget below counts too; a real scheme may pay something for each recruit, but the newest members have brought in no one and had nothing back. And when the country runs out, the people left are too few to give the level above its six each; shared out, they come to fewer than two each, so that level is waiting too. At the widget’s last step, 93% of Nepal is waiting. Whenever it stops, and it has to stop, that is who is left holding the loss.

Asking each member to bring in fewer people only buys a few more levels. With two each, Nepal is used up at level 24; with ten each, at level 8. Try it below.

Try it

Run a pyramid

Each bar is everyone in the scheme so far; each line is 100 times the one below. Dashed: all of Nepal, babies included.

110010,00010 lakhNepal15101520level
Each brings in:
In it so far
259
Share of Nepal
under 0.01%

Level 3: the 216 newest have paid, 83% of everyone in it. Each is paid only once they bring in 6 more, 1,296 people in all.

Real pyramids never get near the end of the table. They stall at a few thousand or a few hundred thousand members, when recruiting gets hard in a town or a group of friends. But the shape is the same at any size. When it stops, the people at the bottom outnumber everyone above them put together, and their money has already gone up the levels.

Network marketing here

Not every company that sells through people is a pyramid. Some sell soap, toothpaste or health drinks through ordinary people who buy from the company and sell to their neighbors, and earn a share of what they sell. That is direct selling, or network marketing when sellers also earn on the sales of the sellers they bring in. The US Federal Trade Commission draws the line this way. If a company is not a pyramid, it pays you for what you sell to real customers, without your having to recruit anyone. If your income comes mostly from how many people you recruit, not from what is sold, it is a pyramid. And even in honest network marketing, it adds, most people make little or no money.

Nepal tried banning the whole business first. After Unity Life and others, the government declared multi-level marketing illegal, and on 28 March 2012 the Supreme Court upheld it, saying the public might be cheated; it left room for the business to be allowed under a proper law. That law came in 2017, the Direct Sale of Goods (Management and Regulation) Act, 2074 (वस्तुको प्रत्यक्ष बिक्री (व्यवस्थापन तथा नियमन गर्ने) ऐन, २०७४).

Under it, nobody may sell goods this way without a license from the Department of Commerce, now the Department of Commerce, Supplies and Consumer Protection (section 3). A licensed company must give the Department its plan of commissions and bonuses before using it (section 15), and its sellers must be at least 18 (section 2). Section 11 bans any business based on a pyramid, and lists what counts as one: having to become a member before you may buy; having to buy a bundle of goods to get the one you want; being unable to get back what you paid, in money or goods, unless you recruit other sellers under you; taking fees just for making new sellers, with no goods traded; luring members with gifts or commissions no business could really pay; a plan that discourages selling to ordinary customers; and selling something as a miracle that will bring unexpected gains or cure a disease medicine says can’t be cured. Section 12 adds that a company may not make bonuses depend on buying a fixed amount, and must let customers buy the items they want, one by one.

Running a pyramid is an offense under section 36, and section 37 sets the punishment: a fine equal to the money involved, or where that isn’t known Rs. 5 lakh to Rs. 10 lakh, or three to five years in prison, or both. Anyone who knows of one can complain, within ninety days, to the Department’s inspection officer or the nearest police office (section 38). The Consumer Protection Act, 2075, says the same from the buyer’s side: selling goods through more levels of sellers than the rules allow is an unfair trade practice (section 16(2)(त)).

Unity Life’s case shows how slowly this can go. In May 2010 the government declared its network business illegal and canceled its registration, and the police filed a case against 112 people connected with it. In 2014 the Lalitpur District Court convicted twelve of its leaders. In 2017 the Patan High Court acquitted them. In April 2025 the Supreme Court convicted nine, most for two and a half years in prison, and, as its spokesperson reported, ordered that the people who could prove what they had lost be paid back from the company’s assets. They were convicted under the laws of their own time, before the 2017 Act. Fifteen years passed between the ban and the verdict, and of more than 368,000 members, only 4,144 had come forward with a complaint.

Seven warning signs

The Securities and Exchange Commission lists seven signs that a scheme may be paying old investors with new investors’ money. Each is the missing earner showing through.

The first is high returns with little or no risk. Every real investment carries some risk, and the higher the return, usually the higher the risk, so the word “guaranteed” deserves the most suspicion of all. The second is returns that are too steady: real investments go up and down, and one that pays the same every month whatever the market does is being paid from somewhere other than the market. The third is an investment that isn’t registered, and the fourth a seller who isn’t licensed. In Nepal the lists to check are Nepal Rastra Bank’s for anyone taking deposits, SEBON’s for anyone selling shares or managing funds, the Nepal Insurance Authority’s for insurance, and the Department of Commerce’s for direct selling; Chapter 12 tells what happened to savers who trusted a name that wasn’t on the list they needed.

The fifth is a strategy that is secret or too complicated to explain. Postal coupons, foreign exchange, “an international trading platform”: if you can’t find out how the money is earned, you can’t find out whether it is. The sixth is trouble with the paperwork, statements with mistakes in them, or that don’t add up. And the seventh is trouble getting paid, especially when the scheme answers a request to cash out by offering even more to stay in. A scheme with no earner can’t let everyone leave, so it pays people to wait.

To those seven, the Commission’s page on pyramids and the Federal Trade Commission’s add the mark of a pyramid: the money comes from recruiting people, who pay to join or must buy goods to stay in, rather than from selling to people outside the scheme. Section 11 of Nepal’s Act names the same things.

None of the signs needs any expertise. Each one is a way of asking the same thing as Ponzi’s sum: if this money isn’t coming from a borrower, a customer or a tenant, where is it coming from? A real business can answer that question in a sentence. A scheme answers it with a story, a deadline or a bigger promise.

What a reader should take from this chapter is short. Every return is paid from something the money earns or from someone else’s money, and only the first can last. Ponzi’s notes cost him more than they brought in from the day he began, so his scheme needed new money to grow without end, and it stopped the moment new money slowed. A pyramid is the same sum with recruiting built in: with six each it passes all of Nepal in ten levels, and at every level about five members in six are the newest, who lose when it stops. Selling through people is legal in Nepal under the Direct Sale of Goods Act, but only with a license, only for real goods sold to real customers, and never as a pyramid, which that Act punishes with a heavy fine, years in prison, or both. Before you pay anything, ask who pays you, and from what.

Try this

  • Next time an offer reaches your family’s phone with a return in it, write down who it says will pay that return, and compare it with a bank’s fixed-deposit rate in the box below. If you can’t name the payer, you have found the answer.
  • On paper, work out a pyramid where each member brings in 5. How many new members does level 7 need, and how many members are there in all by then? (78,125 new, and 97,656 in all.)
  • If someone invites you to join a selling scheme, ask how much of what its members earned last year came from selling to customers outside the scheme. Ask whether it holds a license from the Department of Commerce, Supplies and Consumer Protection, and check before you pay.

Found a mistake? Please tell us.