In 1920, there was an Italian man in Boston named Charles Ponzi. He promised investors 50% returns in 90 days.

Where did the money come from?

He said it was international postal arbitrage. Buy Italian postal reply coupons with American money, exchange them back, and pocket the difference.

The explanation sounded very professional. But no one actually verified it.

Because the returns were real.

You invest a hundred dollars, three months later you get one-fifty. You tell your friend, your friend invests too. He also gets paid. So more people invest.

Ponzi ended up serving fourteen years in prison. His name became a word: Ponzi scheme.

But I've always had a question.

What was so special about Ponzi's model? Humans didn't start conning people with him.

Later I learned he really wasn't the first.

In 1869, there was a woman in Germany named Adele Spitzeder. She opened a private bank promising high returns. Most depositors were poor people. Where did the money come from? From new depositors.

Going further back, in 1880, in Boston—yes, the same city—there was a woman named Sarah Howe. She opened a "Ladies' Deposit" that only accepted unmarried women. 8% monthly interest.

They all preceded Ponzi by decades.

But no one remembers their names.

Ponzi became a noun not because he invented something. It's because of his scale. Because of the media. Because of timing—post-WWI America, people wanted to get rich, wanted to get rich fast.

He's remembered not because he was the worst, but because he was the loudest.

When I was researching stablecoins, I had to seriously think about this question: what exactly is a Ponzi scheme?

The definition seems simple. Three characteristics:

One, returns come from new investment, not real profits.

Two, promises of high returns with low risk.

Three, no actual business.

Sounds clear. But if you think carefully, you'll find many things fit partially.

I once drew a diagram—a mathematical model of chain letters.

Chain letters are the simplest Ponzi structure. You receive a letter with a list of names. You send money to the first person on the list. Then you add your name to the bottom and send the letter to ten friends.

If everyone follows through, you'll eventually receive a large sum of money.

If.

I wrote it as a formula. Suppose each person sends to M people, the list has a maximum of N names, and there are L total layers.

How much does the first person earn? He stays on the list longest, receives the most money.

How much do people in the middle earn? They only stay on the list for N layers, receiving M to the power of N payments, minus the one payment they made.

What about the last few layers? They only pay out, never receive. Because there is no next layer.

This is inevitable.

What's interesting is the return rate.

You'd think the more people, the more the early entrants earn?

Not exactly.

It depends on the relationship between list length N and distribution number M.

If the list is short and people are many, the first person's advantage isn't that big.

If the list is long and people are few, the first person's advantage is enormous.

In other words, the inequality of a Ponzi scheme can be tuned.

Smart con artists adjust this parameter. Make it look "fairer," make it last longer.

Then I thought of a question.

In a chain letter, everyone knows where the money comes from—from the next person down. No one is "deceived," everyone knows the rules.

So why is it illegal?

The answer is: information asymmetry.

The people in the last few layers don't know they're in the last few layers. They think there's still a next layer. They think there's still time.

That's the con.

Not lying about where the money comes from. Lying about how many layers are left. Lying about your position.

Ponzi didn't lie to investors about giving returns—he actually did.

What he lied about: he said returns came from postal arbitrage. They actually came from the next batch of fools.

This lie made people believe the system could last forever.

What if you don't tell that lie?

What if a system explicitly states: returns come from new investors. Would you play?

Some would.

MLM. Direct sales. Amway. Most participants know where the money comes from. They bet they're not the last layer.

This is called "rational Ponzi."

Economists have studied this. If interest rates are low enough and the population is growing, a Ponzi structure can theoretically run forever.

Sounds crazy. But think about pensions. Think about social security.

The money you pay isn't saved up waiting for you to retire. It's being paid out right now to people currently retiring.

When you retire and withdraw money, you're depending on the next generation to pay in.

If that's not a Ponzi scheme, what is?

What's the difference?

The difference is: whether people are being born.

Pensions' "next layer" is the next generation. As long as people have children, the system keeps running.

But chain letters' "next layer" is friends of friends of friends. Your social network is finite. After a few layers, it's exhausted.

A Ponzi scheme must collapse because growth must stop.

But pensions don't have to collapse because humans are still reproducing.

You could say human reproduction is the biggest Ponzi scheme. We use descendants who don't yet exist to support current beliefs.

When I was writing my paper, I divided economic activities into three categories.

First category: no output from start to finish. Money just transfers between people. Pure Ponzi.

Second category: no output at first, then there is. Like a startup that burns money early, then becomes profitable. This isn't fraud, it's called "success."

Third category: output at first, then none. Business keeps losing money but keeps raising funds. This is called "becoming a fraud."

Most companies that fail spend some time in the third category before dying.

Minsky called these "Ponzi enterprises." If a company can't even pay its interest and can only borrow new debt to pay old debt, it's entered the Ponzi stage.

It didn't start out wanting to defraud. It just couldn't survive anymore, so it had to.

What about stablecoins?

I researched DAI. It's a cryptocurrency that claims to always be worth one dollar.

How does it do this? Collateral.

You deposit Ethereum, it gives you DAI. If Ethereum drops, the system automatically sells your collateral to ensure DAI doesn't collapse.

Doesn't sound like a Ponzi.

But look closely—where do its returns come from?

From lending interest, from liquidation penalties, from appreciation of governance tokens.

Why do governance tokens appreciate? Because people believe the system will continue. Because new capital is coming in.

When new capital stops entering, governance tokens fall, and confidence in the entire system shakes.

DAI isn't a pure Ponzi. But some parts of it—especially the part where speculators make money—depend on growth.

In 2022, Terra collapsed.

It was another stablecoin. Its design was: use another coin it issued to "support" itself.

This is called self-reference.

Like a person trying to lift themselves by pulling on their own shoelaces.

A few days before the collapse, Terra's founder was mocking skeptics on Twitter: "Poor people."

Then forty billion dollars evaporated.

Some said it was a Ponzi. Terra's people said it wasn't.

What's the difference?

I thought about it for a long time. My conclusion was:

The definition of a Ponzi scheme isn't "using new money to pay old money." Many normal economic activities do this.

The definition of a Ponzi scheme is: claiming there's an income source that doesn't exist.

Ponzi said he was doing postal arbitrage. He wasn't.

Terra said its algorithm could maintain stability. It couldn't.

DAI says it has real collateral. Most of the time, it does.

The difference isn't structure, it's the lie.

The structure can be identical—one is legal, one is illegal.

It depends on whether you lied to people.

This made me think of a deeper question.

What is money?

Chen Qiming said meaning is prepaid. You do something, give yourself a reason. The reason needs validation. But validation is always in the future.

Money is also prepaid. You hold a hundred dollars—why is it worth a hundred dollars? Because you believe tomorrow someone will trade something for it.

No one actually verifies. No one can verify. You can only believe.

What sustains this belief?

Forgetting.

You forget that you never actually verified. You forget that this paper has no value itself. You forget that you're participating in a centuries-old gamble.

Currency is the most successful Ponzi scheme humanity ever invented.

It's so successful that everyone forgets it's a Ponzi scheme.

I'm not saying you should withdraw all your money.

I'm saying, when you call something a "Ponzi scheme," think clearly about what you mean.

Are you saying it uses new money to pay old money? Many things do.

Are you saying it will eventually collapse? Everything eventually collapses. The sun will explode too.

Are you saying it deceived people? Then the question is, deceived them about what?

Ponzi's crime wasn't his model. It was his lie.

I finished writing this and made a cup of tea.

It was raining outside the window. I looked at the numbers in my bank account.

Are those numbers real?

They're a string of electronic signals. Their value depends on: tomorrow someone willing to recognize them. The day after, someone willing to recognize them. A year from now, ten years from now, someone willing to recognize them.

I have no way to verify this. I can only believe.

Then I forget that I'm only believing.

Everyone does.

Maybe this is how we keep living—participating in a game we can't think too hard about.

The tea went cold.