← Maths & Science

The People Coin

The second half of a thought experiment. The Gross People Product scores every person from −1 to +1 according to whether their basic, stable and comfortable needs are met, and adds everyone up. Now suppose the dollar were replaced by a crypto-currency tied to that number, so that money itself tracked how many people are doing well. Would the economy still work? Would the coin even survive? This page takes the idea seriously enough to backtest it over a hundred years.

The short answer. Tying the quantity of money to the GPP would have worked surprisingly well for a century and then fail, because the GPP has a ceiling and recessions would destroy money exactly when it is needed. Tying the coin's price to the GPP is a Ponzi scheme with a better story. Tying only the distribution of new money to the GPP is sound, could be built today, and could be worth about as much as all the world's foreign aid — provided someone can measure the GPP honestly, which turns out to be the hardest part.

What money has to do

Money does three jobs: it is a unit of account (prices are written in it), a medium of exchange (people accept it in payment) and a store of value (it still buys something next year). A currency that fails at any one of them gets abandoned, usually fast.

The tool for thinking about what a rule for creating money will do is Irving Fisher's equation of exchange:

M · V = P · Y

Money in circulation (M) times how often each unit changes hands in a year (V) equals the price level (P) times real output (Y). Take growth rates and hold velocity steady, and inflation is simply the gap between how fast money grows and how fast the economy grows:

inflation ≈ money growth − output growth

Every rule for issuing money is therefore two decisions at once: a decision about the price level, and a decision about who receives the newly created money — the profit of issuing it, which economists call seigniorage. “Tie a coin to the GPP” can mean three quite different things depending on which of those it governs.

Quantity peg

The number of coins in existence tracks the GPP: coins are minted when people climb a rung and destroyed when they fall.

Precedent: The gold standard (supply set by mining), Bitcoin (supply set by a schedule), Ampleforth (supply rebased daily).

Verdict: Works for a century by luck, then turns structurally deflationary and destroys money in every recession.

Price peg

The coin's value tracks the GPP: one coin buys more as humanity climbs, so holders profit from progress.

Precedent: Algorithmic stablecoins such as TerraUSD, whose promised yield had no cash flow behind it.

Verdict: Unbacked: nothing pays the promised appreciation, so the peg holds only while new buyers arrive.

Distribution peg

The coin keeps a stable value; the GPP decides only who receives newly created money, weighted toward the lowest rungs.

Precedent: Chile's Unidad de Fomento (an indexed unit of account), dollar-backed stablecoins, Cash on Delivery aid.

Verdict: Viable, and buildable today — but small, and only as honest as the measurement behind it.

Design 1 — the quantity peg

The literal version: the number of coins in existence follows the GPP. It hits a wall at once. As specified, the world's GPP has been negative for the whole century — about −2.1 billion points in the mid-1980s, about −160 million today — and a money supply cannot be negative. The clean fix is to count rungs climbed instead of points. On the four-rung ladder of the companion article, every person stands on rung 0, 1, 2 or 3, and the total number of rungs humanity has climbed carries exactly the same information as the GPP (it is ³⁄₂ × (GPP + population)) but is never negative. The rule becomes: one coin is minted for every rung someone climbs, and one is burned for every rung someone falls.

A hundred-year backtest

Suppose the People Coin had existed since 1925. Its money supply would have followed humanity's total rungs; the economy would have followed world real GDP. The chart sets both to 100 in 1925.

People Coins in circulation (total rungs climbed) World real output
0600120018002400300019251935194519551965197519851995200520152025
Hover a year to compare the lines.

Money supply proportional to the total number of rungs climbed by the world's population; output from rounded decade averages of world real GDP growth. Both indexed to 1925 = 100.

The surprise is how closely the two lines travel. Over the century the coin supply would have grown about 17.5-fold and output about 25.6-fold. With velocity steady, prices in People Coins would have fallen by about 31% over a hundred years — roughly 0.4% a year. Over the same period, US consumer prices rose about 18-fold.

It is not luck. Both lines are population times something per person. The coin supply is population (×4.2) times the average rung (×4.2); output is population times output per head (about ×6.1). As long as development and income per head rise together — and over a century, across the world, they broadly do — a rung-based money supply grows at about the speed of the economy. That is a better record than gold, whose supply depended on where prospectors happened to dig, or Bitcoin, whose supply follows a schedule written in 2009 that knows nothing about the economy.

People Coin (implied) US dollar (consumer prices)
-6-303691935194519551965197519851995200520152025
Hover a year to compare the lines.

Average inflation over the ten years ending in each year shown. People Coin: money growth minus output growth, velocity held constant. US dollar: BLS consumer price index (2025 estimated).

Decade by decade, the coin would have been remarkably stable — never more than about two per cent a year from zero, against the dollar's swings from the deflation of the 1930s to the seven per cent a year of 1975–85. The two most deflationary decades are telling, though: the 1940s, when war knocked people down the ladder while factories ran flat out, and the decade just ended, when output kept growing but rung-climbing slowed. That second one is a preview.

Why it breaks: the ceiling

The average rung cannot exceed 3. Today it is 1.47. The UN projects world population to peak at about 10.3 billion in the 2080s and stand near 10.2 billion in 2100. So even in the best possible future — every person alive in 2100 on the top rung — the total number of rungs can grow from 12.0 billion to at most 30.6 billion: about 2.5-fold, or 1.3% a year. If the world economy grows at 2.5% a year, prices in People Coins must fall by at least 1.2% a year, every year, to 2100 — and faster once population starts to shrink and the money supply shrinks with it.

Permanent deflation sounds pleasant and is not. Debts are fixed in coins, so their real burden grows every year — Irving Fisher's debt-deflation mechanism, which he worked out from the wreckage of 1929–33. Simply holding coins earns a real return, so spending and investment are postponed. And interest rates cannot go much below zero, so in a slump the central lever of modern economic policy is jammed. The world has run this experiment: under the gold standard, prices in Britain and the United States drifted down for two decades after 1873, with chronic farm-debt crises and a political revolt against “hard money” that culminated in William Jennings Bryan's “Cross of Gold” speech in 1896. A People Coin would make that era permanent, and make it worse the more successful humanity became.

Why it breaks: recessions burn money

In a downturn people fall down rungs, so the rule burns coins — the money supply shrinks exactly when the economy most needs it to hold steady. In 2020 COVID pushed about 70 million people back into extreme poverty and many more from stable to merely getting by; the People Coin would have answered by destroying money. Milton Friedman and Anna Schwartz's account of the Great Depression is the story of a central bank that let the US money stock fall by about a third between 1929 and 1933. A quantity peg would write that mistake into the code. Forbid burning to fix it, and the coin no longer tracks the GPP at all.

Burning is also mechanically ugly. Coins sit in people's wallets; to destroy some, the protocol must either shrink every balance at once (a “rebase”, as the Ampleforth token does daily) or make money decay over time, like Silvio Gesell's stamped scrip used in the Austrian town of Wörgl in 1932–33. Either way, your savings change because of events on another continent — a hard sell for a store of value.

Why it breaks: the incentive points the wrong way

Hold output fixed and look at who gains. When people climb, new coins appear and every existing coin is worth a little less; when people fall, coins are burned and every remaining coin is worth a little more. The largest holders of the world's money — its richest people and institutions — would hold an asset that rises when humanity has a bad year. A coin meant to align money with human welfare would, at the margin, align its biggest owners against it.

Design 2 — the price peg

The opposite idea fixes that incentive: make the coin's value rise as the GPP rises, so that holders profit from progress. The trouble is the question every financial product must answer: who pays? A currency's value rises only if demand for it grows faster than its supply, or if an issuer buys coins back with assets that earn a return. The GPP rising generates no cash for anyone. An issuer promising appreciation must either hold real assets — and then the coin is just an investment fund with a mission statement — or pay early holders with later holders' money.

This has been tried, with a different story. TerraUSD, an “algorithmic” stablecoin, offered holders around 20% a year through its Anchor protocol with no cash flow behind the yield. When withdrawals began in May 2022 the peg broke, and tens of billions of dollars of value vanished within days. A GPP-appreciating coin is the same machine with a nobler label. Not viable.

Design 3 — the distribution peg

The version that survives separates the two decisions. Let the quantity of money do what sound money should — grow with the demand for it, so prices stay stable — and let the GPP decide only who receives the newly created money. In practice: a stablecoin worth one dollar, or one basket of basic goods, fully backed by safe assets such as Treasury bills; the interest those reserves earn (or, for an unbacked currency, the new issuance that a growing economy absorbs without inflation) is paid out according to the GPP — to the people on the lowest rungs.

The economics are not hypothetical. Tether, the largest dollar stablecoin, reported a net profit of about $13 billion for 2024, earned mostly on the Treasury bills behind its coins. That is seigniorage, and today it belongs to the issuer's owners. A People Coin is the same machine with the profits pointed at the bottom of the ladder.

How big could it be? Suppose the coin reached $10 trillion in circulation — a small fraction of the world's money — and yielded about 3% of that a year to distribute, roughly the growth rate of the world economy or the interest on safe reserves. That is about $300 billion a year, more than all official development assistance (about $212 billion in 2024). Paid to the 0.8 billion destitute, it comes to about $366 each a year — roughly $1.00 a day, a third of the way to the $3 line. Spread over the 4.9 billion people on the bottom two rungs, it is about $61 each. Meaningful, not transformative.

Three ways to decide who gets it, each with a known failure:

  • One share per person. The simplest and hardest to game by governments: a universal dividend. But it needs a global registry of unique humans. Projects such as World (formerly Worldcoin), which scans irises, and GoodDollar have tried; World's scanning has been suspended or challenged by regulators in several countries on privacy grounds.
  • Weighted by rung. More to the destitute, less to those getting by, nothing above. Fairer, but it requires knowing each person's rung — surveillance, or self-reporting that everyone has a reason to shade downward.
  • Bounties for climbs. Pay governments or charities a fixed sum for every person verifiably lifted a rung — the “Cash on Delivery” aid model Nancy Birdsall and William Savedoff proposed in 2010. It pays for results rather than plans, and it runs straight into the next problem.

The oracle — what breaks every version

A crypto-currency is only as good as the data fed into it, and the GPP is built from household surveys run by governments, published with a lag of two or three years and revised for a decade after. Once money depends on the number, the number becomes a target, and Goodhart's Law takes over: a measure used as a target stops measuring. There is a precise precedent. From 2007 to 2015 Argentina's statistics office understated inflation, cutting what the state owed on inflation-linked bonds, until the IMF formally censured the country in 2013. A GPP coin that pays for progress would pay governments to report it.

Even honest statisticians move the number. When the World Bank updated its extreme-poverty line in June 2025, about 125 million people were added to the 2022 count. Under the quantity peg, that methodological change alone would have burned about 1.0% of the world's money supply. Any serious design needs measurement that no single government controls — satellite and phone data, randomised audits, heavy smoothing, caps on how fast any country's figures can move — and whoever runs that would sit on the most powerful committee in the world.

If it really replaced the dollar

  • Prices would drift down gently for a while, then fall structurally as the world approached the ceiling. Savers and creditors win; borrowers, young households and firms that invest lose.
  • No country could run its own monetary policy, as with the euro or with dollarisation (Ecuador in 2000, El Salvador in 2001). Recessions would be fought with budgets alone.
  • Demand for the coin would have to come from somewhere. Dollars, euros and yuan are wanted partly because taxes must be paid in them. A private coin has no tax base pulling it into use unless states agree to accept it — and accepting it means surrendering the printing press.
  • Redistribution would be built into money creation: a small, permanent, automatic transfer from the forgone interest of coin holders to the bottom of the world's income ladder.
  • Power would move to whoever controls the measurement, which is why the oracle, not the cryptography, is the real design problem.

Verdict

As a replacement for the dollar, not viable. The quantity peg would have tracked the economy for a hundred years and then delivered permanent deflation and money that vanishes in every recession; the price peg is unbacked by construction. As a unit of account, viable: contracts can be written in an indexed unit, as Chile has done with its Unidad de Fomento since 1967, and a unit defined as “the cost of meeting one person's basic needs for a day” is close to what Benjamin Graham proposed in 1937. As a stablecoin whose earnings go to the bottom rungs, viable and buildable today — limited by how many people choose to hold it, and only as honest as its measurement.

The lesson generalises. Money should be boring: its quantity should follow the demand for money and nothing else. The GPP should be ambitious: it is a fine rule for deciding where new money goes. Tie the GPP to the distribution of money, never to its quantity.

Appendix — assumptions and sources

A backtest of a currency that never existed. The money supply follows the reconstruction in Gross People Product, whose rung shares are rounded estimates. The quantity equation is applied with velocity held constant, which no real currency achieves — velocity moves with interest rates, payment technology and confidence. The backtest shows what the issuance rule implies, not what a real coin would have traded at. The projection to 2100 assumes output growth of 2.5% a year and the UN's medium population scenario; the $10 trillion circulation and 3% yield are round illustrative figures.

Sources

  • Irving Fisher. The Purchasing Power of Money (1911) for the equation of exchange MV = PT, and “The Debt-Deflation Theory of Great Depressions”, Econometrica, 1933.
  • Milton Friedman & Anna Schwartz. A Monetary History of the United States, 1867–1960 (1963) — the money stock fell by about a third from 1929 to 1933.
  • US Bureau of Labor Statistics. consumer price index, annual averages (the 2025 value is an estimate).
  • Friedrich Hayek. Denationalisation of Money (1976) — the case for private, competing currencies.
  • Benjamin Graham. Storage and Stability (1937) — a currency backed by a basket of basic commodities.
  • Silvio Gesell. The Natural Economic Order (1916), and the Wörgl stamp-scrip experiment in Austria, 1932–33 — money that loses value by design.
  • Banco Central de Chile. the Unidad de Fomento, an inflation-indexed unit of account in use since 1967.
  • Nancy Birdsall & William Savedoff. Cash on Delivery: A New Approach to Foreign Aid (Center for Global Development, 2010).
  • OECD Development Assistance Committee. official development assistance of about $212 billion in 2024 (preliminary).
  • Tether. reported net profit of about $13 billion for 2024, earned mostly on the Treasury bills backing its stablecoin.
  • Terra / Luna collapse, May 2022. an algorithmic stablecoin paying an unbacked yield lost its peg; tens of billions of dollars of value disappeared within days.
  • Ampleforth (2019). a token whose supply is rebased daily toward a price target.
  • World (formerly Worldcoin) and GoodDollar. crypto projects distributing tokens per verified human — the proof-of-personhood problem in practice.
  • Argentina, INDEC. the official inflation figures understated from 2007 to 2015; the IMF formally censured the country in 2013.