Impactism
日本語 Join Impactism
Manifesto The Impact Standard Foundations Objections Read For organizations 日本語
Join Impactism
Antique gold and silver coins
Photo by Elena Mozhvilo on Unsplash

Essay

Money Is Memory — and What It Forgets

June 15, 20268 min read

We want to start by being fair to money, because the case against it is usually made dishonestly.

Money works. Strip away the resentment and the romance, and at bottom it is a kind of shared memory. A portable record that you did something for someone, which a stranger on the other side of the world will honor. You bake bread; you carry away a coin; weeks later, in another town, a person who has never met you hands you a coat for it. The coin remembers the bread. It is a record that you contributed, redeemable by people who were not there.

This is not a metaphor we invented. The economist Narayana Kocherlakota proved a version of it formally in 1998, in a paper whose title is exactly this: “Money Is Memory.” His result is that anything money does in an economy, a complete and trusted record of past transactions could do instead. Money functions as a kind of societal ledger — a way of keeping track of who has given what, so that giving can be repaid by someone other than the person you gave to. The full equivalence has qualifications, and we will not overclaim it. But the core idea is real economics, not poetry. The anthropologist David Graeber made the historical companion case in Debt: ledgers of obligation — who owes whom — come first, and coins arrive later to make those ledgers portable. Credit precedes coinage. The memory came before the metal.

If that is what money is, then the largest economic fact of the last generation looks less like a miracle and more like good bookkeeping at scale. Across recent decades, the world saw the largest decline in extreme poverty in recorded history. We say this as accompaniment, not as proven cause — markets running on that ledger were present while it happened, and honesty forbids us from claiming they alone caused it; the measurement lines have also been redrawn over the years, which is why we don’t lean on the raw percentages. But the direction is not in dispute, and anyone who wants to burn the ledger should sit with that number first. Money let strangers cooperate at a scale no village ever managed. We are not here to burn the ledger.

We are here to point at what the ledger forgets.

What the memory leaves out

Money records what was sold. It has almost no memory of what was given.

Think about the most important work happening within a hundred meters of where you are sitting. Someone is raising a child. Someone is caring for a parent who no longer remembers their name. Someone is cooking a meal that will be eaten and never invoiced. None of it shows up in the national accounts — and not by accident or oversight. It is excluded by construction. The international rulebook for measuring an economy, the System of National Accounts, draws a line it calls the production boundary, and unpaid household and care work falls on the far side of it. The economy, officially, does not see it.

The political economist Marilyn Waring made this the center of her 1988 book If Women Counted. Her point was not sentimental. It was structural: a measure that defines an entire category of essential human work as economically nothing will, over time, treat the people who do that work as nothing too. The ledger is not lying. It is doing exactly what it was built to do — recording priced transactions. The failure is not moral. It is a recording failure of the unit.

You can watch the same blind spot in a place that looks nothing like a kitchen.

The maintainer nobody pays

Open-source software is, by most measures, one of the better recognition systems humans have built. The person who wrote the code that runs your bank, your phone, half the internet, can be looked up by name. Their contribution is visible in a way a caregiver’s almost never is. Recognition, here, tracks real contribution better than most systems manage.

And yet. The currency of open source is the GitHub star — and a star measures popularity, not load-bearing importance. The most-starred projects are not always the ones quietly holding up the most. Behind a great deal of critical infrastructure sits a single unpaid maintainer, doing the work for free, watched by thousands, paid by no one. Nadia Eghbal’s Working in Public documented the shape of this: vast public dependence resting on private, uncompensated labor. Tidelift’s surveys of maintainers found a majority of them unpaid, and a majority who have quit or considered quitting — burnout in plain sight, on the most measurable contribution there is. (Treat those percentages as indicative; the sample is self-selected.) Even when the work is visible, the unit that measures it counts the wrong thing. Stars count attention. They do not count the weight someone is carrying.

Diamonds and water

There is an old puzzle that explains why money keeps making this mistake, and it is as old as economics itself.

Nothing is more useful than water; you die in days without it. A diamond is, by comparison, useless — you can live a full life having never touched one. Yet water is nearly free and a diamond costs a fortune. Adam Smith famously posed this in 1776 (the puzzle is older than him), and it sat unresolved for nearly a century until the marginalists worked out the answer in the 1870s. Price does not track total value. Price tracks marginal value under scarcity — what the next unit is worth when the thing is plentiful or rare. Water is abundant, so its price falls toward nothing, no matter that it is the most valuable substance on earth.

This is the whole problem in miniature. The things that matter most are often the things we have enough of to make cheap — air, water, a parent’s attention, the unglamorous maintenance that keeps everything running. Price is a measure of scarcity, not of worth. It was never designed to tell you what is valuable. It was designed to tell you what is scarce. We have spent a few centuries confusing the two.

To be fair, once more

We don’t want to leave money under-credited, because that would be its own dishonesty.

Money does not only fail to record value; it can also create it, durably. When Swedish lottery winners were followed for years afterward, their life satisfaction rose and stayed risen — not a sugar high, but a lasting gain. Windfalls help. Anyone who tells you money doesn’t matter is selling something, usually to people who already have enough of it. A measure that pays out real, durable well-being is not a measure to be sneered at. It is a measure to be kept — and asked to do less.

Because that is the precise claim. Not instead of money. Alongside it.

The case for a second measure

Put the four things side by side. The care work the accounts exclude by design. The maintainer the stars can’t weigh. The water the price can’t value. And, underneath all of it, money’s own honest origin: a memory of who contributed, kept so that contribution could be repaid by strangers.

The lesson is not that the ledger is corrupt. The lesson is that it is partial. It remembers one kind of contribution — the kind that passed through a price — with extraordinary fidelity, and it forgets the rest almost entirely. A society that runs entirely on that one memory will, slowly and without anyone deciding it, come to treat the unpriced halves of life as if they were worth nothing. Not because anyone is cruel. Because the instrument cannot see them.

So the conclusion we draw is modest, and we mean it as modest. Keep money. It is one of the most useful tools our species ever built, and the people most eager to abolish it have usually never gone without it. But build a second measure for the half of human contribution that money was never able to record — the given, not the sold. Make it work like honor and never like a price: conferred by the people who saw the work, carried, never cashed. Make it a plural ecology, never one global score — a nurse’s ward and an open-source project will never share one number, and they shouldn’t. Let meaning lead, with the measure following behind. A curve, not a switch.

That second measure is what we are building. We lay out the argument for it, step by step and with the contested evidence flagged, on Foundations; the mechanism — how a unit of recognition can work like honor instead of money — on The Impact Standard.

Money is memory. We are not here to erase it. We are here to write down what it forgot.

Sources

← All essays