Still Waters: The Case for a Flow Economy

TL;DR: Opportunity comes from money moving, not from money piling up. Our economies demand constant growth because vast pools of money have stopped moving, and raising the water level is the only way anyone downstream gets a drink. A century of half-buried experiments shows money can be made to move. This post makes the case for a flow economy: a tax schedule that makes money cost more to hold than to use, collected through the same machinery that already collects GST and VAT, and aimed at one outcome, the constant creation of opportunity. No programmable currency. No printing. And no growth requirement.

The town that made money rot

In the winter of 1932 the town of Wörgl, four thousand people in the Austrian Tyrol, was broke: a third of its workers had no work, and the town couldn’t pay for roads, let alone wages. Its mayor, Michael Unterguggenberger, a former railway mechanic with a shelf of unfashionable economics books, tried something strange. Wörgl issued its own money.

The notes paid wages on public works, and they carried a catch. Printed on the back was a grid of boxes, one per month, and to keep a note valid its holder had to buy a stamp worth 1% of the note’s face value and glue it into the current box. Holding money for a year cost you 12%, while spending it made the stamp someone else’s problem. Every shopkeeper checked for the stamp, because accepting an unstamped note meant buying the stamp yourself, which is why everyone spent before stamp day.

Wörgl built a bridge, a reservoir, a ski jump and new roads. The standard account says the stamped notes changed hands around fourteen times faster than the Austrian schilling, and that local unemployment fell by a quarter while it climbed everywhere else in the country.

Treat that account with care. The famous velocity figure traces to Irving Fisher, the celebrated American economist who was also the scheme’s loudest promoter. Much of the revival was plain public-works spending that any deficit would have bought, and there was no twin town next door running the same works without the stamps. One town, fourteen months, contested numbers, no control group.

The ending, at least, is beyond dispute: the Austrian central bank saw a rival mint, went to court, and shut the experiment down in September 1933, while some two hundred towns were preparing to copy it.

Wörgl proves nothing, but it hints at something: that money’s willingness to move may matter as much as its quantity, and that a small, boring fee can transform that willingness.

Fast water going nowhere

Picture the economy as a body of water. Money is the water, payments are the current, and the life the water supports (crops, mills, towns) is the real economy: goods made, services rendered, wages paid.

Hydrologists learned long ago not to trust speed measured at a single point. A river’s centre channel runs fast while the banks barely move, and behind every boulder sits an eddy: water spinning quickly in a circle, all motion and no transport. That is why hydrologists measure through-flow instead, the volume that actually passes a cross-section and travels downstream.

Finance is full of eddies. A share that changes hands a thousand times a day between trading algorithms is water spinning behind a boulder, enormous local speed with nothing carried downstream, while a dollar spent at a bakery moves once and pays a baker’s wage. Through-flow, in a river or an economy, is what feeds the valley.

Still water rots

The danger in stagnant water comes from what flow carries: oxygen. Cut the flow and the oxygen goes; algae bloom, the depths turn septic, and the pond starts breeding mosquitoes that bite the whole neighbourhood.

Money behaves the same way, because idle capital doesn’t sit politely waiting to be useful. It seeks return without motion, buying what others must rent, cornering what others must buy, and fencing what used to be open. A stagnant pool does worse than fail to water the fields: it breeds things that bite.

And an asset bubble is an algal bloom, sudden lurid growth in the stillest part of the water, a sign of sickness that photographs like life.

Lakes have layers

Deep lakes stratify. The sun warms the surface until the warm layer floats on the cold depths, the two stop mixing at a sharp boundary, and the surface can be busy with wind and waves while the bottom sits still for months.

Modern economies stratify the same way, with a surface layer where money churns among assets (shares, property, stakes trading hands at speed) and a deep layer of wages, invoices and shop tills. Between them sits a boundary that money rarely crosses.

We ran the experiment that proves it, expensively. After 2008, central banks created new money on a scale never before attempted and injected it at the top, by buying bonds from financial institutions. The new water landed in the surface layer and stayed there, and asset prices roughly doubled over the decade while consumer prices and wages barely stirred. There’s a textbook equation for this, the only one this essay needs: MV = PQ, the money stock times how often it changes hands equals prices times real output. The decade after 2008 tripled M and watched V collapse to cancel it, because printing raises M and prays, and the praying went unanswered. Water poured onto a stratified lake spreads across the top, and new money enriches whoever stands nearest the tap. The tap was installed in the penthouse.

Lakes do mix eventually, when storms and the turn of seasons break the layers and turn the water over, and economies historically got their turnover the same ugly ways: wars, crashes, great inflations. The engineering question of this essay is whether we can build gentle, continuous turnover instead of waiting for the storm.

Money flows uphill

Here the metaphor inverts, and the inversion is the point. Water flows downhill and spreads out. Money flows uphill and concentrates. Returns compound, money attracts money, and left alone every dollar in the system drains toward the existing pools, the way every drop of rain eventually finds the sea. A pure gravity system ends with all the water in the ocean and every river dry.

Rivers run anyway for one reason: the sun lifts water out of the sea, against gravity, and rains it on the mountains. No sun, no rivers, anywhere, ever.

The economy has a sun of a kind in innovation, which acts on money the way heat acts on water. It thins it, cutting the friction of moving it (consider what payment technology alone has done in a decade), and it evaporates the stagnant pools: a real breakthrough pulls value straight out of incumbent assets and rains it down as new headwaters of firms, wages and work. Kodak’s fortress became a puddle and the value fell elsewhere. Economists call it creative destruction; a hydrologist would recognise evaporation and rainfall.

But innovation is weather, not climate: it heats unevenly and on no schedule, decades of overcast and then one scorching summer, and no river system survives on weather alone. The pools defend themselves besides. Notice the word finance uses for a protected market position: a moat, still water dug on purpose to keep the heat out. Regulatory capture, patent thickets, buying the raincloud before it bursts: the dead zones of an economy do not wait passively to be evaporated.

Growth raises the water level

Now the piece this essay exists for. What do you do with a lake whose pools are dead, whose layers won’t mix, and whose weather won’t cooperate? You raise the water level. If the level rises everywhere, even the stagnant margins get wet, and everyone downstream drinks a little. That is what growth does for us, and that is why we cannot do without it: every boat lifts and the dead zones stay hidden underneath. And the moment inflow stops, the stagnation that was always there starts to bind, which is why a zero-growth year reads as an emergency rather than a rest.

We have the causality backwards. Growth is the compensation we pay ourselves for circulation that has failed. A river doesn’t need more water every year to sustain the life along its banks; it needs the same water to keep moving.

An economy built on that principle, where opportunity comes from circulation rather than inflow, deserves its own name. Call it flow capitalism. Nothing about it abandons markets, property or profit; it asks for one change, that money cost more to hold than to use.

We already charge for holding money, badly

Wörgl’s stamps came from Silvio Gesell, a German-Argentine merchant who published the design in 1916. Gesell noticed that everything a farmer brings to market rots, rusts or goes stale except the money the buyer brings, and that this asymmetry lets money charge a toll for turning up. His remedy was money that rots on schedule, and his second remedy, less remembered, was a land tax, for a reason that is about to matter.

Keynes read Gesell seriously, called him “an unduly neglected prophet”, and judged that the future would learn more from him than from Marx. Then he drove in the knife. Rot the cash, Keynes said, and hoarding simply moves house: into bank balances, gold, jewels, land, foreign money. The fee chases the hoard from one vessel into the next, and the stillness survives the eviction.

He was right, and it has been proven since. In 2012 a cryptocurrency called Freicoin implemented Gesell faithfully, with an automatic 5% annual decay, and holders did what Keynes predicted: they left for money that didn’t rot, and Freicoin died of emigration. A holding fee that applies to only one vessel is a tax with a thousand exits. Hold that thought, because the design below is shaped by it, and so was Gesell’s: the land tax was his exit-blocker.

Meanwhile, we run a holding fee of our own and pretend we don’t. Central banks target 2% inflation partly so that cash quietly loses 2% a year and hoarding it never quite pays. Inflation is the same fee wearing a disguise, and the disguise makes it crueller, because a hidden fee lands on whoever can’t dodge it. The pensioner’s fixed income and the worker’s wage between pay rises take the hit in full, while the asset-rich escape, because houses and shares ride the price level up. We chose the version of Gesell’s fee that spares the pools and taxes the current.

After 2008 the fee briefly came out of hiding. For most of a decade the central banks of Europe and Japan charged banks negative interest for parking reserves: a holding fee, applied with a firehose, flat across everything. It underwhelmed. Push a flat fee deep enough and banks lend less, not more, because lending itself stops paying, and Sweden left first, concluding the side effects outweighed the stimulus: a flat fee on everything is the wrong shape for the tool.

Three states of money

Shape it properly, then. Watch a single dollar for one quarter of the year and ask two questions: did it move, and if it moved, what did it buy?

Three answers are possible. The dollar sat: call it idle. The dollar moved and bought something that already existed (a share from another holder, a house already built, a bar of gold): call it churning, motion in an eddy. Or the dollar moved and bought goods, services, labour or new capacity (a wage, an invoice, newly issued shares, a construction loan): call it working, the dollar that passed the cross-section and travelled downstream.

I didn’t invent that split. Richard Werner published it in 1997, separating money that flows into transactions counted in GDP from money that flows into trades of existing assets, and showing that the second stream is what inflates bubbles (he called it the quantity theory of disaggregated credit, and built it on Japan’s bubble years). Werner aimed his split at bank lending, arguing regulators should steer credit toward the first stream. The proposal here ports his split to the other side of the ledger: not what banks may lend into, but what holding money costs.

The schedule: idle money pays the full holding fee, say 4% a year. Churning money pays half. Working money pays nothing. And the first stretch of every household’s balance pays nothing regardless, because an emergency fund is insurance, and insurance is a real service. The fee is aimed at the parked billions, not the rainy-day thousands.

Notice what three tiers do to Keynes’s objection. His migration argument killed Gesell’s design because only cash rotted, so value fled into everything else, but here the fee follows the money into the escape routes. Flee cash into asset churn and you meet the half rate. Buy land and sit on it and you are idle again in a different vessel, which is where Gesell’s land tax re-enters the full design. Each escape route still exists, and each now carries a toll.

Dumb money, smart tax

This proposal requires no crypto or programmable money: the currency stays dumb, no token expires, no wallet freezes, and no transaction is blocked. The classification and the fee live where classification and fees already live: in the tax system.

The precedent runs the checkout of nearly every country on earth. GST here, VAT in about 175 countries: every transaction is already classified at the till by its use (taxable, exempt, zero-rated), self-assessed by the parties, reconciled quarterly, audited by sampling, and disputed in front of a tribunal while the money stays put. Classifying what money is doing, transaction by transaction, is already the world’s dominant tax design, and nobody calls a quarterly VAT return a surveillance apparatus. The holding fee is assessed the same way: banks already report balances and interest, so the fee on idleness is one more line on the same statement, and nothing new gets switched on. The one major economy without this rail is the United States, which has no VAT, and honesty requires saying an American implementation would be the hardest one.

Enforcement borrows a second old trick. Taxes on trades fail by migration: Sweden taxed share trades in the 1980s, the trading promptly moved to London, and the tax died within seven years. Taxes on title survive: Britain has taxed share ownership through stamp duty for over three centuries, because dodging it means you don’t legally own the share. So the churning tier is levied where ownership is registered (the land titles office, the securities depository, the company register), not where the trade executes, and an asset stays anchored to its jurisdiction even when its trading doesn’t.

The deepest anchor belongs to the currency itself. A wealth tax follows the person, and people cross borders, which is why most European wealth taxes are dead. A currency’s stock can’t emigrate: sell your dollars in disgust and someone buys them, and the deposits remain inside the dollar system with a different name attached. A fee attached to the currency follows the currency. The honest leak is offshore booking, because dollars held in foreign accounts have escaped domestic rules before (it’s how the eurodollar market was born), and the idle tier will leak there too, though two decades of international account-reporting have narrowed the channel. The fee works best on large domestic balances, which fortunately is where the diagnosis says the disease is.

One more property worth stating plainly: this is a national instrument. Growth economics needs the world to cooperate: export markets, global demand, everyone’s water rising together. A flow economy concerns the internal hydrology of a single currency zone, so one country can fix its own mixing without waiting for a treaty. Wörgl didn’t ask permission from Vienna. That turned out to be the problem, and a nation doesn’t have a Vienna.

Where this breaks

Every proposal in this territory has been attacked for a century. The serious objections deserve plain answers.

“A dollar today beats a dollar next year, fee or no fee.” The oldest attack, pressed hardest by the Austrian school of economics. Their case: interest exists because people are impatient, since almost everyone prefers a sure dollar today to a promised dollar next year, and no tax can repeal impatience, so a fee that tries to abolish interest will only warp prices on the way to failing. Agreed, and permanently: this fee doesn’t try. It prices one behaviour, idleness, the way a congestion charge prices driving into the city without abolishing cars. Interest survives, but sitting still stops being free.

“Banks create money when they lend. The action is in new credit, not old balances.” The schoolbook picture has banks collecting deposits and lending them out, but modern banking runs the other way: when a bank approves a loan it creates the money on the spot, and the deposit appears as the loan is spent. Economists who start from that fact object that a fee on balances tinkers with money that already exists, while the economy is actually steered by the money being created, loan by loan. Also true, and out of scope: a fee on balances doesn’t direct what banks lend into, and doesn’t claim to. That steering is Werner’s own credit-side proposal, and the two instruments are complements. One governs how the existing stock behaves, the other where new lending goes.

“Velocity is an outcome, not a policy dial.” This one comes from the monetarists, the tradition of Milton Friedman, and in plain terms it says that how fast money moves is a reading, not a setting. Each of us decides how much cash to keep on hand and for how long, the economy’s speed is the sum of those private choices, and a government can no more decree it than it can warm a room by legislating what the thermometer says. Agreed, and the objection defeats a proposal nobody here is making, because nothing in this design legislates speed. The fee changes what holding money costs, and speed responds through the channel their own theory describes: make holding dearer and people hold less, for less time. We don’t say “investment is a private choice, so interest-rate policy is impossible”; the central bank sets a price and the choices move. Inflation is the existence proof that the channel works on money too, because every inflationary episode shows money speeding up as holding it grows costlier, with hyperinflation as the flood-stage limit. The fee is the same force at a hundredth the dose, steady instead of panicked, and the difference between flood and flow is exactly why the schedule has three tiers instead of one.

“Money is fungible. The categories will be gamed.” Fungible means interchangeable: every dollar is identical to every other, so the objection says labels like idle, churning and working describe nothing real, and drawing walls through something that mixes like water invites people to relabel their dollars rather than change what they do with them. The strongest objection, and partly right. A wash trade dressed as an invoice turns churning money into working money on paper, and related parties can sell each other services that don’t exist. Conceded: the boundary is imperfect and always will be. But it is the identical fraud surface that GST/VAT presents (fake invoices, carousel schemes), and the world’s tax offices police that surface tolerably well across 175 countries. Imperfect boundaries with familiar policing beat the pristine boundaries of instruments that were never built.

A wealth tax that wants to be avoided

The wealth-tax debate has returned everywhere at once: Piketty made inequality’s arithmetic famous, economists tour proposals for minimum taxes on billionaires, and every few months another country floats a levy on large fortunes. The instinct is sound, because the pools are real and they are growing. The designs keep failing, though, and the failures rhyme: the tax follows the person, the person moves, and a decade later the levy is repealed having collected less than promised. Norway raised its wealth tax and watched its richest citizens decamp to Switzerland; France and Sweden repealed theirs.

This essay’s proposal is a wealth tax, if you want to call it that, with two differences. It follows the currency and the asset register rather than the person, which closes the airport exit. And it taxes the only property of wealth that harms anyone else: its stillness. A fortune paying wages, funding new firms and buying services pays nothing under the schedule, whatever its size, while the same fortune parked pays every year until it moves.

Every tax breeds avoidance, and tax design is mostly the art of surviving it. This one inverts the problem: the only way to avoid the fee is to put the money to work, so avoidance is the policy. We should tax wealth, but the wealth that sits still.

What would change my mind

I hold this as an argument, not a faith, so the exits should be marked. If new research overturns the finding that flat negative rates suppressed lending, the case for shaping the fee weakens, and a flat fee would be simpler. If the half rate on churning turns out to choke the resale markets that let early investors exit, and new investment dries up upstream as a result, the churn tier needs rethinking, because an eddy behind a boulder is part of how a river works. If independent measurement ever pins down what Wörgl or its modern Bavarian descendant (the Chiemgauer, a stamped local currency running since 2003) actually did to velocity, the historical evidence firms up or dissolves. And if any jurisdiction ships a graduated, use-linked fee and money fails to move, the theory has met its control town at last.

The question to ask

A limnologist, a scientist of lakes, walking your shoreline would not ask how much water the lake holds, nor how fast it moves at the boat ramp. She’d ask where the dead zones are and what blocks the mixing.

Ask it of an economy and the growth debate rearranges itself. The degrowthers and the boosters are arguing about the water level, and the level was never the problem. Below the busy surface, enormous pools of money sit still, breeding what still water breeds, while everyone downstream waits for rain.

So the whole essay reduces to two sentences. Opportunity is what money does when it moves, so an economy that wants opportunity without end must keep its money moving. And money moves when sitting still is the most expensive thing it can do, which a modest fee on idle wealth, collected through the tax machinery we already run, is enough to arrange.

Wörgl stirred its stagnant pool with a 1%-a-month stamp. Let’s find a way to stir ours.