How it works
Start with the two-sector cartoon. Firms hire households and pay them wages. Households take those wages and buy the things firms make. That revenue pays next month's wages. Nothing is created or destroyed in the loop; the money just keeps moving, and the speed at which it moves is roughly what we mean by "the economy".
Real economies leak. Households save some of their income rather than spending it. Government takes some in tax. Some is spent on imports, which sends it out of the domestic loop. Against those leakages sit injections: business investment, government spending, exports. When injections match leakages, the flow is stable. When leakages exceed injections, every turn of the loop is smaller than the one before.
The multiplier, in invented numbers
These figures are made up and round, chosen to show the mechanism rather than to describe any actual economy.
Say national income is 100. Households spend 90 and save 10, and firms invest exactly 10. The loop is steady at 100 forever. Now firms cut investment to 5. The first round of income is 95, of which 85.5 gets spent, so the next round is smaller again, and so on. The flow does not settle back at 100. It settles where saving out of the smaller income equals the smaller investment: at a 10 per cent saving rate, that is an income of 50.
A cut of 5 in one part of the loop took 50 out of the whole thing. That is the multiplier, and it runs in both directions.
What people get wrong
The loop does not care why a piece of it broke. A factory closing, a tax rise, a wave of caution, an algorithm doing the job faster — the arithmetic downstream is identical, and no amount of arguing about whether the cause was good for productivity changes it.
The second thing is that money can leave the flow without leaving the country or even the bank. Income that piles up with people or firms who do not spend it has functionally stepped out of the circle. Circulation is what matters, not ownership.
Why it matters to this crash
Our displacement thesis is a circular-flow argument end to end. Wages are the largest single channel through which money re-enters the loop, and our own spiral piece puts US wages and salaries at $12.98 trillion in 2025, on BEA figures, against roughly 162 million people employed as of July 2026. Displace a quarter of that workforce and about $3.2 trillion of annual income leaves the flow, or nearer $4 trillion once you weight it towards the better-paid occupations most exposed to a machine that works through a screen.
The circular flow is the reason that loss does not stay with the displaced. Their forgone spending is someone else's forgone revenue, and that person's job is next. This is the , and it is not a metaphor. It is the multiplier running downhill.
It is also why the obvious fix is harder than it sounds. The parable in that dispatch is pure circular flow: the state hands out $100, people spend it at the one firm that makes everything, the state taxes the owner at 90 per cent and recovers $90. Next cycle it has $90 to distribute, then $81, then $73. Nobody cheated. A share extracted every cycle from a closed loop simply drains it, and the only rate that closes the circle is 100 per cent. The fiscal arithmetic points the same way. Individual income tax was $2.60 trillion of $5.24 trillion in federal receipts in FY2025, or 49.6 per cent, while corporate income tax was $497 billion. The event that creates the need for replacement income is the same event that removes half the revenue base.
What would make this dangerous
The observable that matters is aggregate wage income, not the unemployment rate. Total wages and salaries falling in nominal terms while corporate profits hold up would mean the loop is being rerouted away from spenders and towards savers, which is the precise condition under which the multiplier turns negative.
Four things would sharpen that. Employment falling in high-wage screen occupations rather than broadly, since the income lost per job in those fields is well above the all-occupation average. The household saving rate rising among those still employed, which is precaution and is itself a leakage. Discretionary consumer spending falling while headline GDP is held up by capital expenditure, which is what a loop breaking under cover of an investment boom looks like. And income replacement that covers a visible fraction of lost wages rather than the whole thing, because partial replacement slows a spiral without stopping it.
The falsifier is equally concrete. If displaced workers are re-employed at comparable wages within a few quarters, the flow repairs itself, the multiplier runs the other way, and this is an ordinary adjustment rather than a spiral. Watch re-employment wages, not re-employment counts.