Where Does the Return Come From?

Every graduation date on this site — every "about eleven years" — rests on a single input. That input is the return the community's businesses earn on the capital invested in them. So far we have assumed a healthy figure and moved on. That is a placeholder. And a placeholder is exactly the kind of thing an honest model must, sooner or later, cash out or drop.

A skeptic's first question is the right one: where does that return actually come from? If the answer is "we assumed it," the whole edifice is a wish. So here we replace the assumption with a mechanism. The good news: the mechanism is simple enough to hold in your head. It is also honest enough to tell you when the number is unreachable.

Start with the honest quantity. It is not "how many members spend how much." It is what a real business actually earns. A business is built with some capital and runs at some margin. The profit it throws off is a fact of its economics, and that profit is where the return lives. The investors who funded the business take a slice of that profit (call the slice PSP). The rest stays inside the business to graduate its own people. The return is a slice, not the whole pie. It is earned, not assumed.

So how much does a real business really make? It depends on which of two things it is. Grocery is the low-margin doorway (8–20%). But a pooled buying operation cycles its capital astonishingly fast, because orders are collected before the goods are bought. And the businesses members build after the doorway run genuinely fat margins. Processed food runs 50–85% (hot sauce, pickles, frozen paratha). Telecom runs 40–60%, cleaning products 50–70%, services and trades 30–50%. All are capital-light. Why do such economics hold here and almost nowhere else? Because captive member demand — guaranteed customers, zero marketing spend — meets real margins. That profit, earned from real commerce, is the raw material of the return. We make both shapes precise below.

There is no money from nowhere. Every business runs the same model. Its customers are often the community's own people in a different role, so the profit traces to real commerce at a real margin. The one discipline: a business's profit funds either the community's investment pool or its investors' return. Never the same dollar twice.

Define the quantities precisely. A business serves m members. Each routes spend dollars of their monthly purchasing through this business — not their whole household budget, just the part this business captures. The business runs at a net margin and was built with capital C. It pays its investors a governed slice PSP of its profit. Then the annual return on the capital is:

That is the demand reading — member commerce in the numerator, capital in the denominator. But the same identity factors into a form every accountant will recognize. The reason: m·spend·12/C is just the business's annual revenue over its capital — its capital turns:

Three factors, three different owners. Margin is a property of the business type. Grocery retail runs 8–20%, services 30–50%, processed food 50–85%. Capital-turns is how many times a year the capital cycles through commerce. This is exactly where member spending enters the physics. PSP is a governance choice.

Here is the scientific claim, stated so it can be checked. In the open market, a business's profit really does behave as a function of its capital. Why? Because capital must buy not only capacity but the fight for uncertain demand: marketing (commonly 10–20% of revenue), sales, churn, idle capacity. Competition then bids returns down toward market rates. In this model the fight is deleted: demand is membership. Members' routed spending is committed before the business is built. So the business runs near capacity from day one, and the cost of winning customers approaches zero. Capital still sets the ceiling per cycle — a business cannot process more commerce than its build supports. But member spend is what guarantees the cycles. Profit still scales with capital once demand fills the capacity. What the community controls is the rate: margin × turns × slice.

And the equation is falsifiable in both directions. It says out loud when a target is impossible: a capital-heavy business serving few members cannot clear a high return at any honest slice. It also says when a dial combination is a fantasy. Turns have physical reference classes — a factory does not cycle its capital weekly — and the calculator below flags you when you leave them.

Because ROI = PSP × margin × turns, there are exactly two honest ways a community business clears a high return. They sit at opposite corners of the margin-turns plane:

• The distributor: thin margin × fast turns. The buying club itself is the archetype. Orders are pooled first, then the goods are bought. So capital is mostly a working-capital float that cycles weekly or faster, with no shelf risk. Real-world reference: grocery and produce distributors turn inventory 20–75 times a year. At a 10% margin and ~50 turns, the business's own profit runs ~500% of its small float in a year — because the float is tiny, not because money multiplies. A 10% investor slice of that is ~50%.
• The producer: fat margin × modest turns. The businesses members build after the doorway — hot sauce, frozen paratha, cleaning products, services. A $30k frozen-food operation doing $150k a year is 5 turns. At a 60% margin that is $90k of profit — 300% on the capital as business profit, and a 30% investor return at a 10% slice. Even halved for caution, the target survives.

Both shapes share the same enabling condition: member-density — committed demand per dollar of capital. What kills a candidate business is the third corner: capital-heavy and thin-margin and slow. The equation prices that honestly, at a return no slice can rescue. That is the discipline the CDF applies before funding anything. The target return is not hoped for. It is selected for.

Turn the dials — including the margin, because it is the anchor, not a footnote. The full causal chain is computed live below the dials: routed spend → revenue → profit → investor slice → return on capital. The capital-turns readout tells you which reference class your combination lives in. It also tells you when the combination has left physical plausibility altogether.

Everything above is a single business paying its investors from its ongoing profit. The real shape is a recursive network. Each business reinvests part of its return to spawn a child business, owned by the same people. So returns flow up the branches. A member holds stakes across many businesses, and each throws off its own slice. Founding a root means owning the whole tree it grows.

Two honest warnings, because we found them the hard way when we simulated the network:

• The compounding is not free. The recursion doesn't only add. There is an optimum reinvestment: around 70% held, 30% paid out. Reinvest too little and the tree is undercapitalized. Reinvest too much and the payout starves, and graduation slows. A real U-curve, not "more is always better." (Businesses That Spawn Businesses.)
• A tempting version is a bubble. One could pay investors from the surplus of the newcomers a business brings in (a "delayed enrollment" of new customers). That income is transient. It dries up the moment the market fills, and the whole thing collapses. The return has to rest on permanent, productive profit. Newcomer contribution can only be a transient accelerant on top. (The Network Sustains.)

So the equation above is the per-business floor, grounded in real profit. The network compounds it — and does so sustainably when it is built on productive businesses and tuned near the reinvestment optimum. We do not assume that here. We simulated it, failure modes included, in the studies linked above.

From here, the return stops being an input we assume. It becomes an output we measure. The economy closes honestly. Real businesses earn real profit at a real margin. Investors take a disciplined slice. No dollar is spent twice. That is a far stronger claim than "assume 60%." It is falsifiable, and it invites you to check the arithmetic rather than trust the author.

And we did check it, all the way up. We let a business actually be a community — one that pays its investors out of its own profit while graduating its own members. Then we built a network of them. We asked whether everyone still reaches independence when the return is earned rather than granted. The answer, with the failure modes caught and corrected, is yes. The network sustains, and everyone graduates. That is no longer the next step. It is the capstone finding.