“What an organism feeds upon is negative entropy.”
— Erwin Schrödinger, What is Life?, 1944
Download the Slides Deck: Capital_Absorption
Listen to the deep-dive discussion – The Impossible Triangle of Compounding
I. The Cult of the Compounder
No word in investing carries more reverence than compounding. The compounder is the prize — the business that turns one dollar into ten, then ten into a hundred, while the patient owner does nothing but wait. And the definition has hardened into a single sentence, repeated until it sounds like law: a compounder is a business that reinvests its earnings at a high rate of return.
That sentence describes a single business and mistakes it for a universal law. Compounding is not a property a business either has or lacks, and it is not one mechanism shared by all who earn the label. It is a consequence — the visible result of a deeper force. How a business compounds, how fast, and by what means is determined by the direction in which capital flows through it, and the rate at which it does. Change the direction of that flow and you change the nature of the compounding. This essay is about that underlying force, and about why the celebrated compounder of the cliche — the one that reinvests everything at a high rate forever — is not the rule but the rarest exception, structurally almost impossible to find.
II. The Toll, the Wave, and the Direction of Absorption
Every business is a flow of capital between two reservoirs. The A is the toll — the cash the business collects today, short duration, cash now. The B is the wave — the growth that makes tomorrow’s toll larger than today’s, long duration, cash later. Almost every business carries both, in proportions that vary; together they form its dual duration.
These two reservoirs have different time horizons, and together they give every business a dual duration. The A is short duration — cash collected now. The B is long duration — cash that arrives later, as the growth matures. This matters for compounding because compounding acts on the long-duration component: it is the B that compounds, the growth that builds on itself over years, while the A is harvested in the present. A business that is almost all A and very little B has little long duration to compound; a business pouring its toll into a large, growing B is mostly long duration. The balance between the two durations is the first clue to how — and how fast — a business will compound.
Between these two reservoirs, capital flows in one of two directions, and the direction is universal — it applies to every business model without exception. In a business of quality, the A absorbs the B: the toll generates more than the growth consumes, and the surplus overflows, returned to owners. In a weaker one, the B absorbs the A: the costly growth devours the cash the toll produces. This is the spectrum of absorption, and nothing escapes it.
The word absorption invites a wrong picture, and it is worth dispelling before it misleads. Absorption here is not the black hole, which draws matter past a horizon from which nothing returns — that is destruction, the pole where the B absorbs the A and the structure is devoured by what it takes in. Nor is it gravity, which merely aggregates, mass drawing mass into a larger inert heap without transforming anything, and without any direction in time. Capital absorption is neither. It is incorporation: the toll takes in the growth the world offers and converts it into a larger toll tomorrow — the flow is not swallowed and not merely piled up, but converted. What enters as B leaves as compounding.
This is why the direction of the flow decides everything. A system left to itself runs down — it drifts toward uniformity, exhaustion, the flat equilibrium that in a living thing is death and in a business is the slow erasure of returns. The only escape is to draw order in faster than it is spent. That is what Schrödinger meant: what an organism feeds upon is negative entropy — order captured from its environment to hold off its own decay. A business is no different. The world supplies the B; the great business is the one whose A captures that order and compounds it, staying permanently ahead of the drift toward exhaustion.
This is not a metaphor borrowed from physics. It is the same form of law, instantiated in a different substance. The physicist Ilya Prigogine showed that an open system — one that exchanges matter and energy with its surroundings — maintains and builds its order only so long as it draws in more order than it dissipates within. A living cell obeys this; so does a business. Neither is the other, and capital is not energy. But both are open systems sustained by a flow they capture from outside themselves, and both are governed by the same structural law: absorb the flow and organize, or lose it and dissolve. When the A absorbs the B, the structure compounds. When the B absorbs the A, the flow reverses and the structure drifts toward dissolution. Compounding is what a business does while it remains on the ordered side of that flow.
Even the business with almost no growth sits on this spectrum. A magnificent toll with a tiny B — a See’s Candies, a Coca-Cola — is simply the pole where the B approaches zero and the overflow approaches its maximum. It does not stand outside the frame; it occupies its far edge. And as we will see, a business’s position on this spectrum of absorption is precisely what determines how it compounds — the mechanism, and the speed.
III. The Dollar Test
Buffett gave the discipline its sharpest measure. Retaining earnings is justified, he wrote, only when each dollar retained creates at least a dollar of market value over time. The test is not whether the business earns a high return on the capital it already holds — it is whether the capital it holds back and reinvests produces incremental value. The word that matters is incremental. A business can earn thirty percent on its existing capital and only eight percent on the next dollar it reinvests. Compounding lives on the second number, not the first.
But the dollar test contains a hidden assumption: that the business can reinvest at all, and in size. How much a business can retain and redeploy — and how fast that redeployment compounds — is not a free choice. It is dictated by where the business sits on the spectrum of absorption. And that is where the single sentence breaks apart into a spectrum.
IV. Compounding Is a Spectrum, Not a Switch
The speed and mechanism of compounding are a reading of the direction and intensity of absorption. Three quantities move together, and their relationship defines how a business compounds: how much it overflows (the surplus the toll throws off beyond what the growth consumes), how much it returns (the cash sent to owners), and how fast it compounds (the rate at which the earning base itself grows).
These three are bound by a structural tension. A business that overflows enormously has little growth to fund, so it compounds slowly. A business that compounds fast must be pouring its cash into growth, so it overflows little. You cannot, as a rule, have both the largest overflow and the fastest compounding at once. Each business model resolves the tension differently — and lands at a different point on the spectrum. Walk the spectrum, and the single word “compounder” dissolves into distinct species.
V. Two Variables, and Why Speed Misleads
Two distinct variables govern the flow, and confusing them is the source of most errors about compounding. The first is direction: whether the toll absorbs the growth or the growth absorbs the toll — whether the business is on the road to value or the road to ruin. The second is rate: how intensely the absorption runs in that direction. And the two are independent.
Consider two businesses both firmly on the right road — in both, the toll covers its growth and value is created. They share the same direction. Yet the intensity of their absorption could not be more different. A capital-heavy compounder — a railway reinvesting in its network, a wide-moat business plowing earnings into expansion — absorbs its B intensely: it takes the cash its toll throws off and pours almost all of it back into a growing base, incorporating its growth rather than releasing it. See’s Candies does the opposite: its toll is enormous and its internal growth is tiny, so there is almost nothing to reinvest in — it can absorb only a sliver of what it earns, and nearly everything overflows to the owner. Same direction, opposite intensities. And here is what the single word conceals: absorption and overflow are the two halves of one flow. What a business absorbs, it reinvests and compounds; what it cannot absorb, it overflows and returns. The railway absorbs almost everything and overflows little; See’s absorbs almost nothing and overflows nearly all. Absorption and overflow run in opposite directions.
Here intuition is set right, because the intensity of absorption runs with the speed of compounding, not against it. The business that absorbs its growth most fully — the railway, reinvesting nearly all its cash into an expanding base — compounds the fastest, because compounding is simply absorption sustained over time: capital taken in, redeployed, and taken in again. The business that absorbs its growth barely — See’s, with almost nothing to reinvest, overflowing everything — compounds the slowest, because a business that returns its cash is not building on it. The strongest absorption produces the thinnest overflow and the fastest compounding; the barest absorption, the largest overflow and the slowest. What overflows is not compounding; only what is absorbed compounds.
But speed is not a function of absorption alone, because a second factor sits between them: the return earned on whatever is reinvested. Two businesses can absorb their B just as intensely — both plowing most of their cash back — and yet compound at entirely different speeds if one reinvests at fifteen percent and the other at eight. Absorption sets the quantity reinvested; the incremental return sets what that quantity earns; and the speed of compounding is the product of the two. This is exactly what Buffett’s dollar test measures in a single stroke — not just whether a dollar is retained, but whether each retained dollar comes back as more than a dollar. Quantity and return, asked together.
And this is why speed, for all the worship it receives, is the most overrated dimension of compounding. Because the value created is exponential, it is decided far more by how long the compounding lasts than by how fast it runs. A business compounding at twelve percent for thirty years buries one compounding at twenty-five percent for four. The fastest compounder — the capital-heavy reinvestor — is also the most conditional, lasting only as long as its growth keeps clearing the cost of capital and its moat keeps holding. The highest speed carries the shortest certainty. This is the quiet case for the satisfactory compounder over the spectacular one: a moderate rate that endures, behind a durable barrier, outruns a high rate that fades.
VI. One Pole: The A That No Longer Absorbs
At one end of the axis stands the business whose toll is magnificent but whose growth is spent. See’s Candies, Coca-Cola — a fortress brand, superb margins, returns on capital few businesses approach. But the internal B is exhausted: the market for boxed chocolates or sugared water barely grows, and no amount of reinvestment changes that. There is almost nothing left to absorb. So the A stops reabsorbing and begins to overflow — it returns its cash, and the most powerful return is the repurchase of its own shares, each buyback shrinking the count and lifting the value of the shares that remain. The compounding here is real but slow, and largely mechanical rather than organic. This is the pole the cliché cannot explain: a wonderful business that compounds gently because it has run out of B to feed on. It overflows almost everything, and compounds little.
VII. The Middle: Where Most Businesses Gravitate
Between the poles lies the vast middle, where nearly every business sits, its position fixed by the mix of its dual duration — how much short-duration toll it collects now against how much long-duration growth it still funds. A business here absorbs part of its cash into a living B and returns the rest; its place on the axis is simply the ratio of the two. And the middle has its own breadth. At its heavy end sits the reinvestor whose toll absorbs a large and living B intensely — little overflows, because most of what it earns is plowed back at modest returns into an ever-widening base, compounding slowly in rate but enormously in absolute terms, for as long as the base keeps widening. At its light end sits the business that can absorb only a little of what it earns — its growth is nearly spent, so most of the cash overflows and is returned.
The businesses of the middle are best understood as points in motion. A company’s position is set by its dual duration — the mix of short-duration toll and long-duration growth it carries — and that mix shifts over time as the business matures, migrating along the axis. Berkshire Hathaway funds a vast long-duration B out of its short-duration insurance and operating tolls; Alphabet does the same in a different register, its established tolls funding the build-out of new franchises. Neither is fixed: as a long-duration B matures into a short-duration toll, the business drifts toward the overflowing pole; as it opens new long-duration growth, it moves back toward the reinvesting one. The middle is not a category but a passage — the stretch of the axis every enduring business travels as its durations rebalance. Absorption itself asks for no moat; a business can absorb its growth, or be absorbed by it, with no barrier at all. But how long a favorable direction lasts, and how cleanly the surplus overflows rather than being spent to hold the position, is where the franchise tells. The moat governs not whether a business absorbs, but the purity and duration of the absorption once it does — and it is that protected, enduring absorption, not the mere fact of it, that turns a favorable flow into real and lasting value creation.
VIII. The Other Pole: When the B Absorbs the A
At the far end the flow runs the other way. Here the growth is costly and the toll cannot yet contain it: the B absorbs the A, the reinvestment consumes more than the business earns, and cash flows out faster than it comes in. This is the direction of ruin — unless it is the direction of a deliberate wager. A business may pour capital into a growth it does not yet earn, betting that the toll it is building will one day be large enough to absorb the B it is now feeding. Amazon spent years in this posture: its retail and infrastructure spending devoured its cash, and the market watched a business that seemed to consume everything it took in — until the direction reversed, and a toll it had been constructing all along began to absorb the growth that had been absorbing it. The wager is precisely a wager on the direction of the flow: that the A will come to absorb the B before the B exhausts the A. Most businesses in this posture never reverse. The rare one that does is the reason the posture is attempted at all.
IX. Why the Perfect Compounder Is the Rarest Point
Now the perfect compounder of the cliché comes into focus — the business that reinvests a large quantity of capital, at a high incremental return, for a very long time. It would have to maximize three ingredients at once: a high return on reinvested capital, a large quantity that can be reinvested, and a long duration over which both persist. And these three are in structural tension.
The tension is this: a high return on capital usually comes from being capital-light — and being capital-light means there is little capital that can be absorbed. The very property that produces the high return forecloses the large quantity. The business at the exhausted pole earns spectacular returns but has almost nothing left to reinvest, so it overflows what it cannot absorb. The heavy reinvestor absorbs capital in size, precisely because its toll funds a large and living B — but at more modest returns, and only while its moat holds. What a business cannot absorb, it returns; the recycling spread — the gap between what a great toll generates and what it can profitably reinvest — is exactly this law made visible. Each point on the axis buys one virtue by surrendering another. Almost no business escapes the trade-off.
And duration is the scarcest ingredient of all. Compounding is exponential, so its value is decided less by the rate than by the number of years it survives — and the years are bought only by a moat that competition cannot cross. A high return without a barrier fades quickly; the compounding stops before it matters. The perfect compounder must hold all three — rate, quantity, and duration — against a market that is forever competing each of them away. This is why it is the rarest point on the spectrum, and why Buffett, who spent a career searching for it, found so few.
X. Compounding Is a Consequence
Return to where we began. The cliche treats compounding as a thing a business possesses — a high-reinvestment engine bolted on, present or absent. But every business we have walked through compounds as a consequence of one prior fact: the direction and intensity with which its toll absorbs its growth. At one pole, the exhausted toll can absorb almost nothing, so it overflows completely and compounds only slowly, through buybacks. In the middle, a business compounds in proportion to how much of its growth it absorbs — the more it absorbs, the more it plows back and the faster it builds; the less it absorbs, the more it overflows and returns. At the far pole, the wagering business absorbs more than it earns, and compounds only if its direction reverses. Different compoundings, each dictated by a single underlying variable — not chosen, but determined by where the business sits on the axis of absorption.
This is why the perfect compounder is a myth rather than a target. It would require a direction and rate of absorption that no real business sustains — a toll large enough to overflow, a growth cheap enough to fund without draining, a return high enough to matter, and a moat durable enough to hold all of it against competition for decades. The ingredients are not independent; they are bound in tension by the same flow that produces them. The high return comes from being capital-light, which forecloses the large reinvestment. The fast compounding comes from heavy reinvestment, which forecloses the overflow. The flow that gives with one hand takes with the other.
So do not ask whether a business compounds. Ask in which direction its capital flows, and how fast — because that, and not a label, is what compounding actually is. The cliche names a single rare point on a spectrum and mistakes it for the whole. The investor who understands that compounding is the consequence of absorption sees the entire spectrum, and knows that what looks like a search for the perfect compounder is really a choice among trade-offs that the direction of the flow has already set.
Watch the Video
This content is educational and reflects a personal analytical framework. It does not constitute investment advice.