Listen to the deep-dive discussion – Why Moody’s beats the Coca-Cola moat (19:37 min)

 

“We lived in a torrent of money, and we were constantly deploying it.”

— Charlie Munger, Berkshire Hathaway Annual Meeting, 2016

What if the torrent deployed itself?

I. Two Holdings, One Portfolio

Warren Buffett has held Coca-Cola since 1988 and Moody’s since its spin-off from Dun & Bradstreet in 2000. Both are classified as permanent holdings. Both have appeared in the Berkshire Hathaway portfolio for decades. Both carry the same label in Buffett’s vocabulary: wonderful businesses with durable competitive advantages, purchased at fair prices, held forever.

The criteria are identical. The metrics are similar. Both have high returns on equity. Both have wide margins. Both generate predictable cash flows. Both pay dividends. The spreadsheet sees two variations of the same theme — blue-chip compounders with impregnable moats. Morningstar gives both the same rating: wide moat.

The framework sees two different planets.

*     *     *

II. The Moat That Costs $4 Billion

Coca-Cola’s moat is its brand. The brand is the most recognized in the world. It is worth tens of billions in any valuation. And it costs $4 billion per year to maintain.

Four billion dollars in advertising, marketing, sponsorships, and promotional spending every year. Not to grow. To maintain. To keep the brand in the consumer’s mind. To defend shelf space. To counter the shift toward health-conscious beverages, water, kombucha, and functional drinks. The $4 billion is not an investment in the future. It is the cost of preventing the erosion of the present.

Stop the $4 billion and watch what happens. In year one, the brand holds. In year two, the shelf space begins to shrink. In year three, the competitors — who did not stop spending — have advanced. In year five, Coca-Cola is a regional brand with a fading logo. The moat is not self-reinforcing. It is maintained by an annual expenditure the size of a small country’s GDP.

This is a mechanical moat. It must be rebuilt every year. It is wide — $4 billion buys a very wide moat — but it is expensive. And the width depends on the spending. Reduce the spending and the width narrows. The moat is rented, not owned.

Coca-Cola’s ROE is 25%. Its growth is 2–3%. The gap — 22 percentage points — exits the business as dividends and buybacks. The 15/15 is broken. The cash recycles. Return OF capital. The business earns magnificently on existing equity and has nowhere to deploy the surplus because the B is dead. The consumer is migrating toward health. The GLP-1 drugs accelerate the migration. The addressable market is slowly, invisibly, shrinking.

*     *     *

III. The Moat That Strengthens With Crises

Moody’s moat is regulatory, institutional, and self-reinforcing. The regulators — Basel III, Dodd-Frank, global banking standards — require credit ratings from Nationally Recognized Statistical Rating Organizations. There are nine NRSROs in the world. Building the institutional credibility to become one takes decades. No AI model, no startup, no competitor can replicate a century of institutional trust with a better algorithm.

And the moat does not weaken with crises. It strengthens. After 2008, the regulatory response was not to eliminate ratings but to require more of them. After every credit event, the demand for independent assessment increases. The crisis validates the toll. The toll becomes more mandatory, not less.

The B costs nothing. The expansion of global credit markets — emerging market debt, corporate bonds, structured products, sovereign issuance — delivers volume to Moody’s without Moody’s spending a dollar to create it. The digitalization of the delivery mechanism means the marginal cost of an additional rating approaches zero. The wave is structural. The wave is free.

Moody’s ROE exceeds 30%. Revenue growth is 12% per year. The 15/15 holds at zero cost. And the surplus — the cash that a mechanical business would have spent on growth — is returned to shareholders through buybacks and dividends. Return ON infinity. The cash overflows by force.

*     *     *

IV. The Hierarchy Within the Moat

Buffett’s moat framework distinguishes between wide and narrow moats. A wide moat is durable. A narrow moat is vulnerable. The distinction is binary — the moat is either wide enough to protect the business for decades or it is not.

But within the category of “wide moat,” Buffett does not distinguish further. Coca-Cola is a wide moat. Moody’s is a wide moat. Both are durable. Both will exist in twenty years. The label is the same.

The framework introduces a hierarchy within the wide moat category. Not all wide moats are created equal. The width is the same. The nature is different.

A sand moat is created by external authority and dissolved by external authority. FICO’s moat was regulatory — Fannie Mae required FICO scores. When the FHFA approved VantageScore, the moat cracked. The sand dissolved. FICO lost 55% of its value.

A mechanical moat is maintained by continuous spending. Coca-Cola’s moat requires $4 billion per year. The moat is wide as long as the spending continues. It is rented from the market through advertising.

A coral moat is built by usage and strengthened by usage. Every transaction on Visa’s network makes the network more valuable. Every credit rating Moody’s delivers reinforces the institutional trust. The moat does not need to be maintained because the users maintain it by using it. The cost of maintenance is zero. The moat grows on its own.

Buffett’s formative period — the 1950s through 1980s — was the golden age of the mechanical B. American Express. See’s Candies. Coca-Cola. Gillette. Every great business of that era grew by reinvesting capital into brands, distribution, and products. The B was always mechanical. The B always cost something. And the moats that protected these businesses — brand, scale, distribution — were all mechanical moats, maintained by spending. In this world, the distinction between sand, mechanical, and coral did not exist because coral moats had not yet emerged. Every wide moat was mechanical. The hierarchy was invisible because only one level of the hierarchy existed.

The coral moat emerged with the digital age. Visa’s network became self-reinforcing as the world digitalized payments. Moody’s institutional trust deepened as global regulation expanded. MSCI’s index became the lingua franca as passive investing exploded. These moats did not exist in their natural form when Buffett was building his framework. They emerged in the 2000s and 2010s — after the vocabulary had been set, after the criteria had been established, after “wide moat” had become a binary label that made no distinction between a moat that costs $4 billion a year to rent and a moat that maintains itself for free. The hierarchy became visible only after the world produced the coral. And by then, the label was already fixed.

*     *     *

V. The Same Metrics, Two Architectures

	Coca-Cola (KO)	Moody’s (MCO)
ROE	~25%	~30%+
Revenue growth	2–3%	~12%
Moat type	Mechanical — brand maintained by $4B/yr	Coral — regulatory + institutional, self-reinforcing
Cost of B	High positive — B consumed by maintenance	Negative — B free + surplus returned
The 15/15	Broken — ROE 25%, growth 2%	Holds at zero cost — ROE 30%+, growth 12%
Cash behavior	Recycles — overflow by weakness	Overflows by force — growth AND return
PE (Spring 2026)	~25	~29
Morningstar	Wide moat	Wide moat
Buffett label	Wonderful business, hold forever	Wonderful business, hold forever
Framework	A alone — the visible trap	A × B — the invisible gift

 

Morningstar: wide moat, wide moat. Buffett: wonderful, wonderful. The framework: two different planets. The spreadsheet sees the same quality. The framework sees opposite architectures.

KO at PE 25. MCO at PE 29. The market charges four points of PE more for a business with 6× the growth, a self-reinforcing moat, and a negative cost of B. Four points of PE is the market’s price for the difference between a business that recycles and a business that compounds at infinity. The framework says the gap should be twenty points, not four.

The compression reveals a deeper pattern. The market systematically overvalues what it can see and undervalues what it cannot. KO’s value proposition is entirely visible — the 3% dividend yield, the 25% ROE, the brand recognition, the decades of uninterrupted payments. Every metric is clear, measurable, and reassuring. The market pays PE 25 for this visible safety. MCO’s deepest value is invisible — the self-reinforcing moat that costs nothing to maintain, the B that delivers 12% growth for free, the surplus that overflows by force rather than weakness. None of this appears in a standard screen. The market pays PE 29 — barely more than KO — because the invisible cannot command a premium in a model that only prices the visible. The dividend reassures. The coral is silent. And the silence is structurally undervalued.

*     *     *

VI. The Torrent and the Deployment

At the 2016 Annual Meeting, Buffett and Munger described the architecture of Berkshire with characteristic clarity. Munger called it “a torrent of money” — the float, the premiums, the constant inflow of capital at zero cost. And then the work: “we were constantly deploying it.”

The torrent is the float. $170 billion of capital that policyholders pay Berkshire to hold. The greatest A ever built — not merely a toll but a toll where the customers pay you to collect it. The cost of the float is negative in good underwriting years. The policyholders fund the capital. Berkshire invests it. The spread between the cost (negative) and the return (positive) is the engine.

“People didn’t — for a long time — appreciate the value of float.”

The market did not see the float for decades. It took Buffett years of explaining in his annual letters before the market understood that Berkshire’s insurance operation was not a business — it was a source of free capital. The float was invisible. The market eventually learned to see it. The market has not yet learned to see the free B.

But the torrent requires deployment. Every dollar of float must find a home. An acquisition. A stock purchase. A bond. A Treasury bill. Each deployment is a decision. Each decision carries the risk of error. Munger acknowledged this: “we were wising up as we went along.” Wising up implies prior unwisdom. Kraft Heinz. Precision Castparts. Dexter Shoes. The torrent absorbs the errors because it never stops flowing. But the errors are real.

“Sometimes it doesn’t look like we’ve accomplished much, and we haven’t accomplished much.”

Buffett admitted that some years the B produces nothing. The torrent flows. The targets do not appear. The cash pile grows. In 2016, this was a modest confession. In 2026, it is a structural diagnosis. Berkshire holds approximately $340 billion in cash and Treasury bills — more than the GDP of Finland. That $340 billion includes the $170 billion float — the accumulated premiums that policyholders have paid Berkshire to hold. The other $170 billion is accumulated profit that even the greatest allocator in history has not found a way to deploy. The torrent continues. The roads have run out.

The Freesurfer never has a year where it has not accomplished much. The B delivers every quarter. The digitalization does not pause. The indexation does not pause. The expansion of global credit does not pause. There is no dry year. There is no torrent waiting for deployment. The deployment is automatic, structural, and free.

“It’s allowed for a lot of mistakes… You don’t have to be smart. You just keep the standard stupidities out.”

Munger described the first chapter of The Infinite Investor without knowing it. Survival first. Inversion. Avoid ruin. The torrent is so powerful that avoiding stupidity is sufficient. Intelligence is a bonus. The architecture does the heavy lifting.

The framework recognizes its own foundation in Munger’s words. But it asks the question Munger did not ask: what if the architecture eliminated the mistakes instead of merely absorbing them? What if the B required no decisions at all? What if the torrent deployed itself?

*     *     *

VII. The Two Summits

Berkshire Hathaway is the summit of the mechanical world. The greatest A ever built — a toll where the customer pays you to collect it, where the revenue precedes the cost, and where the combined ratio below 100% means you are paid to receive debt. The greatest allocator who ever lived. Sixty years of compounding. A portfolio of wonderful businesses. The proof that the mechanical B, guided by exceptional judgment and financed by an A whose revenue arrives before its cost, can produce the most successful investment record in history.

The Freesurfer is the summit of the natural world. No float needed. No allocator needed. No decisions needed. The B is exogenous. The moat is self-reinforcing. The toll collects. The wave pushes. The cost is negative — the business pays you to hold the growth.

The two summits coexist. Both are investable. Both compound. But the first depends on the torrent and the pilot. The torrent continues after Buffett — the float is structural. The question is the pilot. Greg Abel inherits the torrent. The question is whether the mechanical world still offers enough B of sufficient quality at $340 billion of scale.

The second summit does not ask this question. The B arrives without a pilot. The scale is irrelevant — the wave is the entire global economy digitalizing, indexing, and expanding credit. There is no scale at which the B stops arriving.

Berkshire is paid to receive debt. The policyholder hands over the premium. The combined ratio below 100% means Berkshire profits from the transaction before the capital is even deployed. The float accumulates. The pilot deploys it into mechanical B. The system works because the A is temporally inverted — revenue before cost — and the pilot is the best who ever lived. The Freesurfer is paid to receive growth. The wave delivers volume. The toll collects. The buyback returns the surplus. No debt is created. No sinistre is owed. No pilot is needed. Berkshire is paid to carry risk. The Freesurfer is paid to carry certainty. The same gesture — being paid to receive — with opposite natures underneath.

This is not an argument against Berkshire. It is an argument for understanding what Berkshire is — the greatest monument ever built in the mechanical world — and what the Freesurfer reveals about the world that is emerging beside it. A world where the torrent deploys itself.

*     *     *

Watch the video

averagingup.com