Quality compounder investing

The most influential modern school is really late Buffett, generalized. Buy a small number of exceptional businesses that can reinvest at high returns for a very long time. Pay a sensible price, which needn't be a cheap one. Then mostly do nothing while compounding does the work. It's Fisher's growth-quality with a sharper focus on return on capital and the reinvestment runway, and it connects straight back to your Phase 4 moat work. A compounder is just a wide moat plus a long reinvestment runway, held. Three practitioners give it three memorable formulations.

What is Chuck Akre's three-legged stool?

Chuck Akre (Akre Capital, the Akre Focus fund) describes his ideal investment as a three-legged stool. Take away any leg and it falls over:

1. An extraordinary business, one that earns high returns on capital durably, protected by a real moat. This is the "compounding machine."
2. Excellent management: people who are skilled operators and also honest, shareholder-aligned capital allocators. They treat you as a partner and reinvest cash intelligently.
3. Reinvestment, the runway. This is the leg most investors forget, and Akre puts it at the center. The business must be able to reinvest its free cash flow at those same high rates of return for years. That can mean reinvesting in the business itself or in sensible new ones. A great business that can't reinvest hands back cash that you have to redeploy yourself, and pay tax on. A great business that can reinvest at 25% compounds internally and tax-deferred, and that's the eighth wonder. Akre calls this "the nature of the return": over a long holding period your return as a shareholder approximates the business's return on capital. So you want the highest sustainable, reinvestable ROC you can find. The returns-on-capital work promised this idea would come back in Phase 6, and here it is as a portfolio philosophy.

What is Terry Smith's three-sentence philosophy?

Terry Smith launched Fundsmith in 2010 on a three-sentence philosophy, "buy good companies, don't overpay, do nothing," and it's deeper than it looks:

1. Buy good companies. Smith's definition is quantitative and strict. He wants high returns on capital employed and high gross margins, a sign of pricing power or differentiation. He wants strong cash conversion, meaning profits that turn into cash, and growth funded from retained earnings instead of debt or dilution. And he wants resilience to technological change. He screens the world for the few dozen companies that clear this bar and largely ignores the rest.
2. Don't overpay. Valuation matters. Smith's controversial refinement is that for a high-ROCE compounder you can pay a high multiple and still do well, because over long horizons the compounding overwhelms the entry price. That's the duration point from reverse DCF, pushed hard. "Don't overpay" is a real constraint, but it tolerates multiples that would horrify a Graham value investor, and that's the quality-versus-price argument, a cousin of the concentration debate. The tension comes back in the common mistakes. "Don't overpay" can slide into "any price is fine for quality," and that's how quality investors get hurt when multiples finally compress.
3. Do nothing. Once you own a portfolio of exceptional businesses, trading is the enemy. It brings costs and taxes, and the near-certainty of selling a compounder too early, the error the selling lesson warns about. Fundsmith's turnover is famously minimal. "Do nothing" takes effort, because you're actively resisting the urge to act, and it's the hardest of the three to carry out. One caveat matters a great deal. "Do nothing" applies to winners whose thesis is intact. It's never an excuse to hold a business whose economics have broken, and that's the line between conviction and stubbornness.

What are scale economies shared?

Nick Sleep and Qais Zakaria ran the Nomad Investment Partnership from 2001 to 2013 and returned roughly ten times the money over its life, far ahead of world indices. Then they closed it and made the Nomad letters public. They're free, they're among the finest investing documents ever written, and you should read them. Their central discovery was a specific, powerful business model they called "scale economies shared." An ordinary firm that gains scale keeps the savings as higher margins. A scale-economies-shared firm deliberately gives the savings back to customers as lower prices. That drives more volume, which creates more scale, which lowers costs further, which funds still lower prices. It's a self-reinforcing flywheel (Greenwald and Helmer's scale economies, turned into a weapon) that builds an ever-widening cost moat along with a reservoir of customer goodwill no competitor can match.

Sleep's exemplar was Costco. Its membership model and deliberately capped markups pass scale savings to members. He held it, Amazon, and Berkshire in enormous concentration for years, practicing what he called "destination analysis." Ignore the quarterly noise, ask where this business is going over a decade, and if the flywheel is intact, hold.

Where does this school show up in India?

The scale-economies-shared model has a near-perfect Indian embodiment in DMart (Avenue Supermarts). It runs an everyday-low-price grocery model: thin gross margins, high inventory velocity, stores it owns instead of rents, ruthless cost control. It passes scale savings to customers to drive footfall and volume, which is Sleep's Costco flywheel with an Indian cost structure. Treat this as illustrative, and study the model in the annual report and in the consumer and retail lesson.

India has also grown a quality-compounder school of its own. Marcellus, under Saurabh Mukherjea, built a franchise on "consistent compounders," clean, high-ROCE Indian franchises held for the long run. Pulak Prasad's Nalanda Capital runs an extremely concentrated portfolio of high-quality, low-debt Indian businesses, bought to hold almost permanently, with a Darwinian emphasis on avoiding big mistakes over chasing big winners. Prasad's book What I Learned About Investing from Darwin (2023) is a modern classic. These are Akre's stool and Smith's screen in an Indian idiom, and they lead straight into the concentration debate.

When does this approach work, and when does it fail?

The quality school works magnificently over long horizons in stable, high-ROC industries (consumer, software, payments, medical, niche industrials) where moats are durable and reinvestment runways are long. It's the closest thing to "buy and forget" that actually compounds.

The last decade made its three failure modes vivid.

• Overpaying, the "quality at any price" trap. Even a wonderful business bought at 60× earnings can deliver mediocre returns for a decade if the multiple halves while earnings double. That's the 2022 quality-growth drawdown.
• Misjudging durability. A "compounder" whose moat has worn away sees its reinvestment runway shorten and its ROC fade to the hurdle, and then the premium multiple collapses.
• Using "do nothing" as an alibi for not selling a broken thesis, and holding a former compounder out of loyalty.

The habit from your moat work keeps this school out of trouble: the multiple is a duration claim, so check the CAP you're paying for against the evidence and keep watching the falsifiers.