Every school on the tour takes a side, openly or not, on the question that most divides serious investors: how many stocks should you own? Hold this debate evenhandedly. State the strongest case on each side and work out what actually decides the answer, so you don't just inherit a slogan. Here are both cases, argued fairly.
What is the case for concentration?
The case for concentration (Munger, Buffett, Pabrai, Sleep, Keynes). The argument is disarmingly simple. Your best idea is far better than your tenth, so why dilute the best with the mediocre? Charlie Munger championed "focus investing," a portfolio of perhaps ten holdings instead of a hundred, and he dismissed broad diversification for the informed as achieving nothing but the index minus costs. Buffett has two aphorisms for it. One is that "diversification is protection against ignorance; it makes little sense if you know what you are doing." The other is the "20-punch card." Imagine a card with room for only twenty investment decisions in your whole life. You'd choose each one with enormous care, and that scarcity would make you rich.
Keynes, running his college endowment, wrote that real conviction is rare: "there are seldom more than two or three enterprises at any given time in which I personally feel myself entitled to put full confidence." Pabrai's "few bets, big bets," Sleep's three-position concentration, and Nalanda's ~10 to 15 names are the same conviction in practice.
Underneath it runs Kelly, which you met with Pabrai. If you can size by edge, a large edge demands a large position, and equal weighting throws information away. Concentration is how deep knowledge turns into outsized returns.
Advertisement
What is the case for diversification?
The case for diversification (Graham, Schloss, and the index). The opposing argument is just as rigorous, and it starts from humility about your edge. Benjamin Graham advised broad diversification. He bought his net-nets as a group of 20 to 30, because any individual statistical bargain might fail while the basket reliably worked. His edge was statistical, so it needed breadth to show up. Walter Schloss, Graham's disciple, held ~100 deep-value bargains in small positions and slept well, because no single mistake could hurt him.
The index and efficient-market case, which the active-versus-passive debate in portfolio construction takes up, goes further. Most investors have no reliable edge. Individual businesses can go to zero for reasons no analysis catches (fraud, black swans, technological death). And broad diversification captures the equity premium while removing idiosyncratic risk essentially for free, the one "free lunch" in finance.
Diversification is the right answer whenever you can't reliably tell your best idea from your worst. The diversifier would add that this describes almost everyone, almost all the time.
What does the maths say about the debate?
The debate is more than philosophy. It has a mathematical spine that both sides have to respect. The decision-science and position-sizing lessons make it rigorous:
- Kelly says size by edge, but use a fraction, because you don't know the true odds. Full Kelly maximizes long-run growth if your probability estimates are exact. They never are, and overestimating your edge, which is the overconfidence bias, turns full Kelly into a ruin machine. So serious concentrators use fractional Kelly, which already pushes you away from extreme concentration and toward more, smaller positions than raw conviction suggests.
- Ergodicity and ruin cap concentration. A bet that's attractive "on average," with positive expected value across many parallel universes, can still ruin you on your one actual path through time. All it takes is a loss big enough to knock you out of the game, because after ruin there's no compounding left to recover. So avoiding permanent, portfolio-level impairment is a precondition, and you don't get to treat it as a preference. No single position, however high the conviction, should be able to ruin you if its thesis proves catastrophically wrong, through fraud, say. That's the hard mathematical limit on concentration. It's why even Buffett caps individual positions, and why "bet big" always means "bet big within survival."
- Hidden correlation can make "diversified" a lie. The number of names isn't the number of bets. Thirty stocks that are all Indian NBFCs, or all US high-multiple software, or all levered to the same interest-rate or commodity factor, are one bet held thirty times. The portfolio looks diversified and behaves like a single position when the shared factor turns. IL&FS and DHFL took the whole NBFC complex down together, as the NBFC lesson showed. Real diversification spreads across uncorrelated risk factors, and ticker symbols don't count. Turn it around and a "concentrated" portfolio of ten uncorrelated businesses may carry less risk than thirty correlated ones. That one idea reframes the whole debate. Stop asking "how many names?" and ask "how many independent bets do I have, each sized so that being wrong on it, or on the factor they share, can't ruin me?"
When should you concentrate, and when should you diversify?
So what decides the right answer? There's no universal number. There's a dial, and where you set it depends on factors you have to assess candidly. Turn it toward concentration when these hold:
- your edge is deep understanding of a few businesses
- you do the work to know each holding cold
- the businesses are high-quality and durable, so the permanent-loss risk per name is low
- your temperament can hold through the violent drawdowns concentration guarantees without capitulating
- your holdings really are uncorrelated.
Turn it toward diversification when these hold:
- your edge is statistical: deep value, quant, factor baskets where individuals fail but the group works
- your information is unreliable because fraud and governance risk run higher (promoter and accounting-quality risk give Indian investors a structural reason to diversify more or dig far deeper)
- you can't reliably rank your ideas
- your temperament needs the sleep-at-night breadth Schloss described
- you simply lack a demonstrable edge, in which case broad diversification or indexing is the correct choice.
Where do most great investors end up?
The common heuristic "concentrate to build wealth, diversify to preserve it" captures part of the truth, since outsized fortunes are usually built concentrated and kept diversified. Treat it as a prompt and never as a law. The quality and reliability of your edge, set against the survival constraint, always decides it.
The mature position, which most great investors end up at, is neither extreme. Concentrate enough to express real conviction and let your edge matter. Diversify enough that no single mistake, and no single shared factor, can permanently impair you. Then set the exact number by your real edge (not the one you imagine), your temperament and the reliability of your information. For a skilled investor doing deep work on durable businesses, that's often somewhere in the range of 8 to 20 really independent positions. That's concentrated by index standards and diversified by Munger's. Position sizes come from conviction × downside, which is fractional-Kelly logic. One hard rule says no position's worst case can sink the ship. You'll write your own concentration policy down when you reach portfolio construction, and the arguments to defend it are all here. The last word is a candid one. This is a contested question, reasonable masters sit at both ends, and the right answer is the one you can carry out under pressure, given who you actually are.