Morgan Stanley's Ridham Desai On Where Indian Markets Are Headed

ELI5 / TLDR

Ridham Desai, Morgan Stanley’s India strategist of 35 years, spends most of this conversation not on stock tips but on how he thinks. His core idea: the biggest profits sit exactly where things feel most uncertain — when nobody can tell whether the next outcome is a doubling or a halving, that’s the moment worth buying. He thinks the worst was already behind us by April 2026, that retail Indian investors are unusually patient and getting it right, and that the price-to-earnings ratio everyone obsesses over tells you almost nothing about what a stock is actually worth.

The Full Story

This is a long, meandering podcast — closer to a fireside lecture than a market call. Desai is the kind of person who reads eleven hours a day and would rather explain Claude Shannon’s information theory than give you a Sensex target. The interview rewards patience. Here is the spine of it.

Reading is for entertainment, not for getting good at things

The clip that went viral had Desai saying “reading is a waste of time.” Here he walks it back and sharpens it. His point was never anti-reading — he reads constantly. His point is that some things cannot be learned from a page.

“Cooking, plumbing, running, meditating, investing — you can read forty or fifty books and still not know how to do any of them. You learn by doing.”

He read roughly 40-50 investing books — Schroeder on Buffett, Peter Lynch, Benjamin Graham, Howard Marks — and insists none of them made him an investor. Practice did. His rough split: 80% of investing skill comes from doing it, 20% from reading. And most of his own reading, he says, is for entertainment, the way other people watch films. He doesn’t watch TV or movies; he reads instead.

Why read at all, then? For subjects where practice isn’t available. History (which “doesn’t repeat but rhymes,” and which you need to make sense of today’s geopolitics). Physics, chemistry, maths. The Charlie Munger argument: read widely across disciplines so your model of the world gets richer, and any new situation becomes legible.

A worry about AI and “intellectual laziness”

A digression worth flagging. Desai is deep into David Deutsch’s The Fabric of Reality and The Beginning of Infinity. Deutsch argued there’s infinite progress ahead because of human creativity. Desai now thinks that claim can be challenged — because AI may make people “intellectually lazy at scale.”

“AI is doing the thinking, so humans think less. And if human beings stop thinking, then ironically AI will slow down the progress of mankind.”

His generation will be fine, he says; the question is the next two. If they outsource their intellect, progress stalls.

The market is a hospital, not a physics lab

Desai likes to locate the stock market among the sciences. People assume it’s mathematics, because it’s all numbers and formulas. He disagrees. Maths is precise — two plus two is always four. Markets are not. A stock on a P/E of 5 can underperform one on a P/E of 100.

The closest science, he argues, is medicine. A doctor has general principles, but every patient is unique, so the prescription is custom each time. Same with companies:

“Every company is like a different body. You apply the principles, but the decision you make about that company is distinct — you can’t transplant it to the next one.”

That said, physics donates a few powerful concepts.

Entropy. Disorder in the universe only ever increases; time runs one way; a smashed cup doesn’t reassemble. In markets, disorder keeps rising and you have to spend energy to impose any order — and even then it slips.

Ergodicity. A system is ergodic when its average over time equals its average across many parallel instances. Markets are non-ergodic — the two are very different. His illustration is Russian roulette: across 100 separate people the expected payout is calculable, but for you playing once, there’s no meaningful “average.” The probability over time and the probability right now are not the same thing. For a portfolio: the return it shows today and the return it delivers over time can diverge wildly.

The Kelly criterion, and why sizing beats stock-picking

This is the heart of the episode. Desai traces a lineage from Claude Shannon (Bell Labs cryptographer, father of information theory) to John Kelly, who turned a special case of information theory into a betting formula — originally for horse racing.

The simplified version, via William Poundstone’s book Fortune’s Formula:

“The percentage of your bankroll that should go on a bet equals your edge divided by your odds.”

Edge is the extra information or insight you have — the gap between what you believe and what the market believes. If you think a company grows 20% and the market thinks 10%, your edge is that gap. Odds is the payoff if you’re right. Worked through: a 20% edge with a stock that doubles if you’re right implies putting a large chunk of your bankroll on it. (In practice people use “half-Kelly” because full Kelly is aggressive.)

The punchline, and the most quotable idea in the talk:

“Identifying a winner is not as difficult as deciding how much to put in. Winners are all over the place — you’ll find ten in a year. But you can’t size them, so you don’t get wealthy. Position sizing is far more critical than identifying the winner.”

His read on the late Rakesh Jhunjhunwala fits here: the differentiating factor was conviction translated into size — going all in (and even levering) when conviction was high. Desai claims most successful investors — Buffett included — are using Kelly’s logic, knowingly or not.

Frank Knight: maximum profit lives at maximum uncertainty

The other load-bearing idea comes from 1920s economist Frank Knight: all profit resides in uncertainty. If the future were certain, you’d earn a bond-like, fixed-deposit return — and that’s not really profit. Profit is the reward for bearing uncertainty.

Desai’s refinement: the point of maximum profit is the point of maximum uncertainty. His examples are March-April 2020 (COVID, when nobody knew if they’d survive, let alone what markets would do) and the present geopolitical moment.

“If it feels uncertain, there has to be a lot of reward waiting for you.”

How to operationalise “uncertainty,” which is otherwise vague? Look at the gap between the bull case and the bear case for a situation. When that gap is wide — bad outcome halves your money, good outcome doubles it — that’s high uncertainty and high potential reward. When the gap narrows (everything feels certain, calm), there’s little left to make. He points to mid-2024, post-election: “everything became certain, three months later the market made its peak.” His claim is that the hardest moment was already behind us — around end-March/early-April 2026.

His actual market view (briefly)

For the institutional crowd, Morgan Stanley sets index targets twice a year (currently June 2027), built on 10-15 assumptions — “we don’t actually know the future; it’s a scenario.” But he’s emphatic that retail investors should ignore one-year targets entirely.

“If you have less than five years, you shouldn’t be in the stock market at all. Want a one-year return? Buy a bond. Compounding happens over 5, 10, 15 years.”

His most genuinely interesting observation is about Indian retail behaviour. The current generation of Indian retail investors, he argues, is more long-term than institutions — they didn’t churn during the bad stretch, and they bought in March-April when the market was ugly. That’s the opposite of global retail, which typically sells into weakness and buys into strength. He attributes part of this to the SIP (systematic investment plan) habit. He frames a structural tailwind: equity is still only ~7-8% of India’s household savings pool (gold is ~12-13%), versus ~40% in the US and ~50% in Europe — even conservative Japan is around 20%. That share, he expects, climbs over the next 10-15 years.

Value is not the P/E ratio

A recurring riff. People treat the P/E ratio as a measure of value. Desai calls it dangerous shorthand.

“P/E gives you nothing about value. Valuation is the same as price. Value is something else entirely.”

Value, properly defined, is the discounted stream of all future cash flows — the dividends, discounted at the return you require. He credits John Burr Williams (1938 PhD) with formalising this: value a bond by its interest, a stock by its dividends. The trouble with equities is that, unlike a bond, the “coupon” (dividend) isn’t fixed and the duration is endless — it can grow or shrink. That’s exactly what makes equities a hard asset class, and what makes a blunt “buy low-P/E stocks” screen fail: some 5-P/E stocks have shrinking cash flows and go bust; some 100-P/E stocks have real growth ahead and deliver.

His old SBI-vs-HDFC example (SBI at 1x book, HDFC at high price-to-book, yet HDFC not destined to underperform) lands on the single metric he says matters most:

“Return on capital is the most important thing. Return on share price ends up being return on equity. If the company doesn’t make money on its capital, you won’t make money on the stock.”

The second-order point: a high return on capital matters most when the company can reinvest at that same high return — that compounding is what actually creates stock-market wealth. And the offsetting risk discipline, straight from Munger via Graham: pay too much even for a wonderful business and you won’t make money; pay little for a poor one and you still might. Graham started with cigar-butts (stocks below net working capital, free puffs off discarded ends), and Munger pivoted Buffett toward “growth at a reasonable price” once those bargains dried up in the 1970s. Desai’s closing caveat is honest:

“These are all judgments. It’s easy for me to say on a podcast. In real life it’s very hard. Don’t underestimate how difficult this is to actually do.”

Key Takeaways

Maximum profit sits at maximum uncertainty (Frank Knight). Operationalise it by measuring the gap between bull and bear cases — wide gap means high uncertainty and high reward.
Position sizing matters more than stock-picking. Finding winners is easy; sizing them correctly is what builds wealth. The Kelly criterion (bet = edge ÷ odds) formalises this; most great investors use it knowingly or not.
Edge = the gap between your view and the market’s view. No gap, no edge.
P/E ratio tells you nothing about value. Value is discounted future cash flows. A 5-P/E stock can be a trap; a 100-P/E stock can be cheap.
Return on capital is the single most important metric, and it compounds only when the company can reinvest at that same high rate.
Don’t overpay even for great businesses; don’t avoid cheap mediocre ones reflexively — risk management is mostly about price paid.
Retail investors should ignore one-year index targets. Under a five-year horizon, buy bonds, not stocks.
Indian retail is unusually patient — bought into the March-April 2026 weakness, behaving more long-term than institutions, helped by SIP habit.
Equity is ~7-8% of Indian household savings vs 40-50% in developed markets; structural room to grow over 10-15 years.
Markets are non-ergodic — your single lived path differs from the statistical average; and subject to entropy — disorder rises unless you spend energy.
He believes the hardest moment was already behind us (around end-March / early-April 2026).

Claude’s Take

This is a good listen, but you have to mine it. Desai is genuinely well-read and the mental-model scaffolding — Knight’s uncertainty, Kelly sizing, value-versus-P/E — is the durable, transferable stuff, and it’s correct. The Kelly point especially (“sizing beats picking”) is the kind of thing that’s obvious once said and ignored by almost everyone in practice. Worth the price of admission.

A few honest caveats. First, the “maximum profit at maximum uncertainty” framing is true and survivorship-flavoured — uncertainty is also where you lose everything, and “buy when it feels worst” is advice that’s clean in hindsight and brutal in real time. Desai concedes exactly this at the end (“don’t underestimate how difficult this is”), which is to his credit. Second, there’s a soft house-view talking-book underneath: the long, glowing passage on patient Indian retail and the under-allocated savings pool is also, conveniently, the bull case Morgan Stanley sells. It’s probably true, but it’s not disinterested. Third, the actual market call is deliberately thin — he gives an institutional June-2027 target framework but refuses to translate it into anything retail-actionable, which is the responsible move but means the clickbait title oversells.

The transcript is a machine-mangled Hinglish mess (the auto-transcription badly garbles the English finance terms), so some numeric specifics — exact Sensex bull/bear figures, precise dates — couldn’t be recovered cleanly and aren’t reported here. I’ve reconstructed the conceptual content, which survives the noise; treat any specific number as approximate.

Score 7: high signal on frameworks and investor psychology, dragged down by a vague headline promise, a detectable institutional tilt, and a transcript that hides the harder data points. As a thinking-tools episode rather than a market-call episode, it’s a solid 7.

Morgan Stanley's Ridham Desai On Where Indian Markets Are Headed | Kushal Lodha #333