From Bottom to Best: The Incredible Turnaround of 3 Invesco Funds l Power Talks Ep 12

ELI5 / TLDR

A fund manager named Aditya Khemani took over three Invesco mutual funds in late 2023, when they were mediocre and held about ₹9,000 crore. Two years later they are top performers and hold ₹27,000 crore. His explanation is almost boring: build a balanced portfolio, never let one stock get too big, pick a handful of long-term themes (he loves hospitals and quick-commerce), and care as much about how the portfolio behaves on a bad day as on a good one. He also admits the part most fund managers won’t: his own personal stock-picking has probably done worse than the funds he runs, so even he puts his money in mutual funds.

The Full Story

The turnaround came from construction, not stock-picking

The interviewer opens with the obvious question: three funds went from bottom-half to top-of-category, and from ₹9,000 crore to ₹27,000 crore, in two years. What changed?

Khemani’s answer is not “I found great stocks.” It’s that he rebuilt how the portfolios are assembled. Think of it like the difference between buying good ingredients and being a good cook. He says the second one took him longer to learn:

“Maybe 10 years back… you’re not thinking as much about portfolio construction. But today… stock picking is stock picking, it takes time, but I think portfolio construction also takes a lot of time.”

He breaks “balance” into three axes. First, market-cap mix — a deliberate blend of large, mid, and small companies. Second, style mix — he holds both value stocks (cheap, out-of-favour) and growth stocks (expanding fast), plus “growth at a reasonable price” in between. The logic: nobody can reliably predict whether value or growth will be in fashion, so own both. Third, sector spread — don’t bet the farm on one industry.

The 5% rule, and thinking in pairs

The market regulator (SEBI) lets a single stock be up to 10% of a fund. Khemani treats that as far too risky and caps any one holding near 5%. The reasoning is short-term survival: if your biggest bet is 10% and it stumbles, it drags the whole fund for months, even if the stock is fine long-term.

The more interesting idea is that he thinks about how sectors move together. Imagine two dominoes tied with string — knock one and the other tends to fall too. Auto companies and the lenders who finance cars (“NBFCs,” non-bank finance companies) rise and fall as a pair. So if he wants to overweight one, he deliberately underweights the other, otherwise the fund’s swings get amplified. Same logic links stock exchanges, asset managers, and real estate — all do well when markets are buoyant and people feel rich.

“You have to think what can go wrong and you have to construct a portfolio accordingly — not what can go right.”

He stress-tests by watching real bad days: the 2024 election-result crash, the India–Pakistan flare-up. If the fund falls hard on those days, that’s his red flag that he’s taken on too much risk.

Top-down themes, bottom-up stocks

He picks maybe eight to ten long-term themes, decides how much of the fund each deserves (say 10% to hospitals), then hunts for the specific companies to fill that slot — a different hospital chain in the small-cap fund versus the mid-cap fund. The target shape of the whole portfolio: roughly 20% earnings growth and ~20% return on equity (a measure of how efficiently a company turns shareholder money into profit), while keeping valuations sane.

On the worry that his top 10 holdings now make up 40–45% of the fund (up from 25%), he’s unbothered. Beating a cheap index fund requires having strong opinions; a portfolio of 50 stocks is already plenty diversified, and 40% in the top 10 isn’t extreme. Liquidity — being able to sell without crashing the price — only bites in the tiniest companies (under ~$1 billion), and he keeps exposure there below 5–6%.

Why he leans mid and small, and loves hospitals

While other managers fled to large caps fearing mid/small-cap overvaluation, his funds stayed tilted toward mid and small. He freely grants they’re overvalued by maybe 15–20%, but argues that matters only if you care about the next six months. Over five to ten years, the interesting new themes simply live in mid and small caps:

“Electronic manufacturing… hospital… CDMO… auto ancillary… real estate… building products — large cap is very restrictive.”

Large caps, he says, are mostly banks and bloated IT firms that can’t grow fast. The emerging companies are smaller by definition.

Healthcare is his signature overweight. His framing: it’s a form of consumption, but non-discretionary — when money’s tight you skip a new TV, you don’t skip the hospital. Three tailwinds stack up: COVID made people anxious about health, health insurance now lets ordinary people afford top hospitals, and an ageing, richer population spends more on care. Crucially, he’s bullish on the services side (hospitals, contract drug research/manufacturing called CDMO) not the pharmacy side — of ~20% healthcare exposure, only 3–4% is actual drug-makers.

New-age companies, honestly hedged

He owns Swiggy and Zomato-type names, unusual for a traditional fund. His pitch: quick-commerce is a ~$10 billion slice of a $600–700 billion grocery market, in “habit formation” mode — once you’re hooked on 10-minute delivery, you don’t go back, and monetisation follows (just as food delivery eventually settled into a profitable two-player market). But he’s candid that valuing these is guesswork, off by 20–30%, and calls this “one part of the portfolio where I have some risk sitting.” It needs watching every quarter.

On India, and the quiet confession

He’s bullish on India’s macro — best balance of payments, inflation, and forex reserves in 20 years — and waves off AI-envy of the US and China with a line the host loved:

“Everything you can import and export, but people at scale no country can import or export.”

His honesty peaks at the end. Asked about his personal portfolio, he admits his own stock-picking has likely underperformed the funds he runs — partly a conflict-of-interest problem (the best stocks go into the fund, not his account), partly that solo investors end up with ten undiversified names. So he puts his own money into mutual funds, mostly his own. His closing advice to a 30-year-old: the habit of investing matters more than the amount; 15% compounding turns ₹100 into ₹400 over ten years; buy more when markets are bad.

Key Takeaways

• Three Invesco funds went from bottom-half to top-of-category and ₹9,000cr → ₹27,000cr in two years (Nov 2023 onward), driven by portfolio construction rather than individual stock picks.
• Hard 5% cap on any single holding, versus SEBI’s 10% limit — to avoid one bad stock dragging short-term performance.
• “Balance” across three axes: market-cap mix, value-vs-growth style mix, and sector spread.
• Correlated sectors are managed in pairs (auto + auto-financing NBFCs; exchanges/AMCs + real estate) — overweight one, underweight its partner to dampen volatility.
• Construct for what can go wrong, then stress-test against actual bad days (election crash, geopolitical scares); poor performance on a bad day signals too much risk.
• Top-down theme selection (8–10 themes) filled with bottom-up stock picks; target portfolio ~20% earnings growth and ~20% ROE.
• Top-10 holdings at ~40% concentration is considered fine; 50 stocks is already well-diversified.
• Liquidity risk only matters below ~$1 billion market cap; sub-$1bn exposure kept under 5–6%.
• Mid/small-cap tilt justified by where emerging themes live (electronics, hospitals, CDMO, building products), accepting 15–20% overvaluation for 5–10 year horizons.
• Healthcare overweight (~20%) framed as non-discretionary consumption, concentrated in services (hospitals, CDMO) not pharma (only 3–4%).
• Invesco runs a deliberately narrow ~300-stock universe with only 7–8 analysts; a stock averages ~20 days of research, sometimes months, and fund managers act as “gatekeepers” — the analyst must build conviction first.
• Quick-commerce held as a small, high-risk, “habit-formation” bet; valuation admitted to be imprecise by 20–30%.
• The manager admits his personal stock-picking underperforms his funds and parks his own money in mutual funds — partly due to conflict-of-interest (best ideas go to the fund).
• Lowered Indian inflation (~6% → ~4%) means lower nominal sales growth, so expect 12–14% equity compounding ahead rather than the past decade’s 16–17% — similar real returns.

Claude’s Take

This is a clean, jargon-light window into how an actively managed Indian equity fund actually gets built, and it’s refreshingly free of the usual fund-manager hype. Khemani’s framework is genuinely useful: the 5% cap, the paired-sector thinking, and the “judge the portfolio on a bad day” discipline are real, transferable ideas, not slogans. The single most credible moment is his admission that his own stock-picking probably lags his funds — a working fund manager saying “buy index/mutual funds, including over my own hands” is the opposite of a sales pitch and earns trust.

Caveats worth holding. This is a PowerUp Money channel interview, and PowerUp sells investing products, so the framing is friendly. The whole conversation is a survivorship story — we’re hearing the construction philosophy of funds after two great years; the same process during a mid/small-cap drawdown would read very differently, and 2023–25 was a tailwind period for exactly his tilts. His healthcare and quick-commerce theses are plausible but he’s talking his own book, and he half-admits the quick-commerce valuation is a guess. Nothing here is falsified or misleading, just necessarily one-sided.

A 7: substantive, honest, and a good primer on portfolio construction, but it’s a single practitioner’s narrative during a favourable window rather than tested or independent analysis.