The Diversification Fallacy: Why More Isn't Always Safer

Asset Allocation Series · Risk Layer. For the framework, see the overview. The last post covered how drawdown determines your allocation. This one takes apart the most common illusion: what you think is diversification probably isn't.

01One: A Portfolio That Looks Diversified

Imagine an investor who proudly tells you his portfolio is "very diversified" —

He holds 30 stocks (covering tech, consumer, healthcare, finance), 5 different funds, and some ETFs. He says: "I'm so diversified that a single company blowup can't hurt me. I'm safe."

It sounds bulletproof. 30 stocks + 5 funds — how is that not "diversified"?

But here's the brutal truth: he may not be diversified at all. He may only own "one thing," just sliced into 35 pieces.

Why? Because the essence of diversification isn't "having many things" — it's "having things that won't all fall together in a crisis." And those 30 stocks + 5 funds? Chances are — in a real crisis, they'll all crash in lockstep.

That's false diversification — looking diversified (owning many things) but actually concentrated (they share the same risk source and fail together). It's the most common and most dangerous illusion in allocation.

02Two: True vs False Diversification — It's About Correlation

To understand the difference, you need one core concept: correlation.

The real purpose of diversification isn't "owning many things" — it's "owning things with low correlation." When one falls, the other doesn't (or even rises), cushioning the blow.

True diversification: Your assets have low correlation. When A drops, B holds or rises. The whole portfolio's volatility is dampened. That's the real protection.

False diversification: Your assets have high correlation. They look different (different stocks, different funds), but in a crisis they move together. A drops, B drops, C drops — they all fail. This kind of "diversification" offers zero protection in a crisis, because everything collapses at once.

Back to that investor — his 30 stocks may cover tech, consumer, healthcare, finance, but they're all stocks, all sharing the same risk source: a systemic equity market downturn. When a crisis like 2008 hits, all stocks (regardless of sector) crash together — their correlation in a panic approaches 1 (all down). His "30-stock diversification" is, in a systemic crisis, roughly equivalent to holding one stock market index — he hasn't diversified away the biggest risk.

Worse, those 5 funds — if they're equity funds — simply pile another layer of overlap on an already equity-heavy bet. He isn't diversified; he's using 35 different names to place the same bet (the stock market).

03Three: The Most Dangerous False Diversification in 2026 — When You Think You Bought the Index

The most insidious form of false diversification appears in 2026 — when you buy a "broad index," thinking you own the most diversified thing, but you're actually making a hidden concentrated bet.

I've touched on this in the overview, the Munger piece, and "Why the Core Should Be a Broad ETF." Let me lay it out fully here —

You buy the S&P 500, thinking you're diversified across 500 companies. The truth: the top 7 companies (Mag 7) account for roughly one-third of the index's market cap. That means every 100 dollars you put in, 33 dollars are betting on 7 highly correlated tech giants.

Deeper still — these 7 aren't just concentrated in size; their correlation is also extremely high (they're all tech stocks, all driven by interest rates, AI narrative, and tech sector sentiment). So — when tech goes through a systemic correction, all 7 fall together, and because they make up a third of your index, they drag the entire index down.

You thought you bought "S&P 500" with 500 companies' diversification. In reality, it looks more and more like a concentrated bet on 7 tech giants, with 493 minor characters along for the ride.

That's the most common false diversification of 2026: hundreds of millions of investors think they've achieved diversification by buying an index, but their real risk exposure is heavily concentrated in a handful of tech mega-caps. If those few companies suffer (valuation mean-reversion, AI narrative collapse, regulatory crackdown), every index investor who felt "diversified" will feel the pain together.

04Four: How to Break Through — Look at Your Real Risk Exposure

So how do you avoid false diversification? There's one core method: look through your portfolio to see "real risk exposure," not "nominal holding count."

I pointed this out in the risk-parity section of the framework (the biggest insight from risk parity is focusing on real risk, not nominal allocation). Here's how to do it —

First, ask "where does my risk really come from?" rather than "how many things do I hold?" Group all your holdings by the risk sources that actually drive their ups and downs — for example, "systematic equity market risk," "interest rate risk," "single-sector risk," "single-country risk." Then see how much of your money is concentrated in one source.

That investor with 30 stocks? Once he looks through, he'll see that 90% of his risk comes from one source: systematic equity market risk. His diversification is fake.

Second, ask "in a real crisis, would all these things fall together?" This is the most direct test. Put your portfolio into an imagined 2008-style crisis — which ones would fall together? If almost everything would, you have false diversification.

Third, deliberately allocate to things with "low correlation." True diversification requires actively adding assets that "don't necessarily fall when stocks fall" — cash (rock-solid in a panic), gold (hedge against systemic collapse), some bonds (hedge against growth panics), assets in different countries (diversify single-country risk). This is exactly what the "asset layer" does — achieve real diversification by mixing asset classes with different correlations, not by "false diversifying" within equities.

Fourth, look through your index too. Even if your core is an index, periodically check its "top concentration." If the index itself is becoming more concentrated in a few stocks (like the S&P 500 in 2026), use some tools to offset — allocate a bit to equal-weight indices, value factors, or overseas (remember the overseas piece) to balance the overexposure to the top.

05Five: My Difference from the Mainstream — Diversification Is a Matter of "Quality," Not "Quantity"

The mainstream understanding of diversification is almost entirely about "quantity" — "the more things you hold, the more diversified and safe you are." That's a fundamental misunderstanding.

Diversification is a question of quality, not quantity.

Holding 100 highly correlated stocks is far less diversified than holding just 4 asset classes with low correlation: stocks + bonds + gold + cash. The first is "100 slices of the same risk"; the second is "4 different risk sources." Real diversification depends on the diversity of risk sources, not the number of holdings.

And the mainstream misses an even deeper point — excessive "quantitative diversification" can actually be harmful.

If you hold too many things (50 stocks, 20 funds) just for the sake of diversification, two bad things happen: First, you don't understand any of them well enough (attention diluted) to judge their quality. Second, your portfolio drifts toward the market average, but you pay active-management costs without getting the benefits of an index (you'd be better off just buying the index). This kind of "diversification for its own sake" is fake diligence — it makes you feel safe (you own so many things), but it doesn't truly spread risk (they're all correlated), and it dilutes your knowledge and energy.

My stance — pursue "few but high-quality true diversification" (a few low-correlation asset classes) instead of "many but messy false diversification" (a pile of highly correlated holdings). This goes back to Munger's "circle of competence" and concentration — better to hold a few asset classes you understand, with low correlation, plus a self-renewing broad index, than to hold 50 things you don't understand that all move together. Quality always beats quantity.

06Six: Final Thoughts

False diversification is the coziest illusion in allocation.

It's cozy because it gives you a feeling of "I'm safe" — you own so many things, spread across so many sectors and funds, clearly "not putting all eggs in one basket." That feeling lets you sleep at night.

But it's an illusion. Because — if all those "different baskets" are actually hanging on the same fragile rope (the same risk source), when the rope breaks, all baskets fall together. Your "diversification" across 35 baskets, when that rope snaps, is no different from owning one basket.

That's the most dangerous thing about false diversification: it gives you false security in calm times, letting you relax your guard. Then in a real systemic crisis, it reveals the truth: you were never diversified — you just sliced the same bet into many pieces.

Real diversification isn't comfortable. It requires you to do something counterintuitive: hold things that "drag you down" in bull markets (cash, gold, bonds). These things make you less money when markets surge, make you look "less aggressive," and get you mocked by the fully invested crowd. But precisely these — the things that don't move with stocks, that have low correlation — are what provide real protection when a systemic crisis hits and all stocks crash together.

The price of true diversification is "earning less" and "feeling uncomfortable" in normal times. Its reward is not being dragged down together in a crisis. And most people, unable to stomach the normal-time discomfort, choose false diversification (all highly correlated equities) — and then pay a heavy price in the crisis.

Diversification is about quality, not quantity. Look through to risk. Don't count how many things you hold. Ask: "Will my money all fall together in a crisis?"

If the answer is yes, then no matter how many things you hold, you only own one.

Next post: the third and most insidious killer in the risk layer — sequence risk: why "when you lose" matters more than "how much you lose."