The hedge usually looks perfect right up until the moment everyone reaches for the same exit.
Credits: Kubera
TL;DR
Diversification sounds great in calm markets because correlations look low, portfolio charts look smooth, and every asset seems to play its role. Then stress hits, liquidity vanishes, and the things that were supposed to protect you start dropping together. Diversification does not fail because the idea is wrong. It fails because most people diversify by label instead of by underlying risk, and crises have a nasty habit of revealing that half the portfolio was really the same bet in different clothes.
The comforting lie
One of the most comforting ideas in finance is that if you just spread your money around, you will be okay.
Some stocks. Some bonds. Maybe a little real estate. Maybe a little gold. Maybe an alt strategy if you want to feel sophisticated. You are not betting on one thing, so you must be safer than the person who went all in.
That logic works often enough to become doctrine. It also breaks in exactly the moment people care about it most.
I think that is why diversification feels so emotionally unfair. It does not usually fail on some random Tuesday in a boring market. It fails when headlines are ugly, volatility is spiking, and you are opening your app specifically because you thought at least one part of your portfolio would be holding up.
Then everything is red.
I think a lot about 2022
2022 is still the cleanest modern example of this.
For years, the default grown-up portfolio was some variation of 60/40. Stocks for growth. Bonds for foundation. It was not exciting, but it made intuitive sense. When stocks got hit, bonds were supposed to cushion the fall.
Then inflation showed up and broke the script.
Vanguard noted that target-date and balanced portfolios suffered in 2022 because stocks and bonds fell in tandem, which undercut the historical assumption that investment-grade bonds would act as ballast in rough markets. Morningstar said the same thing more bluntly. The “death of diversification” headlines were exaggerated, but the old stock-bond relationship absolutely failed that year because both sides were being hit by the same force, aggressive rate hikes to fight inflation.
That is the part people miss. Diversification did not fail because investors forgot to own enough different tickers. It failed because the driver underneath the portfolio was the same. Higher rates repriced both long-duration stocks and bonds at the same time. Two asset classes. One macro punch to the face.
If you opened a “balanced” portfolio in 2022 expecting one side to save the other, it was a rough lesson in how quickly a hedge can turn into a co-victim.
Then there is the uglier version
If 2022 was the clean academic example, March 2020 was the more frightening one.
That was the moment when even U.S. Treasuries, the thing people love to call the safest and most liquid market in the world, started selling off during the COVID panic. The IMF later described how longer-duration Treasuries sold off aggressively in mid-March, to the point that they were no longer behaving like the hedge asset investors expected. In its 2025 Global Financial Stability Report, the IMF again pointed to the March 2020 “dash for cash” as a case where Treasury selling intensified once volatility hit certain levels.
That is the version of diversification failure that really matters.
When people are scared enough, they do not sell what they want to sell. They sell what they can sell.
That is a very different market.
Treasuries got hit not because investors suddenly decided they hated the U.S. government. They got hit because funds needed cash, dealers were constrained, liquidity thinned out, and the clean textbook hedge became part of the scramble. The Federal Reserve has warned for years that in crisis conditions, historical correlations break down, liquidity becomes costly or unavailable, and “normal” risk management strategies may stop being useful.
That is the hidden line in all of this. The real enemy of diversification is not just correlation. It is liquidity stress.
And then there is leverage
There is another classic story people bring up for a reason: Long-Term Capital Management.
LTCM was built on smart people, beautiful math, and the assumption that relative value relationships would eventually normalize. The Federal Reserve later described LTCM as a case of excessive leverage and reliance on short-term financing. The 1998 Federal Open Market Committee transcript shows officials explicitly worrying that in stress, “not all the correlations would go to one” had been the assumption built into counterparties’ risk systems. Reality was less polite.
That is another way diversification fails when you need it most.
Leverage takes a good idea and removes its patience.
A diversified book might still work over a year. It might even work over a quarter. But if it is financed with short-term borrowing, daily variation margin, or investor redemptions, then the portfolio does not get to wait for the long-term logic to come true. Stress compresses time. Good positions become bad experiences.
This is why I have always found “correlation went to one” to be true, but incomplete. Sometimes correlations do spike. But often what really happens is worse. The market stops caring about your elegant distinctions because your financing structure and everyone else’s need for cash become the dominant facts on the screen.
What diversification actually is
I think most people quietly define diversification as “owning a lot of stuff.”
That is not crazy, but it is incomplete.
Real diversification is owning exposures that fail for different reasons.
That sounds obvious until you actually test your portfolio against the big categories of pain.
What happens if inflation is the problem?What happens if growth collapses?What happens if both stocks and bonds are expensive?What happens if liquidity disappears and the market sells whatever it can?What happens if regulation or politics hits a specific theme all at once?
Once you ask those questions, a lot of portfolios look less diversified than they did five minutes earlier.
CFA Institute has written that investors obsess over low correlation, but what matters just as much is whether return streams are genuinely different and whether those relationships hold under changing regimes. In a separate 2026 CFA piece, the point is even sharper. Static portfolios fail when volatility, inflation, and macro drivers change the underlying risk relationships. March 2020 was a liquidity shock. 2022 was an inflation shock. Same disappointment. Different reason.
That last part matters. Diversification does not fail in one single way. It fails differently depending on what kind of stress has shown up.
The part nobody likes to hear
Sometimes diversification is doing its job and it still does not feel good enough.
This is the part investors hate.
If you are down 12 percent while the concentrated guy is down 30 percent, diversification worked. It just did not save you from pain. It reduced pain. Those are not the same thing.
A lot of disappointment with diversification is really disappointment that it did not provide comfort when what it actually provides is damage control.
That distinction matters because it stops you from demanding something impossible from a portfolio. No allocation can guarantee that you make money during every crisis. What a good allocation can do is lower the chance that one narrative, one regime, or one liquidity event destroys the entire machine.
That is a much more boring promise than “weather all storms.” It is also more honest.
So why does it fail exactly when you need it most
Because that is when the market reveals what your portfolio was really made of.
In easy periods, everything gets a flattering backstory. Bonds are safe. Private assets are uncorrelated. Alternatives are diversifiers. Gold is insurance. Crypto is digital gold. Foreign exposure reduces domestic concentration. There is always a label ready.
Stress strips labels off.
What matters then is not what the asset was called in the pitch deck. What matters is whether it is liquid, whether it is leveraged, whether it shares the same macro driver as the rest of the book, and whether other people are trying to sell it at the same time.
That is why diversification tends to “fail” precisely when you need it most. That is when everyone finds out whether they built a portfolio of genuinely different risks or just assembled multiple expressions of the same one.
The real takeaway
I do not think the lesson is “diversification is fake.” That is too easy and too dumb.
The lesson is that diversification has to be treated as a live process, not a static identity. It is not enough to own different assets. You have to understand why they should behave differently and what kind of environment would make them stop doing that.
Because when stress hits, the market does not grade you on how many buckets you used. It grades you on whether the underlying drivers were actually independent.
And if they were not, you learn the same lesson a lot of people learned in 2020, in 2022, and long before that. The hedge looked beautiful in the brochure. Then the world changed.
Thank you for reading.
-APL
Footnote
Nothing here is financial, legal, or tax advice. This is a framework for thinking about why portfolios disappoint people under stress, not a claim that any one allocation solves that problem forever. Markets change, correlations move, liquidity can vanish, and even “safe” assets can behave badly at the wrong moment. Do your own research and build something you can actually live with when the screen turns red.
Sources: IMF, Federal Reserve, Federal Reserve, Federal Reserve, Morningstar, Vanguard, BlackRock, BlackRock, CFA Institute
Why Diversification Fails Exactly When You Need It Most was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.