Correlation is the glue that breaks diversification; tail risk is the lesson of every crash.
What you will learn
Understand correlation's role in diversification
Explain tail risk and fat tails
Build a portfolio that survives the tail
Correlation and spread
Correlation and spread
Tail risk in a crash
Tail risk in a crash
Correlation is everything
Diversification only works when assets are uncorrelated — they don't all fall together. Holding 20 stocks is fine until a crash, when correlations spike to ~1 and everything falls at once. The diversification you thought you had disappears exactly when you need it.
💡 The correlation spike
In calm markets, stocks, crypto, and even bonds can look independent. In a crisis, 'risk-off' sends them all down together (except true safe havens). This is why naive diversification fails in crashes — and why you need assets with genuinely different drivers.
Tail risk and fat tails
Returns are not a perfect bell curve — they have fat tails: extreme events happen far more often than the normal distribution predicts. A '1-in-10,000-year' crash can happen in a decade. Tail risk is the risk of these rare-but-devastating events.
Hedging the tail
Tail hedges (out-of-the-money puts, gold, long-volatility, cash) lose a little in normal times but pay off big in crashes. They're insurance — a steady cost that prevents ruin. The question is how much insurance to carry given its drag on returns.
💡 The barbell strategy
One approach: a barbell — mostly ultra-safe assets (cash, bonds) on one end, a small high-risk/high-reward allocation on the other, and almost nothing in the 'medium risk' middle. The safe end guarantees survival; the risky end provides upside. You can't lose more than the risky slice.
The takeaway
Correlation spikes and fat tails mean you can't rely on back-of-envelope diversification. Build a portfolio with truly different drivers, carry tail insurance, and size so the tail doesn't end you. Assume the 'impossible' will happen — because eventually, it does.
❓ Quick check
In a crisis, asset correlations tend to:
A) Fall to 0
B) Spike toward 1
C) Stay low
D) Reverse
Everything falls together.
(Knowledge check — full exam is next)
Key takeaways
Diversification needs uncorrelated assets; correlations spike in crashes
Fat tails: extreme events are more common than the bell curve says
Barbell + tail hedges = survive the unthinkable
📝 Weekly Exam — pass with 80% to unlock next week
10 questions. Review the Deep Dive and courses before attempting.
1. Diversification works best when assets are:
Different drivers.
2. In a crisis, correlations tend to:
Risk-off correlation spike.
3. 'Fat tails' means extreme events are:
Extremes more frequent.
4. Tail risk is the risk of:
The extreme tail.
5. A tail hedge is like:
Crash insurance.
6. A barbell portfolio holds:
Two extremes.
7. The barbell's safe end guarantees:
Survival.
8. Correlation of 1.0 means assets:
Perfectly together.
9. The cost of tail insurance is:
Insurance has a premium.
10. The core lesson of tail risk is to:
Prepare for the tail.
Your score: —
🛠 Weekly Project
Check your portfolio's real correlation.
1
List your top 5 (demo) holdings.
2
For each pair, estimate whether they move together or independently.
3
Identify which assets would fall together in a crash.
4
Write 2 sentences on where you'd add a genuinely uncorrelated asset or a tail hedge.