Your client owns 30 stocks and thinks she is diversified. Every one is a U.S. large-cap tech company. She has variety. She does not have diversification.
Harry Markowitz proved that portfolio risk is not the weighted average of individual risks. When you combine assets that do not move in lockstep, portfolio standard deviation drops below the weighted average. The driver is correlation.
Correlation ranges from -1 to +1. At +1, two assets move identically and diversification does nothing. At -1, you can theoretically build a zero-risk portfolio. In practice, major asset classes fall between 0.0 and 0.8.
HIGH-FREQUENCY: The exam asks: "At what correlation does diversification begin to provide benefit?" Answer: anything below +1.0. You do not need negative correlation.
The two-asset portfolio standard deviation formula makes this concrete. For 60% Stock A (std dev 20%) and 40% Stock B (std dev 25%) with correlation 0.30:
Weighted average std dev = 0.60 x 20% + 0.40 x 25% = 22% Actual portfolio std dev = square root of [(0.36 x 0.04) + (0.16 x 0.0625) +...
Common mistakes
- Confusing asset allocation with asset location. Allocation determines what percentage goes to each asset class. Location determines which account type holds it. "Move bonds from taxable to Roth" wastes the Roth's tax-free growth on low-return assets. Bonds go in the traditional IRA.
- Believing more holdings means more diversification. Thirty technology stocks are not diversified. Diversification requires low correlation, not high headcount. Trap: "portfolio standard deviation equals the weighted average", that only happens at correlation +1.0.
- Rebalancing too frequently in taxable accounts. Selling appreciated positions triggers capital gains. Direct new money and income to underweight classes before selling anything.
Bottom line
- Any correlation below +1.0 provides diversification benefit; you do not need negative correlation
- Expected return is always the simple weighted average, unaffected by correlation
- Strategic allocation sets the long-term mix; tactical allocation makes temporary deviations from market views
- Rebalance tax-efficiently: new money first, redirect income, harvest losses, sell appreciated lots last
Exam shortcut
When the exam gives a client with multiple account types and asks what goes where, use the hierarchy: bonds/REITs in traditional IRA, aggressive growth in Roth, low-turnover index in taxable. Asset location shows up on nearly every form. Remember: "Below +1 = Benefit." Correlation of +1.0 means zero diversification. Anything below +1.0 helps. CML = standard deviation = portfolios only. SML = beta = everything.
The full lesson (about 2,117 words, 14 min read) adds 2 worked examples, all 6 common mistakes, a self-check, free in the app.
Learning objectives
- D.31
Browse all free CFP lessons or jump into free CFP practice questions.