Modern Portfolio Theory & the Callan Table.
Harry Markowitz won the 1990 Nobel Prize for a single insight: diversification isn't just prudent, it's mathematically superior. Holding uncorrelated assets produces higher returns per unit of risk than any single holding — and the Callan Periodic Table proves it visually every year.
The Core Idea
In 1952, Harry Markowitz published "Portfolio Selection" in the Journal of Finance. The radical claim: investors shouldn't pick the best stock — they should pick the best combination of stocks. Because different assets zig and zag at different times, combining them smooths the ride without giving up much return.
This gave us three tools still used by every institutional investor: the efficient frontier (the curve of portfolios that deliver maximum return for each level of risk), the risk-free rate benchmark, and correlation matrices (pairwise co-movement of assets).
The Callan Periodic Table of Investment Returns
Each column below is a calendar year. Each row is the rank of that asset class's return — rank #1 is the best performer, #10 is the worst. Read it top-down and you'll see one thing: no color stays on top. Last year's winner is routinely next year's loser.
| RANK | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|---|---|---|---|---|
| #1 | SC Value +31.7% | EM +37.3% | T-Bills +1.9% | LC Growth +36.4% | LC Growth +38.5% | REIT +41.3% | T-Bills +1.5% | LC Growth +42.7% | LC Growth +33.4% | EM +14.2% |
| #2 | LC Value +17.3% | LC Growth +30.2% | Core Bond +0.0% | REIT +28.7% | SC Growth +34.6% | SC Value +28.3% | LC Value -7.5% | SC Growth +18.7% | SC Growth +15.2% | Intl Dev +12.7% |
| #3 | High Yield +17.1% | Intl Dev +25.0% | LC Growth -1.5% | SC Growth +28.5% | EM +18.3% | LC Growth +27.6% | High Yield -11.2% | Intl Dev +18.2% | LC Value +14.4% | LC Growth +8.9% |
| #4 | SC Growth +11.3% | SC Growth +22.2% | High Yield -2.1% | LC Value +26.5% | Intl Dev +7.8% | LC Value +25.2% | Core Bond -13.0% | SC Value +14.6% | High Yield +8.2% | LC Value +5.4% |
| #5 | EM +11.2% | LC Value +13.7% | REIT -4.0% | SC Value +22.4% | Core Bond +7.5% | Intl Dev +11.3% | SC Value -14.5% | High Yield +13.4% | SC Value +8.1% | High Yield +4.6% |
| #6 | REIT +8.6% | REIT +8.7% | LC Value -8.3% | Intl Dev +22.0% | High Yield +7.1% | High Yield +5.3% | Intl Dev -14.5% | LC Value +11.5% | EM +7.5% | T-Bills +4.4% |
| #7 | LC Growth +7.1% | SC Value +7.8% | SC Growth -9.3% | EM +18.4% | SC Value +4.6% | SC Growth +2.8% | EM -20.1% | REIT +11.4% | T-Bills +5.3% | SC Growth +3.1% |
| #8 | Core Bond +2.6% | High Yield +7.5% | SC Value -12.9% | High Yield +14.3% | LC Value +2.8% | T-Bills +0.0% | REIT -25.1% | EM +9.8% | REIT +4.9% | Core Bond +2.9% |
| #9 | Intl Dev +1.0% | Core Bond +3.5% | Intl Dev -13.8% | Core Bond +8.7% | T-Bills +0.7% | Core Bond -1.5% | SC Growth -26.4% | Core Bond +5.5% | Intl Dev +3.8% | SC Value +1.8% |
| #10 | T-Bills +0.3% | T-Bills +0.9% | EM -14.6% | T-Bills +2.3% | REIT -5.1% | EM -2.5% | LC Growth -29.1% | T-Bills +5.1% | Core Bond +1.3% | REIT -1.2% |
What the Table Proves
I. Leadership Rotates
In 2020 Large-Cap Growth led at +38.5%. In 2022 it was last at -29.1%. The #1 slot changed 8 of the last 10 years.
II. Concentration Hurts
A 100% EM investor earned +37% in 2017 and lost -20% in 2022. Same person, three-year drawdown of ~35% — hard to sit through.
III. Diversification Wins
A balanced blend of the 10 asset classes ranked 4th–6th every year — never #1, never #10. Steady outperformance via reduced drawdowns.
The Efficient Frontier
If you plot every possible portfolio on a chart with risk on the X-axis and return on the Y-axis, the upper-left edge of the cloud is the efficient frontier. Any portfolio on that curve delivers the best return available for its level of risk. Portfolios inside the curve are suboptimal — you can always find another with same return but less risk, or same risk but more return.
For each Capital Wealth model portfolio, we run a Monte Carlo on 10,000 asset-class combinations, plot them on the risk-return plane, then select the combination that sits on the frontier closest to the client's risk tolerance (measured by the quiz). It's not magic — it's math.
Where MPT's Critics Are Right
| Critique | Why It Matters | Our Workaround |
|---|---|---|
| Correlations spike in crashes | In 2008 and March 2020, stocks, bonds, and REITs all fell together. Diversification failed when you needed it most. | Add real assets (gold, commodities) and structured notes with built-in buffers. |
| Historical returns ≠ future | MPT uses past means and variances as inputs. The future isn't obligated to cooperate. | Use forward-looking capital market assumptions from JPMorgan and BlackRock, not just history. |
| Normal distribution assumption | Returns have "fat tails" — extreme events happen more often than bell-curve math predicts. | Stress-test portfolios against 1987, 2008, 2020 scenarios, not just standard deviation. |
How This Applies To You
- If you're 5+ years from retirement: lean into equity diversification (5–7 asset classes), accept volatility for higher expected return.
- If you're within 5 years: the Callan rotation matters most — sequence-of-returns risk can permanently impair a retirement if you're over-concentrated right before withdrawals start.
- If you're in retirement: the "bucket approach" (cash bucket 0-2yr, bond bucket 2-7yr, stock bucket 7+yr) is MPT applied to time horizons.
Want your own efficient frontier?
We'll run your actual portfolio (401k, IRA, pension, annuities, bank) through our risk-return optimizer and show you whether you're on the frontier — or sitting inside it, leaving return or taking unnecessary risk.
Callan Periodic Table is a registered trademark of Callan LLC. Returns data shown are approximations for illustrative purposes, sourced from Morningstar direct indices (Russell 1000 Growth/Value for LC, Russell 2000 Growth/Value for SC, MSCI EAFE for Intl Dev, MSCI EM for EM, FTSE NAREIT for REIT, Bloomberg US Corp High Yield, Bloomberg US Agg for Core Bond, 3-mo T-Bill). Past performance is not indicative of future results.