Why Allocation Dominates Returns
Why Allocation Dominates Returns
The Brinson–Hood–Beebower analysis of institutional portfolios found that asset allocation accounts for 93.6% of return variance, while security selection contributes only 4.6%. This finding has been replicated across decades and markets, proving that your asset class mix matters far more than which specific funds you pick.
Key takeaways
- Brinson et al. found that allocation explains 93.6% of return variance; security selection explains only 4.6%
- This result has held across different time periods, market conditions, and institutional vs. retail investors
- A mediocre fund in the right allocation outperforms a great fund in the wrong allocation
- Market timing—trying to guess when to shift allocations—is even harder than security selection
- Implications: spend 80% of your effort on allocation, 20% on fund selection, and almost no time on timing
The Brinson Study: A Brief History
In 1985, Gary Brinson and Ned Beebower—and later with Lawrence Hood—analyzed the quarterly returns of 91 large U.S. pension funds over the period 1974–1983. They decomposed each fund's return into three parts: allocation effect (the impact of over/underweighting asset classes), selection effect (the impact of picking better or worse securities within each class), and timing effect (the impact of shifting allocations in and out).
The results shocked many investment professionals. The researchers found that 93.6% of the variance in returns between the funds was explained by their asset allocation decisions. Only 4.6% came from security selection—picking the best stocks or bonds within each class. The remaining 1.8% came from timing.
Put differently: if you took any two funds and predicted which would outperform based solely on their allocations, you would be right over 93% of the time. Knowing which specific securities they held added almost nothing to your prediction power.
The study's implications were radical. It suggested that professional investors—people with teams of analysts and real-time market data—were wasting enormous resources on stock picking when they could generate far better results by simply choosing the right asset allocation and holding it.
Why This Result Holds
The Brinson finding isn't a fluke. It has been replicated in hundreds of studies since 1985, across different asset classes, time periods, and investor types. In 1991, Ibbotson and Kaplan extended the analysis to include rebalancing and found similar results: allocation accounts for roughly 90% of variance. Morningstar later ran the analysis on mutual funds and found comparable numbers.
The reason is mathematical. Asset classes have very different return profiles. From 1975 to 2023, U.S. stocks returned roughly 10% annually, while U.S. bonds returned roughly 5.5%. Over 20 years, this compounds to a massive difference. A portfolio that is 80% stocks and 20% bonds will almost certainly outperform a portfolio that is 20% stocks and 80% bonds, regardless of how well either portfolio's securities are chosen.
Within each asset class, however, the best security pickers do only marginally better than random choice, after costs. In a study covering 2,076 actively managed equity funds over the period 2004–2017, only about 6% of funds that beat the market in one decade beat it in the next. For bonds, the picture is even starker: after fees, 96% of active bond funds underperform a simple bond index.
This is not because analysts are incompetent. It is because markets are competitive. Thousands of bright people with data and computers are trying to outpick each other. On average, they cannot; they just trade with each other, paying fees in the process. The returns to security selection, after costs, are near zero.
The returns to good allocation, by contrast, are enormous. An 80/20 portfolio simply will not match a 20/80 portfolio over a 20-year period. The math is in the former's favor. And the data shows that despite decades of evidence, many investors still allocate their money incorrectly for their goals and time horizons.
Allocation vs. Selection: Two Case Studies
Consider two hypothetical investors in January 2018.
Alice is a security-selection devotee. She spends 20 hours a week researching stocks and manages her own portfolio of 15 individual companies. She has indeed beaten the S&P 500 by 2% annually for the past five years. However, she feels she needs equity upside, so she invests 100% of her portfolio in stocks.
Bob is an allocation-focused investor. He has invested zero hours in stock picking. He simply bought a low-cost U.S. total-market index fund and a bond index fund in a 60/40 split. His expected return is probably 1–2% lower than Alice's if she can keep beating the market, but he has not tried to.
In 2020, COVID-19 caused equities to fall 30%. Alice's stock portfolio fell 30%; her portfolio dropped from $500,000 to $350,000. Bob's 60/40 portfolio fell about 18%; his dropped from $500,000 to $410,000. The shock caused Alice to panic and sell at the bottom. Bob, comfortable in the knowledge that a 18% decline was within his expectations, held firm. When markets recovered by June 2020, Alice had locked in her loss. Bob recovered by August.
By 2023, both portfolios had roughly doubled their 2020 lows. But Alice's was still worth less than Bob's because she had sold at the wrong time. The moral: Alice's superior selection skill (a 2% annual edge) was wiped out by her misaligned allocation (100% stocks instead of 60%).
Now consider another case. Charles is a terrible stock picker. He has underperformed the S&P 500 by 3% annually. But he allocates thoughtfully: at age 45 with a 20-year horizon and solid income, he invests 70% in a broad stock index and 30% in a bond index. This allocation is well-suited to his goals.
Daniel is a mediocre stock picker (roughly market-matching) but has an allocation problem. Desperate for returns to fund a lifestyle he cannot afford, he invests 95% in stocks and 5% in bonds. Despite Daniel's better selection skill (market matching vs. underperformance), Charles's portfolio will almost certainly outperform his over 20 years. The allocation difference will overwhelm the selection difference.
The Role of Rebalancing
One reason allocation is so powerful is rebalancing. Imagine you start with 70/30 and equities soar. After a strong year, your allocation drifts to 75/25. If you rebalance back to 70/30, you are selling high and buying low—exactly what disciplined investing requires. This simple rule, applied consistently, adds 0.3–0.5% annually to returns, depending on volatility and rebalancing frequency.
Rebalancing is nearly impossible for active security pickers. If your edge comes from picking the right stocks, you cannot afford to sell your best performers just to maintain an allocation. But for index investors, rebalancing is painless and powerful.
What About Market Timing?
The Brinson study found that timing—trying to shift allocations in and out of the market based on forecasts—explains only 1.8% of return variance. But there is a twist: most timing is negative. Funds that tried to time the market actually reduced their returns.
This is intuitive. Timing requires you to be right twice: when to get out and when to get back in. You might correctly predict a downturn and reduce equities from 70% to 50%. But if you're six months early, you'll miss the 15% rally before the crash. And if you're timing the recovery, you might be 3% too late, catching 97% of the rebound but missing the most violent part.
The historical record is brutal for timers. From 2009 to 2023, the S&P 500 returned roughly 12% annually. But if you missed just the 10 best days—days that were scattered, unannounced, and typically during recoveries—your returns fell to 7%. And the best 10 days usually come alongside the worst days. Avoiding bad markets means avoiding good ones too.
Implications for Your Portfolio
The Brinson finding should simplify your life. It suggests a clear priority order:
-
Get your allocation right. Spend weeks on this if needed. Interview a financial advisor, use a questionnaire, or work through the frameworks in this book. Get to a clear allocation matched to your goals and risk tolerance.
-
Choose low-cost index funds for each asset class. Spend hours on this, not days. Buy a broad U.S. stock index (VTI, SPLG, SWPPX), a bond index (BND, SPLB, SWAGX), and maybe an international equity index (VXUS, SWISX). The difference between the 15th-best fund and the 1st-best is less than 0.1% in fees; the difference between the wrong allocation and the right one is 1–2% in returns.
-
Rebalance once or twice a year. Set a calendar reminder for January and July. If your stock allocation has drifted more than 5 percentage points, trim back and buy bonds. This takes 30 minutes and will slightly improve your returns.
-
Ignore market timing. Do not try to predict recessions. Do not shift allocations based on valuation metrics or Fed policy. The odds that your timing improves returns are less than 10%.
How it flows
Next
Now that you understand why allocation dominates returns, the question becomes: which specific allocation should you choose? The answer depends on a single fundamental trade-off: stocks offer higher long-term returns but larger short-term volatility, while bonds offer stability but lower long-term growth. The next article explores this trade-off in detail.