Asset Allocation Explained — Portfolio Building

Photo: Yan Krukau / Pexels
Asset allocation is the decision of how to divide an investment portfolio among broad categories such as stocks, bonds, and cash. It is usually described as the most important choice in portfolio construction, ahead of which individual stock or fund you pick, but that claim has been repeated so often that its meaning has drifted. This article covers what the research on asset allocation actually measured, how the main model portfolios are built, what happened to them in bad years, and how rebalancing keeps a mix on target.
Educational only, not financial advice. Investing involves risk of loss, and past performance does not guarantee future results. Nothing here is a recommendation to buy or sell anything. A licensed financial professional can account for your tax situation, goals, and circumstances.
What asset allocation means, and what it does not
The U.S. Securities and Exchange Commission describes asset allocation as dividing an investment portfolio among different asset categories, such as stocks, bonds, and cash. The reasoning is that these categories often behave differently under the same market conditions, so a mix can reduce the chance of a large loss compared with holding only one. The SEC also stresses that the right mix is personal, shaped mainly by two things: time horizon, meaning how long the money will be invested, and risk tolerance, meaning how much loss an investor can absorb in pursuit of higher returns.
Asset allocation is easy to confuse with diversification, but they answer different questions. Allocation is the top-level split: how much goes to stocks, how much to bonds, how much to cash. Diversification is the spreading of money within and across those categories, so that no single company, bond issuer, or country decides the outcome. The SEC makes the same point, noting that diversification should happen both between and within asset categories and that an allocation by itself does not guarantee a diversified portfolio.
A cooking analogy helps. The allocation is the recipe: how much flour, how much sugar, how much butter. Choosing a particular stock or fund is closer to choosing a brand of flour. A cake with the wrong proportions will not be rescued by premium ingredients, while a well-proportioned cake made with ordinary flour will mostly turn out fine. The analogy has limits, since investment returns involve luck and no recipe guarantees a result, but it captures why researchers keep returning to the proportions.
The main asset classes
| Asset class | What it is | Role in a portfolio | Main risks |
|---|---|---|---|
| Stocks (equities) | Ownership shares in companies | Growth; highest long-run returns among major classes, with the largest swings | Sharp drawdowns, long recoveries |
| Bonds (fixed income) | Loans to governments or companies that pay interest | Income and, usually, steadier behavior than stocks | Interest-rate risk, inflation, credit risk |
| Cash and equivalents | Treasury bills, money market funds, savings | Stability and liquidity | Low returns, may trail inflation |
| Real estate | Property or real estate investment trusts (REITs) | Income and a different return pattern from stocks | Illiquidity (direct), interest-rate sensitivity |
| Commodities and gold | Physical or derivative exposure | Possible diversifier | Large swings, no income |
| Other alternatives | Private equity, hedge funds, crypto | Diversification attempts, mostly for institutions | Fees, illiquidity, valuation uncertainty |
For how bonds behave when rates move, see how bond prices move with interest rates and the broader primer on bonds and fixed income.
Why asset allocation gets so much credit: the Brinson studies
In 1986, Gary Brinson, Randolph Hood, and Gilbert Beebower published "Determinants of Portfolio Performance" in the Financial Analysts Journal. As later summarized, they examined quarterly returns for 91 large U.S. pension funds from 1974 to 1983 and compared each fund with a hypothetical index fund holding the same average mix of asset classes. A regression produced an average R-squared of 93.6 percent, which was widely repeated as "asset allocation explains 93.6 percent of returns." A 1991 update by Brinson, Singer, and Beebower reported a similar figure of 91.5 percent.
The number is real, and the common reading of it is wrong. An R-squared of 93.6 percent says that 93.6 percent of the variation of a fund's quarterly returns over time tracked the variation of its policy portfolio. It does not say that 93.6 percent of a fund's dollar return came from its allocation, and it does not say that the same share applies to every investor. Brinson himself objected to such readings, saying the figure was an average across the funds studied and not a number that applies to every single portfolio.
Three different questions, three different answers
In 2000, Roger Ibbotson and Paul Kaplan published a paper whose title is its thesis: "Does Asset Allocation Policy Explain 40, 90, or 100 Percent of Performance?" Using balanced mutual fund and pension fund data, they showed that the answer depends on which question is asked.
| Question | Answer found | What it means |
|---|---|---|
| How much of a single fund's return variability over time does its policy explain? | About 90 percent | Month to month, a fund mostly rises and falls with its asset classes |
| How much of the difference in returns between funds does policy explain? | About 40 percent | Two funds with similar rises and falls can still differ, because of costs, security selection, and timing |
| How much of the average return level does policy explain? | About 100 percent | Across all funds on average, the return is roughly what the asset classes delivered |
This is why the claim "allocation determines 90 percent of your returns" circulates in marketing and not in the careful literature. The accurate version is narrower. Most of the ups and downs in a diversified portfolio come from how much is in stocks versus bonds versus cash. Whether one particular fund beats another is a separate matter, where costs and manager choices matter more than the 90 percent figure implies. The Vanguard research team makes a similar point in its work on rebalancing, stating that asset allocation is the major determinant of a portfolio's risk-and-return characteristics, assuming a well-diversified portfolio that engages in limited market timing.
Source: Ibbotson, R. G., & Kaplan, P. D. (2000). Does asset allocation policy explain 40, 90, or 100 percent of performance? Financial Analysts Journal, 56(1). CFA Institute record
What the major asset classes have actually delivered
Historical data give a feel for the differences between classes, with a caveat repeated by every regulator: past returns do not predict future ones. The table below uses the annual return series maintained by Aswath Damodaran at New York University's Stern School of Business, covering 1928 through 2025. The averages are geometric (compound) annual returns that we computed from the ending values of $100 invested at the start of 1928, not figures published by Damodaran.
| Asset (Damodaran series) | Value of $100 at end of 2025 | Compound annual return, 1928-2025 |
|---|---|---|
| S&P 500 including dividends | $1,157,599 | About 10.0% |
| Baa corporate bonds | $53,952 | About 6.6% |
| Gold | $21,025 | About 5.6% |
| 10-year U.S. Treasury bond | $7,753 | About 4.5% |
| Real estate (as defined in the series) | $5,626 | About 4.2% |
| 3-month Treasury bill | $2,578 | About 3.4% |
These are nominal returns, before inflation, taxes, and fees. The gap between the first and last rows is the reward for bearing stock-market risk, and it is large. The $100 invested in stocks grew to roughly 450 times the value of the $100 held in Treasury bills. It is also where the reading of "long-run" needs care: that result includes the Great Depression, World War II, and several lost decades along the way, which investors had to sit through in real time.
The same series shows how rough the ride can be. In 1931 the S&P 500 returned -43.84 percent, in 2008 it returned -36.55 percent, and in 2022 it returned -18.04 percent. A portfolio that holds only stocks is a different experience from one that holds a mix, which is the core argument for allocating at all.
Source: Damodaran, A. Historical returns on stocks, bonds, bills, real estate, and gold, 1928 to 2025. NYU Stern data file
What a mix does in a bad year
An asset allocation shows its character in the years when markets fall. Using the same Damodaran annual returns for the S&P 500 and the 10-year Treasury bond, we calculated the one-year result of four stock and bond mixes, assuming the weights were reset at the start of each year. These are arithmetic illustrations, not funds anyone could have bought at those exact returns.
| Calendar year | 100% stocks | 80/20 | 60/40 | 40/60 | 10-year Treasury only |
|---|---|---|---|---|---|
| 2008 | -36.6% | -25.2% | -13.9% | -2.6% | +20.1% |
| 2022 | -18.0% | -18.0% | -18.0% | -17.9% | -17.8% |
| 2023 | +26.1% | +21.6% | +17.2% | +12.8% | +3.9% |
| 2024 | +24.9% | +19.6% | +14.3% | +9.0% | -1.6% |
Two stories sit inside this table. The year 2008 is the textbook case for holding bonds: Treasuries rose 20.1 percent while stocks lost more than a third, so each step toward bonds softened the blow. The year 2022 is the exception that every allocator now has to explain: stocks and long Treasuries fell together, so the mix offered almost no cushion in this version of the calculation.
Why 2022 broke the usual pattern
Bonds usually cushion stock losses when the shock is a recession, because investors flee to safety and central banks cut interest rates, which lifts bond prices. In 2022 the shock was inflation. Rates rose quickly, and rising rates push existing bond prices down, which is the mechanical relationship between bond prices and yields. Stocks fell at the same time, so the diversifier failed when it was most needed.
The broad-bond version of the story was a little less severe than our Treasury-only calculation. According to InvestmentNews, a 60/40 proxy (the Vanguard Wellington Fund) fell 14.3 percent in 2022. Using ETFs as proxies, the stock portion (SPY) lost 18.2 percent and the bond portion (the iShares Core U.S. Aggregate Bond ETF) lost just over 13 percent. The article called it the worst year for 60/40 portfolios since 2008 and cited Morningstar's Ben Johnson ranking it the sixth-worst year ever for the model. Both figures matter: the exact loss depends on which bond index stands in for "bonds."
Think of the portfolio as a car with two kinds of safety equipment. Most crashes are handled by the airbag, and bonds usually play the airbag. In 2022 the airbag was not the right part for the crash. That does not make the airbag useless in general, but it explains why some investors add inflation-linked bonds, short-term bonds, or other assets to a pure stock-and-bond mix.
Source: InvestmentNews (January 5, 2023), After rough ride in '22, 60/40 portfolio looks poised for solid '23. Calculations by SciHub101 from Damodaran's annual returns.
Classic model portfolios
A model portfolio is a published recipe. None of them is a recommendation, and each trades one kind of risk for another.
The 60/40
Sixty percent stocks and 40 percent bonds is the traditional "balanced" portfolio and the benchmark against which many balanced funds are compared. Its appeal is that stocks drive long-run growth while bonds dampen the swings. In the Damodaran data above, a 60/40 mix lost 13.9 percent in 2008 versus 36.6 percent for stocks alone, and it participated in most of the recovery that followed. Its weakness is the one exposed in 2022: it assumes that stocks and bonds will not both suffer at once.
The three-fund portfolio
The three-fund portfolio, associated with the Bogleheads community and the late Vanguard founder John Bogle's indexing philosophy, uses three total-market index funds: a U.S. stock fund, an international stock fund, and a U.S. taxable investment-grade bond fund. The investor picks the stock-bond split and the U.S.-international split, then rebalances. Its selling points are low cost, broad diversification across thousands of securities, and simplicity. It is the usual starting point for the do-it-yourself investor who wants a mix built from index funds or ETFs.
Target-date funds
A target-date fund is a single fund that automatically shifts from stocks toward bonds as a chosen retirement year approaches. The shift is called a glide path. Vanguard's published description of its Target Retirement series says the glide path begins at 90 percent stocks and 10 percent bonds and reaches its most conservative point, 30 percent stocks and 70 percent bonds, seven years after retirement. Costs vary by provider. In a January 10, 2022 release, Vanguard said a new 2070 fund would charge 0.08 percent a year, which it described as one-fifth the cost of the average target-date fund, based on Morningstar data as of December 31, 2020.
The trade-off is customization. A target-date fund applies one glide path to everyone born in a given period, regardless of other assets, a pension, or a spouse's income. For a person who wants no ongoing decisions, that is the feature. For a person with unusual circumstances, it is the limitation.
Age-based rules of thumb
A common heuristic subtracts age from 100, or from 110 or 120 in more recent versions, to estimate a stock percentage: at 110, a 30-year-old would hold 80 percent stocks and a 65-year-old would hold 45 percent. These formulas are shortcuts without a theoretical basis, and they ignore everything except age. They are best viewed as a way to put a starting number on the page before thinking about the other factors discussed below.
The endowment model
David Swensen, the longtime chief investment officer of Yale's endowment, championed an approach that holds heavy allocations to private equity, venture capital, real assets, and other less liquid investments, with less in plain stocks and bonds. Yale reported a fiscal 2024 return of 5.7 percent, net of fees, for the year ended June 30, 2024, which left the endowment at $41.4 billion. Its chief investment officer said that, given the significant allocation to private assets, Yale expects to lag during strong public-market rallies. Over longer periods Yale reported 9.5 percent a year over 10 years and 10.3 percent over 20 years, 3.8 percentage points per year above a 70/30 stock-and-bond portfolio in both periods.
Those results come from an institution with permanent capital, access to elite managers, and a tolerance for illiquidity. Individual investors generally lack such access and face higher fees, which is why the model is usually described as hard to copy. Yale's own press release lists the asset classes in its strategy but does not give weights, so the allocation percentages in popular summaries should be checked against Yale's annual report.
Risk parity
Risk parity allocates by risk contribution rather than by dollars. Because stocks swing far more than bonds, a risk-balanced portfolio holds more bonds than a 60/40 mix does, often using leverage so that the bond holdings contribute meaningfully to returns. Bridgewater Associates' All Weather strategy, which the firm says was fully formed in 1996 for founder Ray Dalio's family trust assets, is built around four economic environments: rising and falling inflation, and rising and falling growth, each relative to expectations. Bridgewater describes leverage as an implementation tool for equalizing risk. The approach is sensitive to what happens when stocks and bonds fall together, which makes 2022 its stress test.
Lifecycle investing
Economists Ian Ayres and Barry Nalebuff argued in their 2010 book Lifecycle Investing that young savers hold too little in stocks relative to their lifetime wealth, because their future savings are like a large bond they cannot sell. Their proposal was to borrow, up to a 2-to-1 ratio, to hold more stock early in life and then reduce the leverage over time. Reviewers noted that the idea drew skepticism, especially after the 2008 crisis, because leverage magnifies losses and brings margin-call risk.
Comparing the models
| Model | Core idea | Main assumption | Main weak point | Typically suits |
|---|---|---|---|---|
| 60/40 | Balance growth and stability | Bonds cushion stock drops | Stocks and bonds falling together, as in 2022 | Moderate-risk, mid-career savers |
| Three-fund | Own the whole market cheaply | Costs and discipline beat prediction | Requires the investor to rebalance and stay put | Do-it-yourself index investors |
| Target-date | Automate the shift to bonds | One glide path fits a birth cohort | Ignores personal circumstances | Hands-off savers in workplace plans |
| Endowment | Favor private, illiquid assets | Illiquidity is rewarded and managers can be selected | Fees, illiquidity, access | Institutions with permanent capital |
| Risk parity | Balance risk, not dollars | Leverage on bonds is manageable | Rate shocks and correlated sell-offs | Sophisticated investors |
| Lifecycle | Spread stock exposure across a lifetime | Future savings act like a bond | Leverage and behavior in crashes | A small group comfortable with leverage |
How people actually choose an allocation
Both the SEC and the research community converge on the same few inputs. None of them has a single right answer, and different financial professionals weigh them differently.
Time horizon. The SEC defines it as the expected number of months, years, or decades an investor will be investing, and notes that longer horizons generally allow more room for volatile investments. Money needed in two years has less time to recover from a stock decline than money needed in 30.
Risk tolerance. The SEC describes this as the ability and willingness to lose some or all of the original investment in exchange for greater potential returns. The two words are different. Ability is a matter of finances: a stable income and other savings make a drawdown survivable. Willingness is psychological: whether an investor will stay invested through a drop or sell near the bottom. Survey answers about willingness tend to be generous in calm markets, which is why many advisers ask how a person behaved in the last major decline.
Capital needs. A saver who adds money every month can buy during a downturn. A retiree withdrawing from the portfolio is selling regardless of price, which is why sequence-of-returns risk matters for retirement. For more on that, see the 4 percent rule and the broader guide to planning for retirement.
A worked example of the risk side
Take the 2008 column of the table above. An investor who held $100,000 in an all-stock mix would have seen it fall to about $63,500 by the calendar-year return in the Damodaran data. An 80/20 mix would have fallen to about $74,800, a 60/40 mix to about $86,100, and a 40/60 mix to about $97,400. Each investor was "right" for a different tolerance. The costs appear on the other side: in 2023, the same four mixes would have gained 26.1, 21.6, 17.2, and 12.8 percent, respectively. An allocation is a decision about which of those two columns hurts more to miss.
The lesson is not that the middle option is correct. It is that the question to ask before choosing an allocation is how large a loss a person could live with, and for how long, not what return they hope for.
Rebalancing: keeping the recipe from drifting
Markets change the proportions on their own. A 60/40 portfolio in which stocks rise 30 percent while bonds are flat ends the year with a stock share of about 66 percent, a riskier mix than the one chosen. Rebalancing means returning to the target weights by selling what has grown and buying what has lagged, or by directing new contributions to the underweight category. The full process is covered in portfolio rebalancing explained.
Vanguard's 2010 research paper on the subject reached a conclusion that surprises people. It found no optimal frequency or threshold, and showed that risk-adjusted returns are not meaningfully different whether a portfolio is rebalanced monthly, quarterly, or annually. What does differ is the cost. Monthly rebalancing with no threshold would require 1,008 rebalancing events in the period studied, while annual rebalancing with a 10 percent threshold would require only 15. The authors concluded that, for most broadly diversified stock and bond portfolios, annual or semiannual monitoring with rebalancing at 5 percent thresholds is likely to give a reasonable balance between risk control and cost, and that annual rebalancing is likely preferable when taxes or substantial time and costs are involved.
The purpose matters too. The Vanguard authors state that the primary goal of rebalancing is to minimize risk relative to a target allocation, not to maximize returns. A rebalanced portfolio stays closer to the risk level its owner chose, and an unrebalanced portfolio drifts toward whichever asset has done best, which is usually the riskier one. The SEC adds that rebalancing can trigger fees or taxes and that it tends to work best when done relatively infrequently.
Source: Jaconetti, C. M., Kinniry, F. M., & Zilbering, Y. (2010). Best practices for portfolio rebalancing. Vanguard Research. PDF · U.S. SEC, asset allocation overview
Common mistakes and misunderstandings
Treating "90 percent" as a fact about your own portfolio. As the Ibbotson and Kaplan paper shows, the number answers a question about the time series of returns for the average fund. It does not mean that choosing between two funds is irrelevant.
Chasing last year's winner. Gold, for example, returned 66.22 percent in 2025 in the Damodaran series, while its long-run compound return from 1928 was about 5.6 percent. Moving into an asset class after a year like that means buying after the move. Dollar-cost averaging is one way some investors reduce the temptation to time such moves.
Owning many funds that hold the same thing. Several large-cap funds that track overlapping indexes add names to the statement without adding diversification.
Confusing a bond fund with a savings account. Bond funds can fall when rates rise, as 2022 demonstrated. Cash equivalents and short-term Treasury bills behave differently, with a 2022 return of 2.09 percent for 3-month bills in the Damodaran data.
Neglecting costs and taxes. Fees compound against the investor in the same way returns compound for them, and selling in a taxable account can create a tax bill. Account type matters too: the choice between account types is covered in Roth IRA vs 401(k).
Equating a complex allocation with a better one. Adding alternatives increases fees and illiquidity. The endowment model works for institutions with specific advantages; it is not automatically better for an individual.
Holding only the home market. Many investors hold far more of their own country's stocks than its share of the global market would imply. Home-market concentration can work for years and then fail, and the choice between style tilts such as value and growth has its own cycles. The point is not that international holdings must be large, but that the choice should be deliberate.
Asset allocation and long-horizon goals
For goals many decades away, such as early retirement under the FIRE movement, the allocation question stretches over a long accumulation phase and an equally long withdrawal phase. That mix of long horizons and large withdrawals is why researchers study glide paths, and why the stock-versus-bond split can change what a plan survives. The same principles apply at any scale: the allocation should match the horizon, the capacity for loss, and the willingness to stay put.
FAQ
Q: What is asset allocation in simple terms? A: It is how a portfolio is divided among broad categories such as stocks, bonds, and cash. The SEC describes it as dividing a portfolio among different asset categories because they often behave differently under the same market conditions, and it says the right mix depends mainly on time horizon and risk tolerance.
Q: Does asset allocation really explain 90 percent of returns? A: Not in the way the slogan implies. Ibbotson and Kaplan found that policy explains about 90 percent of a typical fund's return variability over time, about 40 percent of differences in returns between funds, and about 100 percent of the average return level. The 93.6 percent figure from the 1986 Brinson study referred to variation in quarterly returns, not dollars earned.
Q: Is the 60/40 portfolio still a reasonable benchmark? A: It remains a widely used reference mix. Its weakness appeared in 2022, when stocks and bonds fell together: InvestmentNews reported a 14.3 percent decline for a 60/40 proxy, the worst year for the model since 2008. Whether that was a one-off or a new regime is debated, and no allocation removes risk.
Q: How often should a portfolio be rebalanced? A: Vanguard's research found no meaningful difference in risk-adjusted returns between monthly, quarterly, and annual rebalancing, and concluded that annual or semiannual monitoring with rebalancing at 5 percent thresholds is likely to balance risk control against cost. Taxes and trading costs are the main reasons to avoid rebalancing too often.
Q: What is the difference between asset allocation and diversification? A: Allocation is the top-level split among categories, such as how much is in stocks versus bonds. Diversification is spreading money within and across those categories so that no single holding dominates. Both are needed, and the SEC notes that an allocation alone does not guarantee diversification.
Q: What is a glide path? A: A glide path is the schedule by which a target-date fund reduces its stock share over time. In Vanguard's Target Retirement series, the glide path starts at 90 percent stocks and reaches 30 percent stocks seven years after retirement. Other providers use different shapes.
Q: Can an individual copy the Yale endowment model? A: Not easily. Yale's returns come from large private-asset allocations, manager access, and the ability to hold illiquid investments for long periods. Yale itself notes that such a portfolio is likely to lag in strong public-market years, and individual investors typically cannot get the same access or fee levels.
The takeaway
The most useful thing the asset allocation research offers is a corrected question. Ibbotson and Kaplan showed that "how much does allocation matter?" has three answers, and the one that holds for a real investor is the first: most of a diversified portfolio's ups and downs come from how it is divided among stocks, bonds, and cash. The Damodaran data show how different those ups and downs can be, from a 2008 in which a 40/60 mix lost about 2.6 percent while stocks lost 36.6 percent, to a 2022 in which nearly every mix fell about 18 percent. Models such as the 60/40, the three-fund portfolio, and target-date funds are different answers to the same underlying trade-off. The practical work is deciding how much loss is tolerable, then keeping the mix from drifting.
Sources
- U.S. Securities and Exchange Commission. Asset allocation overview. sec.gov
- Ibbotson, R. G., & Kaplan, P. D. (2000). Financial Analysts Journal, 56(1). CFA Institute
- Damodaran, A. Historical returns, 1928-2025. NYU Stern
- Jaconetti, Kinniry, & Zilbering (2010). Vanguard, Best practices for portfolio rebalancing. PDF
- Vanguard (January 10, 2022). Press release on the Target Retirement 2070 fund
- Yale University (October 25, 2024). Yale reports investment return for fiscal 2024
- Bridgewater Associates. The All Weather Story
- InvestmentNews (January 5, 2023). After rough ride in '22, 60/40 portfolio looks poised for solid '23
- Brinson, G., Hood, R., & Beebower, G. (1986). Determinants of portfolio performance. Financial Analysts Journal.
- Ayres, I., & Nalebuff, B. (2010). Lifecycle Investing. Basic Books.
See also: Stocks vs bonds: investing basics
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

Value vs Growth Investing — Which Style Wins?
Value vs growth investing compared: definitions, the Fama-French value premium, why value lagged before 2021, 2022's reversal, value traps, and style tilts.

Short Selling Explained — Betting Against Stocks
How short selling actually works: borrowing shares, unlimited losses, short squeezes, and the role of short sellers as fraud detectives.

Stocks vs Bonds — Investing Basics Explained
Stocks vs bonds explained — risk, return, diversification, index funds, and what a century of market data say about long-term investing.

Portfolio Rebalancing — When and How to Rebalance
Rebalancing keeps your portfolio at its target risk. How often to rebalance, threshold vs. calendar methods, tax tricks, and what the research says.

How the Stock Market Works: A Complete Guide
Learn how stocks, exchanges, IPOs, valuation, and market indices work — from bid-ask spreads to the efficient market hypothesis.

Why Stock Prices Move After Earnings
Stocks move after earnings because expectations matter more than headlines. Revenue, margins, guidance, cash flow, and narrative all interact.