Asset allocation by age — why "120 minus age" misses what matters

The most-cited rule of thumb for retirement asset allocation in mainstream US personal finance content is some version of “your equity allocation should be 120 minus your age, with the rest in bonds”. A 30-year-old at 120 minus age holds 90% equity and 10% bonds; a 60-year-old holds 60% equity and 40% bonds; a 75-year-old holds 45% equity and 55% bonds. The rule has the virtue of producing a defensible number quickly, and the glidepath the rule implies is broadly consistent with what major target-date retirement funds actually implement. The rule also misses, by design, the three factors that matter more than age for most households: the stability and floor of the household’s income (including defined-benefit pensions and Social Security), the time horizon to the money’s actual use (which is often much longer than retirement age suggests), and the household’s behavioral capacity to hold equity through a market crash.

This guide walks through what the age-based rules actually say, the three factors that explain why two 60-year-olds can defensibly hold very different allocations, the role of bonds in a portfolio (which is largely about behavior rather than long-term return), and a practical framework for thinking about allocation that produces a more defensible answer than any single number rule. The framework is intentionally simple — six questions, an estimate of each — and is designed to be done at the kitchen table without specialized software.

What the age-based rules actually say

The “100 minus age” rule is the original version, dating to a period when bond yields were higher than they have been at any point in the last fifteen years and life expectancies were shorter. The “110 minus age” rule and the “120 minus age” rule are more recent adjustments to account for longer life spans, lower bond yields, and the broader recognition that equity outperforms bonds over multi-decade windows by enough that a heavier equity tilt is defensible for most investors.

The rules produce numbers in the same general territory as the target-date retirement funds’ glidepaths. A Vanguard Target Retirement 2055 fund, intended for an investor retiring around 2055 (currently age 35), holds approximately 90% equity. At age 50 (target year 2040), the fund is at approximately 75% equity. At age 65 (target year 2025), it is at approximately 50% equity, continuing to glidepath down through 2035 before stabilizing.

The rules and the fund glidepaths converge because they share the same underlying assumption: the appropriate equity allocation for a typical household is primarily a function of time-to-retirement, with adjustments for age-related risk capacity. The shared assumption is broadly correct as a generic baseline. The problem is that the generic baseline assumes a typical household, and very few specific households are typical.

Asset allocation by age — the rules side by side

Before getting to why a single number is the wrong place to stop, it helps to see what the common rules actually produce across a working life. The table below is a reference baseline, not a prescription — the rest of this guide is about the adjustments that matter more than the row you land on. Each cell shows the equity-to-bond split the rule implies at that age.

Age	100 − age (equity / bonds)	110 − age (equity / bonds)	120 − age (equity / bonds)
25	75 / 25	85 / 15	95 / 5
30	70 / 30	80 / 20	90 / 10
35	65 / 35	75 / 25	85 / 15
40	60 / 40	70 / 30	80 / 20
45	55 / 45	65 / 35	75 / 25
50	50 / 50	60 / 40	70 / 30
55	45 / 55	55 / 45	65 / 35
60	40 / 60	50 / 50	60 / 40
65	35 / 65	45 / 55	55 / 45
70	30 / 70	40 / 60	50 / 50

A fourth heuristic, “age in bonds,” says to hold a bond percentage equal to your age — a 40-year-old holds 40% bonds and 60% equity. It is mathematically identical to the “100 minus age” column above and is the most conservative of the common rules; “120 minus age” is the most aggressive and tracks most closely to the glidepaths the major target-date funds actually run, described in the section above. The same equity-and-bond split applies inside a 401(k) — the account wrapper does not change the target you are aiming for — with one wrinkle worth carrying down to the tax-location section near the end of this guide: the least tax-efficient assets, usually the bonds, are the ones best held inside the 401(k) rather than in a taxable account.

Every row in that table, though, assumes the household is average. The next section is about the three things that make a specific household’s right answer diverge from the row it happens to land on.

The three factors that matter more than age

For most households, three factors matter more than age in determining the appropriate equity allocation. The age-based rules ignore all three.

Income floor stability and pension presence. A household with a defined-benefit pension that covers most of retirement spending needs has a substantially different risk capacity than a household entirely reliant on defined-contribution accounts and Social Security. The pension functions as the bond allocation of the household balance sheet — a guaranteed income stream not subject to market volatility — and frees the household’s investment portfolio to be more aggressively allocated to equity without the household’s overall risk profile being out of balance.

A 65-year-old retiree with a $60,000 annual pension from a prior employer, plus full Social Security benefits of $36,000, has $96,000 of guaranteed annual income that fully covers their lifestyle needs. The investment portfolio of $400,000 in retirement accounts is functionally a discretionary fund — used for travel, gifts to grandchildren, possible long-term care costs — not the primary support for living expenses. That household can defensibly hold 80% equity in the discretionary fund without taking on inappropriate risk, because the household’s lifestyle is not exposed to market volatility. The age-based rule would put them at 55% equity / 45% bonds, which is more conservative than their actual situation warrants.

A 65-year-old retiree with no pension, Social Security of $30,000, and $700,000 in retirement accounts that need to support all spending above Social Security is in a very different situation. The portfolio’s volatility translates directly into spending volatility for this household, and the appropriate equity allocation is likely below what the age-based rule suggests, perhaps 40% equity / 60% bonds, to reduce the sequence-of-returns risk in the first ten years of retirement.

Same age, materially different appropriate allocation, because the income floor structure is different.

Time horizon to actual use of the money. The age-based rules implicitly assume the money will be drawn down through retirement to support living expenses, with the drawdown starting at age 65 or so. For households where this assumption holds, the rules produce reasonable allocations. For households where the assumption does not hold — typically high-income or high-asset households where Social Security plus a portion of the retirement portfolio is sufficient to cover lifestyle and the bulk of the portfolio is destined for inheritance or for a much later use — the appropriate allocation is more aggressive.

A 70-year-old with $3 million in retirement accounts and lifestyle needs of $80,000 a year above Social Security can spend the lifestyle portion of the portfolio (~$2 million calculated against a 4% withdrawal rate) and the remaining $1 million is essentially a multi-generational asset that will eventually pass to children or grandchildren. The time horizon on that $1 million portion is the children’s lifetime plus the grandchildren’s window of inheritance — easily 50 years or more. A 50-year horizon argues for a heavy equity allocation regardless of the household’s calendar age.

The Bogleheads community has codified this thinking as “asset allocation should reflect the time horizon of the money’s eventual use, not the holder’s age”. The two coincide for households drawing down to support themselves; they diverge for households whose terminal assets will outlive them.

Behavioral capacity to hold equity through a crash. The honest self-assessment most investors avoid is whether they have the behavioral capacity to hold their equity allocation through a 40% to 50% market drawdown. The historical evidence from the 2008-2009 and 2020 market crashes is that many investors do not. Investors who hold a 70% equity allocation in theory but who sell to cash when the portfolio is down 30% have a realized allocation lower than their stated one, and a realized return substantially worse than the same allocation held with discipline.

The behavioral capacity question is hard to answer truthfully in calm markets, when the prospect of a 40% drawdown is hypothetical. The closest available evidence is the investor’s actual behavior during the most recent crashes (2008-2009, 2020, the briefer 2022 bond/equity selloff). An investor who held position and continued to rebalance through those windows can defensibly maintain a relatively aggressive allocation. An investor who exited to cash or stopped contributions during one or more of those windows has revealed their behavioral capacity, and the right allocation for them is materially more conservative than their theoretical risk tolerance suggests — perhaps 20 to 30 percentage points lower equity than the age-based rule.

The role of bonds in a portfolio — what they are actually for

The mainstream framing of bonds is that they produce income and reduce portfolio volatility. Both are true. The deeper role of bonds in a long-horizon portfolio is behavioral: bonds are the allocation that prevents the investor from panic-selling equity at the bottom of a market crash by providing a stable counterweight to the volatility of the equity allocation.

The expected return on bonds over a multi-decade horizon is approximately 2 to 3 percentage points lower than the expected return on equity. Every percentage point of bonds in a portfolio is a percentage point of expected return given up, in exchange for a percentage point of volatility reduction. For a young investor with a thirty-plus year horizon, the math suggests an aggressive equity tilt is optimal in expected-value terms, because the volatility reduction is not needed (the investor is not drawing on the portfolio for decades) and the foregone return compounds substantially.

The reason young investors should still hold some bonds — perhaps 10% to 20% even at age 30 — is not the expected-value math but the behavioral protection. A young investor with 100% equity who watches the portfolio drop 50% in a crash and then sells is permanently damaged in retirement-asset terms. The same young investor with 90% equity and 10% bonds who watches the portfolio drop 45% has the small bond cushion to rebalance into the depressed equity (selling some bonds to buy equity at the bottom), which both reinforces the discipline and captures some of the recovery. The bond allocation is paid for by foregone return, but the cost is small relative to the avoided behavioral damage if it works.

For older investors approaching retirement and into retirement, the bond allocation serves the more conventional purpose of dampening the sequence-of-returns risk that hits hardest in the first decade of withdrawals. The 60% equity / 40% bond allocation that target-date funds settle into around the retirement year is the conventional answer to this question, and it is broadly defensible for households without a substantial pension floor.

A six-question practical framework

The framework: answer six questions, multiply, get an equity allocation that is more defensible than any single number rule. The framework is not academically rigorous but produces results consistent with academic research while being implementable at a kitchen table.

Question 1: how many years from retirement (or from when the money will actually be needed)? If under 5 years, start with a 40% equity allocation. If 5-15 years, start with 55%. If 15-25 years, start with 70%. If 25+ years, start with 85%.

Question 2: what fraction of retirement spending will be covered by Social Security plus any defined-benefit pension? If above 80%, add 10 percentage points to equity. If 50-80%, no adjustment. If below 50%, subtract 10 percentage points from equity. (Households with very high pension coverage can afford more equity risk; households without pension coverage need more bond cushion.)

Question 3: does the household have separate emergency savings (3-6 months of expenses) outside the retirement portfolio? If yes, no adjustment. If no, subtract 5 percentage points from equity. (Emergency savings outside the portfolio prevent the household from being forced to sell equity at a bad time.)

Question 4: how did the household behave during the most recent market crashes (2008-2009, 2020, 2022)? If they continued contributing and did not sell, no adjustment. If they reduced contributions but held, subtract 5 percentage points. If they sold or moved to cash, subtract 15 percentage points. (Behavior is the dominant variable; do not give the household credit they have not earned.)

Question 5: is the household’s expected post-retirement time horizon long (the money will fund 30+ years of retirement, or pass to heirs) or short (the money will be drawn down rapidly)? Long, add 5 percentage points. Short, subtract 5 percentage points.

Question 6: is the household’s income during the contribution years stable or volatile? Stable (salaried with low layoff risk), no adjustment. Volatile (self-employed, commission-heavy, founder), subtract 5 percentage points. (Volatile-income households need a larger cash and bond buffer to weather income disruption.)

A worked application: a 45-year-old salaried employee, retiring at 65 (20 years out), expected to have Social Security covering 40% of retirement spending and no pension, with 6 months of emergency savings, who held position through the 2020 crash, expects a 30-year retirement, and has stable income.

Starting allocation: 70% equity (15-25 years out). Pension/SS adjustment: -10 (below 50% coverage). Emergency savings adjustment: 0 (has emergency fund). Behavior adjustment: 0 (held position). Time horizon adjustment: +5 (long retirement). Income volatility adjustment: 0 (stable).

Final allocation: 65% equity / 35% bonds.

The same employee but with a $40,000/year defined-benefit pension that would cover most spending: starting 70%, pension adjustment +10, other adjustments same = 85% equity. The pension makes a 20 percentage point difference in defensible equity allocation, which the age-based rule completely misses.

Two common allocation errors to avoid

The first common error is the “I am young, I can afford 100% equity” position. For most young investors, the position is defensible in expected-value terms but ignores the behavioral protection bonds provide. A 5% to 15% bond allocation costs essentially nothing in expected return (perhaps 0.1 to 0.3 percentage points per year) and provides meaningful behavioral cushion. Pure 100% equity is the right answer only for investors with verified behavioral capacity (multiple cycles of holding position through 30%+ drawdowns) and no other risk concerns.

The second common error is the “I am near retirement, I should move to mostly bonds” position. The framing assumes the entire portfolio is needed to support retirement spending, which for most households entering retirement is over-conservative — Social Security and any pension cover a meaningful baseline, and the portfolio supports the discretionary or above-baseline spending. Moving entirely to bonds at retirement exposes the household to inflation risk and longevity risk (running out of money because the bonds cannot keep up with thirty years of inflation-adjusted spending), which are real and frequently underweighted relative to market volatility risk. The target-date funds versus three-fund portfolio comparison covers the implementation question once the allocation itself has been chosen.

The target-date fund glidepath, which settles around 60% equity / 40% bonds at the retirement year and 40% equity / 60% bonds a decade past retirement, is a defensible compromise for most households without strong reasons to deviate. The framework above is the way to identify those reasons and adjust accordingly.

Tax location — which asset goes in which account

The allocation question (what percent equity, what percent bonds) is logically separate from the tax location question (which account holds which asset class), but the two compound when both are done well. The same 70/30 portfolio can produce materially different after-tax returns depending on which asset class lives in the taxable brokerage versus the 401(k) versus the Roth IRA.

The general framework — sometimes called “asset location” to distinguish it from “asset allocation” — places the most tax-inefficient assets in tax-deferred accounts (where their drag does not matter) and the most tax-efficient assets in taxable brokerage (where the preferential treatment compounds tax-free).

Bonds and bond funds belong in tax-deferred accounts (401(k), traditional IRA). Bond interest is taxed at ordinary income rates — the same brackets that apply to W-2 wages, up to 37%. Holding bonds in taxable brokerage means paying full ordinary tax on the interest income every year. Holding the same bonds inside a 401(k) defers the tax until withdrawal, by which time the household is typically in a lower bracket. This single rearrangement can be worth 0.5 to 1 percentage point per year of after-tax return on the bond allocation.

Broad-market equity index funds belong in taxable brokerage. They produce minimal short-term gains (low turnover), most of their dividends are qualified (taxed at the preferential 0%/15%/20% rates explained in the capital gains tax guide), and their long-term gains qualify for the preferential rates plus the eventual step-up in basis at death. A broad-market index held in taxable for thirty years and passed to heirs can be the single most tax-efficient position in a household’s balance sheet, because the latent gain is forgiven entirely at the step-up.

REITs and other high-yield equity belong in tax-deferred or Roth accounts. REIT dividends are largely non-qualified (taxed at ordinary income rates) because of the REIT structural pass-through requirement. Holding REITs in taxable brokerage forfeits the qualified-dividend rate that other equity dividends enjoy. The same REITs in a 401(k) or IRA pay no current tax on the dividend.

Treasuries are best held in taxable brokerage in high-tax states, despite being bonds. The federal interest on US Treasury securities is exempt from state income tax — a meaningful benefit in California (13.3% state rate), New York (10.9%), Hawaii (11%), and similar high-tax jurisdictions. Holding Treasuries inside a 401(k) wastes the state-tax-exemption because the 401(k) wrapper already shelters the interest from federal tax, leaving no marginal benefit for the state-exemption layer.

Foreign developed-markets equity is taxable-friendly because of the foreign tax credit. Foreign withholding tax on foreign dividends can be claimed as a tax credit on Schedule 3 line 1 only when the foreign equity is held in a taxable account. The same foreign equity inside a 401(k) or IRA gets the foreign withholding deducted from the dividend with no offsetting credit available — a permanent loss of the foreign tax paid.

Roth accounts (Roth IRA, Roth 401(k)) belong to the asset class with the highest expected return — typically aggressive equity. Roth tax treatment converts all future growth to tax-free, so the household maximizes the Roth value by putting the highest-growth asset there. Most planners suggest small-cap equity, emerging markets equity, or aggressive sector tilts go in Roth before they go in traditional tax-deferred accounts.

A worked example. A household with 70/30 equity/bond allocation across $300,000 in 401(k), $100,000 in Roth IRA, $200,000 in taxable brokerage:

• 401(k): $300,000 → bonds (the entire $180,000 bond allocation) + $120,000 of total-market equity
• Roth IRA: $100,000 → emerging markets / small-cap aggressive equity (highest expected return)
• Taxable brokerage: $200,000 → total US market index + foreign developed-markets equity (qualified dividends + step-up potential + foreign tax credit)

Same 70/30 allocation, but the tax efficiency over thirty years is materially higher than holding the same 70/30 mix in every account proportionally. Vanguard’s modeling suggests the asset location benefit can compound to 0.5 to 1.5 percentage points of additional after-tax return per year, which over thirty years on a $600,000 portfolio is six figures of additional household wealth.

Sources

• Glidepath methodology for major target-date retirement funds: Vanguard — Target Retirement Funds methodology and Fidelity — Freedom Funds methodology.
• Sequence-of-returns risk research: Wade Pfau — Retirement Researcher maintains an extensive set of articles on sequence-of-returns risk and its implications for retirement-year asset allocation.
• Behavior gap research: DALBAR — Quantitative Analysis of Investor Behavior annual reports on the gap between fund returns and investor returns.
• Asset allocation framework from the Bogleheads community: Bogleheads — Asset allocation.
• US Social Security replacement rate by income: Social Security Administration — Replacement rates and SSA — Quick Calculator.

If a number on this page looks off against current professional consensus, the academic researchers above update frequently; let us know via contact and we will reconcile.