Safe Withdrawal Rates With Equity Volatility: Why the 4% Rule Needs Adjustment

The 4% rule came from Bill Bengen’s 1994 paper and the subsequent Trinity Study. Using U.S. stock and bond returns from 1926 forward, both papers showed that a retiree withdrawing 4% of starting portfolio value, adjusted for inflation each year, would survive a 30-year retirement in every historical cohort.

The rule assumed a 50/50 to 75/25 equity/bond mix, full diversification, and rebalancing. None of those assumptions hold cleanly for a retiree walking in with $3M where $1.8M is one tech stock.

This matters because the 4% rule is the single most-quoted figure in retirement planning. People build their retirement math around it. For equity-heavy retirees, anchoring on 4% without adjustment can lead to portfolio exhaustion in the mid-70s.

What the Trinity study actually tested

The Trinity Study tested withdrawal rates on portfolios of broad-market U.S. stocks and intermediate-term government bonds, rebalanced annually. Success was defined as the portfolio lasting 30 years with inflation-adjusted withdrawals.

Key findings:

• 4% survives 95-100% of historical 30-year periods at 50/50 to 75/25 allocations
• 5% survives only 70-80% of periods, too risky for most
• 3% survives effectively all periods but under-consumes in good scenarios

Forward-looking critiques have argued that the 4% rule overstates safe withdrawal because U.S. market returns since 1926 were an outlier versus other developed markets. The “safe” rate under international return data is closer to 3.3%.

Why concentration makes the rate lower

Three forces push the safe rate down when the portfolio is concentrated:

1. Higher volatility. A single-stock portfolio has 30-40% annualized volatility versus 15-18% for the S&P 500. Higher variance expands the distribution of outcomes and makes bad sequences more frequent.

2. Single-company tail risk. Trinity and Bengen assumed index returns. Index drawdowns are bounded by the broad market. Single-stock drawdowns can be 70-80%+ and in some cases terminal (the company fails).

3. No free rebalancing. The Trinity test assumed rebalancing, selling winners and buying losers. A concentrated retiree can’t rebalance because selling triggers capital gains tax on embedded appreciation.

Academic work using Monte Carlo simulations on single-stock portfolios finds safe withdrawal rates of 2.5-3.0% for 30-year retirements, compared to 3.5-4.0% for diversified portfolios. That difference is real money: on $3M, it’s $30K-$45K less per year.

Dynamic withdrawal strategies

Static withdrawal rates are a simplification. Real retirees can adjust spending based on portfolio performance. Two popular approaches:

Guardrails (Guyton-Klinger). Set an initial rate (say 4%). If portfolio falls 20% below starting value, cut withdrawal by 10%. If portfolio rises 20% above, increase withdrawal by 10%. Allows higher initial rates (up to 5-5.5%) at the cost of spending variability.

Constant percentage. Withdraw a fixed percentage of current portfolio each year. In bad years, withdrawal drops proportionally. Never exhausts the portfolio but spending is volatile.

For concentrated retirees, guardrails with tight downside triggers can protect against sequence risk. A 15% downside trigger instead of 20% forces earlier belt-tightening in bad years.

The role of the Roth bucket

Tax diversification changes the withdrawal math. A retiree with $1M in Roth, $1.5M in traditional, and $500K in taxable can manage withdrawals across buckets to minimize lifetime tax:

• Taxable bucket: sells first, controls capital gains bracket
• Traditional bucket: fills remaining ordinary-income bracket up to some cap
• Roth bucket: last resort, also for unexpected large expenses

In a bad market year, the retiree can draw from taxable (realizing losses that offset gains) and tap Roth without tax consequence. The traditional IRA stays untouched, preserving growth. This cross-bucket flexibility is worth 30-50 basis points of effective withdrawal capacity.

The Social Security bond

Social Security, once claimed, acts as an inflation-indexed annuity. For a retiree expecting $48K/year at 70, that’s the equivalent of a $1.2M inflation-indexed bond at a 4% yield.

Treating SSA as a bond-equivalent lets the portfolio carry more equity risk. A retiree with $48K of SSA plus $80K from portfolio withdrawal has 60% of income from the SSA “bond” and 40% from portfolio. Portfolio volatility becomes less lethal.

Delaying SSA from 62 to 70 increases the benefit by about 8% per year. For an equity-heavy retiree worried about sequence risk, delaying to 70 is usually correct.

IRMAA and tax drag on withdrawal

Withdrawals from traditional IRAs count as ordinary income and feed Medicare Part B and D IRMAA. The 2025 tiers (based on 2023 MAGI for current premiums):

• $106K single / $212K married: Tier 1
• $133K single / $266K married: Tier 2
• $167K single / $334K married: Tier 3

A retiree pulling $160K from a traditional IRA sits at Tier 2 for IRMAA. Annual surcharge: about $2,600 per person for Part B, $500 for Part D. For a couple, that’s $6,200 per year of surcharge, or nearly 4% of the withdrawal.

Mixing in Roth and taxable withdrawals can keep MAGI below Tier 1 even while total spendable income is $200K+. This is another argument for a Roth-heavy retirement.

State tax adjustments

State tax doesn’t scale proportionally to withdrawal rate. A retiree in California pulls $120K and pays 9.3% state marginal on much of it, roughly $8,000 of annual state tax. The same retiree in Florida pays zero. The state-tax swing on a 30-year retirement is $240K.

For retirees with concentrated state-tax exposure in California, New York, or New Jersey, relocating before or early in retirement can add 30-50 bps of effective safe withdrawal rate. The catch: see trailing-nexus rules, RSU income from grants made in California may still be California-sourced even after a move.

Frequently asked

If I’m concentrated, what’s my actual safe rate? Roughly 2.5-3.0% for a single-stock 60%+ concentrated portfolio over 30 years. De-concentrate to <20% and the safe rate returns to 3.5-4.0%.

Does the 4% rule assume inflation-adjusted withdrawals? Yes. The 4% is the first-year withdrawal. Each subsequent year increases by CPI. A 3% CPI environment means the nominal withdrawal grows meaningfully.

Should I annuitize to manage sequence risk? For some retirees, a Single Premium Immediate Annuity covering essential expenses (housing, food, insurance) eliminates sequence risk on that portion. The remainder invests aggressively. Trade-off: annuity returns are low and principal is gone.

How does the §7520 rate affect retirement planning? The §7520 rate (120% of the federal midterm rate) sets the actuarial rate for charitable remainder trusts and grantor-retained annuity trusts. Higher §7520 rates favor CRTs (larger income stream per contribution). Lower rates favor GRATs. Retirees considering these structures should check the current §7520 rate before locking in.

What’s the interaction with HSA distributions? HSA distributions for qualified medical expenses are tax-free and don’t count toward MAGI or IRMAA. For a retiree with a $200K HSA balance, that’s $200K of future medical spending at zero tax and zero IRMAA drag. The 2025 HSA family contribution limit is $8,550 for current accumulation.