The Yale endowment paid out roughly 5% of its portfolio value every year for the last forty years. Through the 1987 crash, the dot-com bust, the 2008 financial crisis, and the 2020 pandemic, the spending kept flowing. The portfolio is larger today than it was at the start. The strategy is older than that, and it works.
The mechanics are simpler than most retirement spending plans. Take a percentage of the current portfolio. Spend it. Repeat next year. There is no inflation calculation, no guardrail trigger, no withdrawal rate drift to monitor. The portfolio decides how much you spend; your only job is to choose the percentage.
This is what percentage-of-portfolio withdrawals look like in practice. It is the most mathematically robust retirement spending strategy, the most psychologically demanding, and the one that fits a specific kind of retirement plan unusually well.
What Percentage-of-Portfolio Actually Does
The mechanic is one line of arithmetic:
annual withdrawal = portfolio value × withdrawal rate
For a 4% rate on a 1,000,000 portfolio:
- Portfolio at 1,000,000: withdraw 40,000
- Portfolio drops to 700,000 after a crash: withdraw 28,000
- Portfolio recovers to 1,200,000: withdraw 48,000
That is the entire strategy. There is no inflation toggle. There is no minimum or maximum. There is no rule for when the portfolio has "earned" a raise. The portfolio sets the withdrawal directly.
The trade-off is binary. You give up income predictability completely. In exchange, you get a strategy that cannot fail because it does not promise anything the portfolio cannot deliver.
The Anti-Depletion Math
Pure percentage-of-portfolio is the only common withdrawal strategy that cannot deplete the portfolio in finite time. The reasoning is straightforward: if the withdrawal is always a fraction of the current balance, the balance asymptotically approaches zero but never reaches it.
Take a portfolio of 1,000,000 with a 4% withdrawal rate and zero returns:
- Year 1: 1,000,000 → withdraw 40,000 → 960,000 remains
- Year 2: 960,000 → withdraw 38,400 → 921,600 remains
- Year 30: roughly 290,000 remains
After a brutal sequence with no growth at all, the portfolio is still about 29% of its starting value. With realistic positive average returns, the portfolio survives indefinitely. This is what makes it appropriate for endowments that exist on perpetual time horizons.
For a retiree, this property is genuinely valuable but partly philosophical. A retiree on a 30-year horizon does not actually need a portfolio that survives forever. They need a portfolio that delivers livable income for 30 years. Those are different goals.
The Income Volatility Problem
The reason most retirees do not use pure percentage-of-portfolio is that the income volatility is severe. The strategy passes 100% of market volatility through to spending. A 30% bear market means a 30% income cut, applied immediately and continuing until markets recover.
For a retiree spending 4,000 per month, a bear market drops income to 2,800. The mortgage is still 1,800. Health insurance is still 600. Groceries are still 800. The variable portion of the budget is now negative. The math says the strategy works; the household's bank account says otherwise.
This is the central reason percentage-of-portfolio is a niche strategy for retirees despite being the safest one for the portfolio. It demands a household structure where most spending is genuinely flexible. Either the fixed costs have to be small relative to total spending, or there has to be a non-portfolio income stream that covers the fixed costs.
Smoothing: How Endowments Handle the Volatility
Yale, Harvard, Stanford, and most major endowments do not use a pure percentage-of-portfolio rule. They use a smoothed version. The most common formulation:
spending = (smoothing weight × prior spending × inflation) + ((1 - smoothing weight) × portfolio × rate)
With a smoothing weight of 70% and a 5% rate, the new year's spending is 70% based on last year's amount (adjusted for inflation) and 30% based on a fresh portfolio-percentage calculation. This blends stability and responsiveness.
The practical effect is that a 30% market drop translates to roughly a 9% spending cut in the first year, not 30%. Over several years, spending fully adjusts to the lower portfolio level if the market does not recover. The smoothing rule does not eliminate the volatility; it spreads it across time.
Retirees can implement the same approach by capping the year-over-year spending change at a reasonable bound (say 10%) and rolling the rest of the adjustment into subsequent years. The math gets slightly more complex, but the income path is dramatically smoother.
How It Compares Across 50,000 Paths
Under a fat-tailed distribution, 30-year horizon, 60/40 portfolio:
| Strategy | Initial rate | Success rate | Income range (10th to 90th percentile, year 15) |
|---|---|---|---|
| Static (4% rule) | 4.0% | 82% | 35,000 to 49,000 (real, with inflation) |
| Pure percentage | 4.0% | 99% | 22,000 to 71,000 |
| Floor-and-ceiling | 4.0% / 70% floor | 91% | 33,600 to 50,400 |
| Guyton-Klinger | 5.0% | 93% | 36,000 to 58,000 |
| Pure percentage | 5.0% | 97% | 24,000 to 84,000 |
The 99% success rate of pure percentage at 4% is real. The catch shows up in the income range. A 10th-percentile retiree under pure percentage is living on 22,000 a year by year 15 - a third of what an upper-decile retiree under the same strategy is spending. That is a livable income for some households and an emergency for others.
The success rate is also slightly misleading. The 1% of "failures" are not portfolio depletions; they are paths where the portfolio shrinks enough that the percentage-based withdrawal falls below a livable threshold. The portfolio is technically still alive. The retiree is functionally broke.
When This Strategy Fits Your Plan
Pure percentage-of-portfolio is not a general-purpose spending strategy. It is the right choice for a specific retirement profile.
You have non-portfolio income that covers fixed costs. Social Security, a pension, or part-time work that produces enough to handle housing, healthcare, food, and basic obligations. The portfolio's role is to fund the discretionary part of retirement: travel, hobbies, gifts, lifestyle. When markets crash, you cut the discretionary spending. When they boom, you upgrade it. This is exactly the trade-off pure percentage handles best.
Your fixed costs are genuinely small. A retiree without a mortgage, with low healthcare costs (perhaps through a paid-up insurance arrangement), and modest baseline lifestyle requirements can absorb a 40% income cut without operational damage. Most retirees cannot, but some can.
You have a long planning horizon. A 50-year retirement (early retiree, healthy, expected longevity) makes the perpetual-survival property of percentage-of-portfolio more valuable than it is for a typical 30-year horizon. The longer the horizon, the worse the static-rule failure rate becomes, and the more the percentage-of-portfolio's mathematical robustness matters.
The Layering Approach
The most pragmatic way to use percentage-of-portfolio is as one component of a layered income plan rather than the entire strategy. The layers:
- Floor income from fixed sources: Social Security, pension, annuity, part-time work. Covers all non-discretionary spending.
- Portfolio withdrawals using percentage-of-portfolio: Funds discretionary spending and grows or shrinks with the market.
- Cash buffer: 1-2 years of expenses to absorb the worst part of a bear market without selling.
This layered approach gets the benefits of percentage-of-portfolio (anti-depletion, automatic market adjustment) without the catastrophic income-collapse risk. The fixed income covers the baseline. The portfolio adjusts the lifestyle. The cash buffer prevents panic-selling at the bottom.
For retirees with this structure, percentage-of-portfolio is not a niche choice. It is the optimal one.
What This Means for Your Plan
Percentage-of-portfolio is rarely the only strategy you need. For most retirees with material fixed costs, floor-and-ceiling withdrawals or Guyton-Klinger guardrails are better fits.
Smoothing matters. A pure year-by-year percentage rule passes through too much volatility. A 3-year trailing average or a capped year-over-year change captures most of the strategy's benefits without the worst of the income swings. Endowments do this; you should too.
Test it against your actual income structure. Run a simulation that includes Social Security, any pension, and the portfolio. The portfolio's job is to fund the gap, not the entire spending plan. If the portfolio gap is small enough, percentage-of-portfolio handles it cleanly. If the portfolio is the entire spending plan, you almost certainly need a strategy with a floor.
Confirm the strategy matches your household profile. How to choose a retirement spending strategy walks through the four inputs that decide whether percentage-of-portfolio is the right call or whether floor-and-ceiling or guardrails fit better.
Watch the 30% bear market scenario. Force a -30% drop and observe the resulting income. If the cut would break your household, the strategy does not fit you. No simulation result changes that. A black swan stress test at age 67 is the cleanest way to see this.
Percentage-of-portfolio is available as a Pro spending strategy in Retirement Lab, with configurable annual rate. It pairs naturally with the income streams feature, which lets you model Social Security and pensions alongside the portfolio withdrawal.
Frequently Asked Questions
- What is a percentage-of-portfolio withdrawal strategy?
- Each year, you withdraw a fixed percentage (typically 3-5%) of the current portfolio value. If markets rise, your income rises. If they fall, your income falls. Mathematically, the portfolio cannot be depleted because you are always taking a fraction of what remains.
- What withdrawal rate should I use for percentage-of-portfolio?
- Common rates are 3-5%. A 4% rate on a 1,000,000 portfolio produces 40,000 in a typical year but can drop to 28,000 after a 30% bear market. Lower rates (3-3.5%) reduce income volatility and grow the portfolio. Higher rates (4.5-5%) extract more income but accept larger swings during crashes.
- Can the portfolio actually run out under this strategy?
- Not in the strict mathematical sense. Withdrawing a fraction of any non-zero number leaves a smaller non-zero number. In practice, however, a 30-50 year sequence of bad returns can shrink the portfolio enough that the percentage-based withdrawal becomes too small to live on. The risk is functional depletion, not literal depletion.
- How is this different from the 4% rule?
- The 4% rule sets an initial dollar amount and adjusts it for inflation, ignoring portfolio performance entirely. Percentage-of-portfolio recalculates the withdrawal based on the current portfolio value each year. The 4% rule offers stable income with depletion risk; percentage-of-portfolio offers no depletion risk with volatile income.
- How do university endowments use this strategy?
- Major endowments like Yale and Harvard pay out roughly 5% annually but use a smoothing rule based on a 3-year (or longer) trailing average of portfolio value. This avoids cutting spending sharply right after a crash while still adapting over time. Retirees can adopt the same smoothing approach to reduce income volatility.