The Trinity Study and the origin of the 4% rule
In 1994, financial planner William Bengen published a study in the Journal of Financial Planning called "Determining Withdrawal Rates Using Historical Data." He used historical US stock and bond returns from 1926 onward to backtest how much a retiree could safely withdraw from a portfolio and still have it last 30 years. The conclusion: 4% inflation-adjusted withdrawals from a 50–75% stock portfolio survived every historical 30-year window tested, including retirements starting in 1929 (just before the Great Depression) and 1966 (the start of the stagflation era).
Four years later, three Trinity University professors — Philip Cooley, Carl Hubbard, and Daniel Walz — replicated and extended Bengen's analysis. Their 1998 paper, "Retirement Spending: Choosing a Sustainable Withdrawal Rate," became known as the Trinity Study. It cemented "4% rule" as shorthand in the financial planning industry. The original study found that 4% withdrawal rates had near-100% success rates for 30-year retirements with stock-heavy portfolios.
The rule's elegance is partly why it spread: a single number gives retirees a clean answer to "how much do I need?" Multiply your annual spending by 25 (the reciprocal of 4%) and that's your nest egg target. For someone wanting $60,000/year, that's $1.5M. The arithmetic is simple enough to do in your head, which is rare in retirement planning.
But the rule also has limitations baked in. It was built on US market returns, which were exceptional in the 20th century. It assumes a fixed 30-year horizon. It does not account for taxes, fees, or behavioral mistakes during downturns. Modern critics — and Bengen himself — have spent decades refining and challenging the original number.
Why sequence-of-returns risk dominates the math
Two retirees can earn the exact same average return over 30 years and end up in radically different places. The difference is the order of returns. If your worst years happen first, you are forced to withdraw from a shrinking balance, locking in losses that cannot recover. If your worst years happen last, you have a fat cushion to absorb them.
A canonical example: a retiree who started in 1966 with a 4% rate barely survived (the stagflation of the 70s did serious damage), while a retiree who started in 1982 (right as a multi-decade bull market began) ended with several times their starting balance — same withdrawal rule, same asset allocation, vastly different outcomes.
This is why the first 5–10 years of retirement matter disproportionately. A 30% drawdown in year two is far more dangerous than the same drawdown in year 20. Practical responses: hold 1–3 years of cash or short bonds as a buffer so you can avoid selling equities in a downturn; consider a "bond tent" that temporarily raises your bond allocation around the retirement date; build in flexibility to cut discretionary spending by 15–20% after a bad year.
Sequence risk is also why the 4% rule includes such a wide margin of safety in average historical paths. The rule is calibrated to survive the worst sequences — meaning in most paths, retirees end with much more money than they started. That "wasted" money is the insurance premium for surviving the bad sequences.
Bengen's 2022 update: is it actually 4.7%?
In 2022, William Bengen revisited his own research and published an update that raised the safe withdrawal rate to 4.7%. The upward revision was driven by two changes in methodology: a more diversified asset mix (including small-cap value, mid-cap, and international equities, rather than just S&P 500 and bonds) and a more careful look at the actual worst-case historical sequences.
His logic is sound: the original 4% rule was conservative because it used a narrow asset allocation. Adding additional asset classes that have historically diversified equity risk does buy you a higher safe withdrawal rate. But the update has caveats. It still assumes a 30-year horizon. It assumes you actually hold and rebalance the more diversified allocation. And it assumes future returns will resemble the historical distribution, which is the assumption most worth questioning.
Other prominent researchers have gone the opposite direction. Wade Pfau's analyses, factoring in lower bond yields and high equity valuations as of the 2020s, suggest 3.0–3.5% is more defensible. Big ERN's exhaustive Safe Withdrawal Rate Series argues for 3.25–3.5% for long retirements.
The honest synthesis: 4% remains a reasonable baseline for a 30-year retirement of someone with a diversified equity-heavy portfolio. Early retirees facing 40–50 year horizons should use 3.25–3.75%. People with significant guaranteed income (Social Security, pensions) can use a higher rate on the remaining portfolio-funded portion of spending. Anyone serious should also have a contingency plan for cutting spending if returns disappoint in the first decade.
Guardrails: making the withdrawal rate dynamic
The 4% rule assumes you withdraw a fixed real-dollar amount every year regardless of how the portfolio performs. That fixed-amount rule is the worst-case scenario for sequence risk — you keep pulling the same dollars whether the market is up 30% or down 30%. Real retirees do not behave that way. They cut spending after bad years and treat themselves after good years.
Jonathan Guyton formalized this in his "guardrails" research. The most-cited version of the rule: if your current withdrawal rate climbs more than 20% above your initial rate (because the portfolio shrank), cut spending by 10%; if it falls more than 20% below (because the portfolio grew), increase spending by 10%. Other guardrails skip the inflation adjustment in years following a portfolio decline.
Backtests consistently show that guardrails can support a higher starting withdrawal rate — 4.5–5% versus 4% — with the same probability of portfolio survival. The trade-off is that your annual spending becomes variable. Some years you cut back; some years you spend more. For retirees with discretionary budget categories (travel, dining, gifts), this trade-off is usually worth it.
A simple version: maintain a baseline budget covering essential spending (housing, food, healthcare, basic transport) that you commit to no matter what, and a discretionary budget that flexes with portfolio performance. The baseline should be coverable by a conservative 3% withdrawal rate plus any guaranteed income, so it survives even severe market drawdowns.
Modern critiques and the 4% rule's future
The most common modern critique is valuation-based: US equities entered the 2020s at historically elevated valuations, and bond yields were near zero for most of the decade. Both factors argue for lower future returns than the historical average. If forward real returns average 3% instead of 5–6%, the 4% rule has less margin and the chance of failure grows.
A second critique is about non-US generalizability. The 4% rule was derived from US returns during the 20th century, which were exceptionally strong relative to other developed markets. Studies that include international equities or extend to non-US retirees find much lower safe withdrawal rates — often 2.5–3.5% — because not every country's stock market delivered US-style returns even in the best century of the modern era.
A third critique is about flexibility. The static 4% rule assumes you will not adjust spending based on portfolio performance. Real humans do adjust. Frameworks like Guyton-Klinger guardrails, the bucket strategy, dynamic spending rules, and "spend-it-down" amortization-based approaches all relax the fixed-rule assumption and produce more efficient spending paths.
For most planners, the practical answer is: treat 4% as a useful default for 30-year retirements, plan for 3.25–3.75% if you are retiring early, build a flexibility plan, and revisit your withdrawal rate every few years based on portfolio performance. The exact rate matters less than having a strategy for when (not if) reality deviates from your projections.
