Financial advisor William Bengen published a paper in the October 1994 issue of the Journal of Financial Planning that quietly revolutionized retirement planning. Using historical market data from 1926 through 1991, Bengen tested every possible 30-year retirement period and asked a simple question: what initial withdrawal rate, with subsequent withdrawals adjusted for inflation, would have survived every period without depleting the portfolio? His answer was 4.15 percent for a portfolio of 50 to 75 percent large-cap stocks and the remainder in intermediate-term government bonds. The number was rounded to 4 percent and entered the financial planning canon as the "4 percent rule." Four years later, the so-called Trinity Study — three professors at Trinity University led by Philip Cooley, published in the February 1998 issue of the Journal of the American Association of Individual Investors — replicated and extended Bengen's work, confirming that 4 percent produced a 95 percent success rate over 30-year horizons with a similar 50/50 stock-bond portfolio. For nearly three decades, the 4 percent rule has been the default framework for retirement withdrawals — and for the past decade, it has been the subject of increasingly urgent debate about whether it still works in a world of lower bond yields, longer lifespans, and more complex market dynamics. The honest answer in 2026 is nuanced: the 4 percent rule remains a reasonable starting point but is no longer the conservative floor it was once considered, and retirees who treat it as a guarantee are taking more risk than they realize.
The original research: Bengen and Trinity
Bengen's 1994 paper, titled "Determining Withdrawal Rates Using Historical Data," tested 51 overlapping 30-year retirement periods starting in 1926. He assumed a portfolio of 50 to 75 percent large-cap U.S. stocks (S&P 500) and the remainder in intermediate-term government bonds, with annual rebalancing. Withdrawals were taken at the start of each year and adjusted for inflation. The worst-performing starting year was 1968, when a 4.15 percent withdrawal rate would have just barely survived the 30-year period. The best-performing starting years (1950, 1960, 1980) would have supported withdrawal rates above 8 percent. Bengen's conclusion: 4.15 percent was the maximum "safe" initial withdrawal rate that would have survived every historical period tested.
The Trinity Study, published four years later, broadened the analysis to include different portfolio compositions (25/75, 50/50, 75/25, 100/0 stocks/bonds) and different time horizons (15, 20, 25, 30 years). The 50/50 stock-bond portfolio with 4 percent withdrawals over 30 years produced a 95 percent success rate — meaning 5 percent of historical 30-year periods would have failed. The 75/25 portfolio improved the success rate to 98 percent. The 25/75 portfolio dropped it to 82 percent. The key findings: stock-heavy portfolios had higher success rates, longer horizons had lower success rates, and the 4 percent rate was approximately the threshold above which failure rates rose sharply. The study became the empirical foundation for the "4 percent rule" as practiced by financial planners.
Both studies used historical U.S. data, and both have known limitations. The U.S. experienced the strongest equity market performance of any major economy over the 20th century — what some researchers have called "American exceptionalism" in financial returns. Bengen's period covered the Great Depression, World War II, the post-war boom, the stagflation of the 1970s, and the bull market of the 1980s, but it also covered a period of structurally higher bond yields (intermediate government bonds yielded 4 to 8 percent for most of the period) that no longer exists. A retiree who began withdrawals in 1980 earned 12 percent on intermediate bonds for the first decade — a tailwind unavailable to today's retiree.
Sequence-of-returns risk: the silent killer
The reason a 4 percent withdrawal rate can fail despite a 30-year average return of 7 to 8 percent is sequence-of-returns risk. The order in which returns arrive matters enormously when you are withdrawing money. Consider two retirees, each with $1 million, each withdrawing $40,000 annually adjusted for inflation, each earning an average 7 percent return over 30 years. Retiree A experiences strong returns in the first decade and weak returns later — the portfolio grows, withdrawals are a small fraction of the balance, and the late-decade weakness has minimal impact. Retiree B experiences weak returns in the first decade and strong returns later — withdrawals consume a larger and larger fraction of the declining portfolio, and by the time strong returns arrive, there is little left to compound. Same average return; vastly different outcomes.
This is sequence-of-returns risk, and it is the primary threat to retirement portfolios. The worst historical sequence for U.S. retirees was 1968 to 1998, when a 4.15 percent withdrawal rate just barely survived. The 1973 to 2003 period was nearly as bad, with the 1973 to 1974 bear market (down 42 percent for the S&P 500) followed by inflation that pushed nominal withdrawals sharply higher. The 2000 retiree cohort faced a similar challenge: the dot-com crash from 2000 to 2002 (down 47 percent) followed by the 2008 financial crisis (down 37 percent), all while inflation continued. A 2000 retiree with a 4 percent withdrawal rate and a 60/40 portfolio would have ended 2023 with about 65 percent of the original balance — not depleted, but on a trajectory that would have failed in the 30-year window without later exceptional returns.
Sequence risk cannot be eliminated, but it can be managed. The most powerful technique is "bond tent" or "rising glide path" asset allocation: hold a higher bond allocation in the years immediately before and after retirement (typically 5 years before to 10 years after), then shift toward equities as the dangerous early-retirement period passes. Research from Michael Kitces and Wade Pfau, published in the Journal of Financial Planning in 2014, found that a bond tent improved success rates by 5 to 10 percentage points across various scenarios. Other techniques include keeping 1 to 2 years of expenses in cash (avoiding forced selling during downturns), using a variable withdrawal strategy (reducing withdrawals when the portfolio is stressed), and annuitizing a portion of essential expenses to remove them from market risk entirely.
Wade Pfau and the new lower safe withdrawal rates
Wade Pfau, professor of retirement income at The American College of Financial Services, has been the leading academic voice arguing that the 4 percent rule is too optimistic for current retirees. His 2010 paper in the Journal of Financial Planning, "An International Perspective on Safe Withdrawal Rates," examined 17 developed countries' equity and bond returns from 1900 to 2000 and found that the maximum safe withdrawal rate was 4.4 percent for the U.S. (the best-performing market) but averaged just 3.0 percent across all 17 countries, and was below 3 percent for several markets including Italy, Belgium, and France. The implication: the U.S. experience that informed Bengen's 4 percent rule was an outlier, and using the U.S. historical average as the global benchmark may overestimate safe withdrawal rates by 25 to 35 percent.
In subsequent research, Pfau and others have argued that current market conditions — particularly low bond yields and elevated equity valuations — point to even lower safe withdrawal rates. Pfau's 2018 analysis, published in his book How Much Can I Spend in Retirement?, estimated that a 50/50 portfolio with a 4 percent withdrawal rate had only a 50 to 60 percent success rate over 30 years given 2018 market conditions, versus the 95 percent success rate suggested by historical U.S. data. His recommended safe withdrawal rate for current retirees: 3 to 3.5 percent, or roughly 25 percent lower than the traditional 4 percent rule.
The Morningstar Retirement Income Model, updated annually, similarly estimates safe withdrawal rates for the current market environment. The 2024 edition estimated that a 50/50 portfolio with a 4 percent withdrawal rate had an 86 percent success rate over 30 years, down from the 95 percent suggested by historical data, and recommended 3.3 percent as the starting withdrawal rate for a 90 percent success rate. The estimate shifts annually with market valuations and bond yields; in years of low yields and high valuations, the estimate drops further. The methodology is more conservative than Bengen's because it forward-tests rather than back-tests, using Monte Carlo simulations on current market conditions rather than assuming the future will look like the past.
Variable spending: the Guyton-Klinger rules
The fundamental problem with the 4 percent rule is its rigidity: a fixed withdrawal adjusted only for inflation, regardless of portfolio performance. Real retirees have flexibility — they can reduce spending when the portfolio is stressed, defer discretionary purchases, or take a part-time job. Variable spending rules formalize this flexibility and dramatically improve portfolio survival rates. The most thoroughly researched variable spending framework is the Guyton-Klinger rules, developed by financial planner Jonathan Guyton and computer scientist William Klinger in a series of papers published between 2006 and 2010.
The Guyton-Klinger rules have four components. First, the initial withdrawal rate is set at a target (typically 4 to 5 percent). Second, withdrawals are adjusted for inflation except in years following a portfolio decline — a "freezing" rule that prevents inflation adjustments during downturns. Third, a "prosperity rule" allows the retiree to increase withdrawals by 10 percent when the current withdrawal rate (withdrawal divided by portfolio balance) drops 20 percent below the initial rate — a sign that the portfolio has grown faster than expected. Fourth, a "capital preservation rule" reduces withdrawals by 10 percent when the current withdrawal rate rises 20 percent above the initial rate — a sign that the portfolio is being depleted faster than expected.
The Guyton-Klinger rules have been tested extensively and dramatically improve outcomes. Guyton's original 2006 paper in the Journal of Financial Planning found that with a 4.5 percent initial withdrawal rate, the rules produced a 96 percent success rate over 40 years — better than the 4 percent fixed rule's 95 percent over 30 years, despite the longer horizon and higher initial rate. Subsequent research by Michael Kitces and others has confirmed and extended these findings, showing that variable spending rules essentially eliminate the failure risk of the 4 percent rule for 30-year horizons and extend safe horizons to 40 or 50 years. The cost is behavioral — retirees must be willing to cut spending during downturns, which is psychologically difficult even when financially necessary.
30 vs 50-year horizons: the longevity question
The 4 percent rule was tested over 30-year horizons, which was appropriate when Bengen published in 1994 and life expectancy for a 65-year-old was about 17 years (to age 82). Today, life expectancy for a 65-year-old is about 19 years (to age 84), and for a 65-year-old couple, there is a 50 percent chance that at least one spouse will live to age 92 and a 25 percent chance one will live to age 97. For early retirees — those who retire at 55 or 60 — the relevant horizon is 40 to 45 years, not 30. The 4 percent rule was not tested for these horizons, and the limited testing that has been done suggests failure rates rise sharply.
Wade Pfau's 2018 analysis found that the 4 percent rule's success rate over 40-year horizons was approximately 85 percent (versus 95 percent over 30 years) and over 50-year horizons was approximately 70 percent. For a 55-year-old retiree expecting to live to 95, a 50-year horizon is realistic, and a 70 percent success rate is not "safe" by any reasonable standard. The implication: early retirees need lower withdrawal rates (typically 3 to 3.5 percent), variable spending rules, or annuitized income to bridge the longer horizon. A 30-year retirement is not the maximum scenario; it is the median scenario for a 65-year-old couple, and a 35 to 40 year retirement is increasingly common.
The longevity risk also compounds with the sequence-of-returns risk. The longer the retirement horizon, the more opportunity for a severe market downturn to occur during the vulnerable early-retirement period. A retiree who experiences a 2000-style sequence in the first decade of a 30-year retirement has two decades to recover; a retiree experiencing the same sequence in the first decade of a 50-year retirement has four decades, but also needs the portfolio to last those four decades. The mathematics of safe withdrawal rates over 50-year horizons favors lower initial rates, variable spending, and meaningful equity allocations to ensure the portfolio outpaces inflation over the long haul.
International diversification: does it help or hurt
Bengen's original analysis used only U.S. stocks and U.S. bonds. The U.S. market has been the best-performing major equity market over the past century, raising the question: would international diversification have improved or hurt safe withdrawal rates? The answer depends on the period and the methodology, but recent research suggests international diversification modestly improves safe withdrawal rates, particularly for retirees in markets outside the U.S.
A 2019 study by Wade Pfau and others, published in the Journal of Financial Planning, tested withdrawal rates using global portfolios (50 percent U.S. stocks, 50 percent international stocks) versus U.S.-only portfolios across 17 developed markets from 1900 to 2015. The global portfolio produced slightly higher safe withdrawal rates than the U.S.-only portfolio in most non-U.S. countries (because it avoided home-country concentration risk) and slightly lower safe withdrawal rates in the U.S. (because non-U.S. returns have been lower on average). The net effect for a U.S. retiree: roughly break-even to slightly negative, but with the benefit of diversification reducing the tail risk of U.S.-specific underperformance in the coming decades.
The case for international diversification in 2026 is stronger than the historical record suggests, for two reasons. First, U.S. equity valuations are at historically high levels (Shiller CAPE ratio of 35 in early 2024, versus the historical average of 17), which is a strong predictor of lower forward returns. International equities are at much more reasonable valuations (Shiller CAPE of 15 to 18 in Europe and emerging markets), suggesting higher forward returns. Second, the U.S. dollar has been strong for a decade, and mean reversion would favor international assets over the next decade. A 60/40 U.S./international split within the equity allocation is now the consensus recommendation among most financial planners, with some arguing for even higher international allocations given current valuations.
Bond yields in 2026: the new reality
The bond market has fundamentally changed since Bengen published in 1994. From 1994 through 2008, intermediate-term U.S. Treasury bonds yielded 4 to 6 percent, providing meaningful income to retirement portfolios. From 2008 through 2022, the same bonds yielded 0.5 to 2.5 percent, historically low and inadequate to support a 4 percent withdrawal rate from a bond-heavy portfolio. The 2022 to 2024 rate increases brought intermediate Treasury yields back to 4 to 4.5 percent, and as of early 2026, yields have moderated to 3.5 to 4.0 percent as the Fed has cut rates. This is higher than the post-2008 era but still below the 1994-era yields that informed the original Bengen research.
The bond yield matters enormously for safe withdrawal rates because bonds are the stable component of the portfolio that funds withdrawals during stock market downturns. Bengen's worst-case scenarios all assumed the retiree could sell bonds during stock downturns; if bond yields are too low to provide meaningful income, the retiree must sell bonds at lower prices (when yields rise, bond prices fall) or sell stocks at depressed prices. The 2022 simultaneous decline in both stocks (down 18 percent for the S&P 500) and bonds (down 13 percent for intermediate Treasuries) was the first time in modern history that both asset classes declined by double digits in the same year, and it exposed the vulnerability of traditional 60/40 portfolios to a rising-rate environment.
The implication for current retirees is that bond allocations need to be more thoughtful than the traditional "intermediate Treasuries" approach. Strategies include: laddering individual bonds to manage reinvestment risk, adding TIPS (Treasury Inflation-Protected Securities) for inflation protection, holding a small allocation to high-quality corporate bonds for yield enhancement, and considering annuities for the portion of expenses that need guaranteed lifetime income. The bond allocation's job in a retirement portfolio is not to maximize return but to provide stability and income during stock downturns — and that job is harder in 2026 than it was in 1994.
Annuitization: the underused tool
Single-premium immediate annuities (SPIAs) and qualified longevity annuity contracts (QLACs) provide guaranteed lifetime income in exchange for an upfront premium. They address the two biggest retirement risks — longevity risk and sequence-of-returns risk — by transferring them to an insurance company. Yet annuities are remarkably underused, in part because of legitimate concerns about high fees, complexity, and the loss of liquidity, and in part because of psychological resistance to "giving up" a lump sum. The academic case for partial annuitization is strong: a 2023 study by Wade Pfau and Michael Finke in the Journal of Financial Planning found that retirees who annuitized 25 to 50 percent of their portfolio had higher success rates and higher lifetime income than those who held all their assets in investment portfolios, particularly in low-yield environments.
The strategy most financial planners recommend is "annuitize essentials, invest for discretionary." Calculate your essential annual expenses (housing, food, healthcare, basic transportation), subtract guaranteed income (Social Security, pensions), and annuitize enough of your portfolio to cover the gap. This typically requires 25 to 50 percent of the portfolio for a 65-year-old, depending on interest rates and annuity pricing. The remaining portfolio is invested for discretionary spending (travel, entertainment, gifts) and growth, with variable withdrawals that can be reduced during downturns without affecting essentials.
SPIA payouts as of 2024 for a 65-year-old are approximately 6.5 to 7 percent of the premium annually — meaning a $500,000 premium produces $32,500 to $35,000 of lifetime annual income. This is substantially higher than the 3.5 to 4 percent safe withdrawal rate from an investment portfolio, reflecting the mortality credits (the pool of annuitants who die early subsidize those who live long). The trade-off is loss of access to the principal and loss of any bequest value. QLACs (which defer income to age 85) provide longevity insurance at lower cost and are increasingly used to address the "what if I live to 95" question. Annuity decisions are complex and benefit from working with a fee-only advisor who does not earn commissions on annuity sales.
Building your own withdrawal strategy
The honest synthesis of the 2026 research is that the 4 percent rule is a useful starting point but not a final answer. A more defensible approach combines multiple techniques: a lower base withdrawal rate (3 to 3.5 percent for 30-year horizons, 2.5 to 3 percent for 40+ year horizons), variable spending rules (Guyton-Klinger or similar) to adjust spending with portfolio performance, a bond tent or rising glide path asset allocation to manage sequence risk, partial annuitization to cover essential expenses, and Social Security delay to age 70 to maximize guaranteed lifetime income. This multi-layered approach is more complex than the simple 4 percent rule but vastly more robust to the risks retirees actually face.
The practical protocol: start by calculating your essential expenses and your discretionary expenses separately. Delay Social Security to age 70 if possible (the 8 percent per year delayed retirement credit is the best longevity insurance available). Annuities enough of your portfolio (often 25 to 50 percent) to cover essential expenses not covered by Social Security. Invest the remainder in a 60/40 to 70/30 portfolio, withdraw at a 3 to 3.5 percent rate for discretionary expenses, and apply Guyton-Klinger-style variable spending rules to adjust if the portfolio is stressed. Maintain 1 to 2 years of cash for immediate spending needs, avoiding forced selling during downturns.
The math is more conservative than the 4 percent rule, but the outcome is more reliable. A retiree who follows this protocol in 2026 has a 90 to 95 percent probability of portfolio survival over a 30-year horizon and a 75 to 85 percent probability over a 40-year horizon — versus the 50 to 70 percent probabilities that the simple 4 percent rule offers in current market conditions. The trade-off is the complexity and the need for ongoing management, which is why working with a fee-only retirement income specialist is increasingly recommended for retirees with portfolios over $500,000. The cost of professional management (typically 0.5 to 1 percent of assets annually) is often offset by improved withdrawal strategies and tax efficiency.
The behavioral dimension: why rules-based matters
Even the best withdrawal strategy fails if the retiree abandons it during periods of stress. The 2008 financial crisis produced a wave of retirees who sold equities at the bottom and never returned, locking in losses and reducing their portfolios by 30 to 50 percent more than the market decline alone would have caused. The 2020 COVID crash was similar — retail investors pulled $326 billion from equity funds in March 2020, according to Morningstar, missing the subsequent recovery. The behavioral discipline to maintain a withdrawal strategy through market downturns is the single largest determinant of long-term outcomes, and rules-based strategies are easier to follow than discretionary ones because the rules remove the emotional decision.
The specific rules vary by retiree, but several principles are universal. First, write down your withdrawal strategy in advance — the specific rate, the variable spending rules, the rebalancing protocol, the contingency plans. Review and update annually. Second, separate your portfolio into buckets (cash for 1 to 2 years, bonds for 3 to 7 years, equities for long-term growth) to provide a behavioral buffer during downturns. Third, work with an accountability partner — a spouse, a financial advisor, or even a trusted friend — who can talk you out of panic decisions during market stress. Fourth, limit how often you check your portfolio balance; quarterly is sufficient, daily is destructive.
The 4 percent rule's greatest legacy may be that it gave retirees a simple, defensible answer to a complex question. The 2026 reality is that the simple answer is no longer sufficient, and retirees need a more nuanced approach. But the underlying principle — that a diversified portfolio, managed with discipline, can support a reasonable withdrawal rate over a 30-year retirement — remains sound. The right withdrawal rate for you depends on your specific situation: your age, your portfolio size, your essential expenses, your risk tolerance, your bequest goals, and your willingness to follow a variable spending strategy. Start with the research, customize to your circumstances, and build a strategy you can actually sustain for the decades ahead.