Sequence-of-Returns Risk: First 5 Years of Retirement

Sequence-of-returns risk is the most underappreciated risk in retirement planning, and it is the technical reason that the 4 percent withdrawal rule has held up empirically since William Bengen first published it in 1994. The arithmetic feels counterintuitive: two retirees with identical 30-year average returns can end up at radically different portfolio values because of when the bad years happen relative to the withdrawals. The mechanism is share count. When equity prices drop 30 percent and you need to withdraw $40,000 anyway, you sell more shares at lower prices. Those shares are permanently gone from the portfolio. When prices recover, you have fewer shares to participate in the recovery. The lost recovery base compounds across the remaining retirement years.

This guide quantifies the risk in the first 5 years of retirement, walks the empirical evidence from Trinity Study and Pfau research, and details the four defenses: cash buffer sizing, variable spending rules, delayed Social Security, and partial annuitization. The conclusion: sequence-of-returns risk is real, measurable, and largely defensible with disciplined planning. The retirees who fail are usually the ones who panic during the first major drawdown and abandon their plan.

The math of sequence-of-returns risk: identical averages, different outcomes

Consider two hypothetical retirees, each starting with $1,000,000 at age 65 and withdrawing $40,000 per year (4 percent rate, adjusted for 3 percent inflation annually). Both experience the same 30 returns over their 30-year retirements, just in different order:

Retiree A: bad years first

• Year 1: -30 percent (portfolio drops to $660,000 after $40K withdrawal)
• Year 2: -15 percent (portfolio drops to $521,000 after $41.2K withdrawal)
• Year 3: +25 percent (portfolio recovers to $610,000 after $42.5K withdrawal)
• Year 4: +20 percent (portfolio reaches $688,000 after $43.7K withdrawal)
• Year 5: +15 percent (portfolio reaches $746,000 after $45K withdrawal)
• Years 6-30: alternating positive and negative years averaging 8 percent
• Year 30 outcome: Portfolio depleted in year 27. The retiree ran out of money 3 years before death.

Retiree B: bad years last

• Year 1: +25 percent (portfolio grows to $1,210,000 after $40K withdrawal)
• Year 2: +20 percent (portfolio grows to $1,411,000 after $41.2K withdrawal)
• Year 3: +15 percent (portfolio grows to $1,580,000 after $42.5K withdrawal)
• Year 4: +15 percent (portfolio reaches $1,773,000 after $43.7K withdrawal)
• Year 5: +20 percent (portfolio reaches $2,083,000 after $45K withdrawal)
• Years 6-30: alternating positive and negative years averaging 8 percent including the -30, -15 years saved for years 28-29
• Year 30 outcome: Portfolio at approximately $1,600,000 at death. The retiree has a meaningful bequest.

Same average return. Same withdrawals. Same retirement duration. Approximately $1.6 million in cumulative outcome difference, entirely due to sequence. The mechanism is what happens in years 1-5: Retiree A's $40,000 withdrawal in year 1 represents 4 percent of $1M, but to maintain that withdrawal after the 30 percent drop, year 2's effective withdrawal rate is 6.7 percent of the reduced portfolio ($40K of $660K). That higher effective withdrawal rate compounds the damage across recovery years.

Why the first 5 years matter most

The Trinity Study (Cooley, Hubbard, Walz, 1998) and subsequent Bengen and Pfau research consistently identify the first 5 years of retirement as the highest-leverage window for sequence-of-returns risk. Three reasons:

1. Portfolio is at its peak dollar size

A 30 percent drawdown on a $1M portfolio is $300,000 of paper loss. The same 30 percent drawdown on a portfolio that has been spent down to $500,000 by year 15 is $150,000 of paper loss. The absolute dollar damage from a drawdown is largest in the first 5 years simply because the portfolio is largest.

2. Withdrawal period is longest

A drawdown in year 1 affects 30 years of subsequent withdrawals. A drawdown in year 25 affects 5 years. The compounding effect of having fewer shares for the entire remaining retirement is asymmetric: early drawdowns have decades to compound; late drawdowns have only a few years.

3. No recovery base exists yet

Going into retirement, the portfolio has not yet had any years of compounding to build a cushion above the starting balance. By year 10, a portfolio with average returns is at approximately $1.3M to $1.5M, providing 30 to 50 percent cushion above the starting $1M. A 30 percent drawdown in year 10 still leaves the portfolio at or above the starting level. A 30 percent drawdown in year 1 cuts directly into the starting capital.

Empirical evidence: which retirement-start years failed historically

Bengen's original 1994 paper analyzed every 30-year period since 1926 to identify which retirement start years would have run out of money at a 4 percent withdrawal rate with a 50/50 stock/bond portfolio. The losers cluster around three episodes:

• 1929-1931: Great Depression. Retirees starting in 1929 with a heavy equity allocation faced approximately 60 percent equity drawdown by 1932. The 50/50 portfolio survived 30 years at 4 percent withdrawals, but barely.
• 1937-1939: Roosevelt recession compounding lingering Depression effects. Started under the 4 percent rule.
• 1965-1973: Stagflation era. High inflation (averaging 7 percent in some years) combined with flat-to-declining real equity returns produced the worst 30-year start in modern US history. A 4 percent withdrawal rate would have failed in some 1965-1969 starts.

All three failure clusters share a common feature: bad equity returns combined with high inflation in years 1-5. The combination forces both nominal withdrawal increases (CPI adjustment) and reduced portfolio capacity (drawdown), accelerating depletion. More recent challenging periods (2000-2003 dot-com crash, 2007-2009 financial crisis, 2022 stagflation scare) showed similar sequence patterns but with lower inflation, allowing 4 percent withdrawals to survive in retrospect.

Defense 1: cash buffer of 2 to 3 years of spending

The simplest defense against sequence-of-returns risk is holding 2 to 3 years of spending in cash equivalents at retirement. The mechanism: when a drawdown hits, you withdraw from cash instead of selling equities at low prices. The portfolio recovery base is preserved.

For a 65-year-old with $1M and $40K annual withdrawals, the cash buffer is $80K to $120K (2 to 3 years). Instruments: high-yield savings ($5K-$50K), money market funds ($10K-$100K, Vanguard Federal MM, Fidelity Government Cash Reserves), short Treasury bills (4-week to 52-week maturities), and brokered CDs (if held under FDIC limits).

In a normal year, the cash earns roughly 4.3 to 4.7 percent in 2026. In a drawdown year, the cash funds withdrawals directly. After the drawdown, the cash is refilled by selling appreciated equities (using the 3-bucket refill rule). This converts sequence-of-returns risk from a portfolio risk into a refill-timing risk, which is much easier to manage.

Studies by Kitces show that the 2 to 3 year cash buffer reduces worst-case portfolio failure rates from approximately 30 percent (no buffer, full equity drawdown) to approximately 10 to 15 percent (3-year buffer) at a 4 percent withdrawal rate over 30 years.

Defense 2: variable spending rules (Guyton-Klinger guardrails)

Static withdrawal rates (4 percent inflation-adjusted) are vulnerable to sequence-of-returns risk because they do not adapt to portfolio performance. Variable spending rules cut withdrawals after large drawdowns and increase them after strong returns, keeping the portfolio's withdrawal rate within a target band.

The Guyton-Klinger decision rules (published 2006) define guardrails:

• Upper guardrail: if the current withdrawal rate falls below 80 percent of the initial 4 percent (i.e., the portfolio has grown such that $40K is now only 3.2 percent of the balance), increase the withdrawal by 10 percent.
• Lower guardrail: if the current withdrawal rate rises above 120 percent of the initial 4 percent (i.e., the portfolio has dropped such that $40K is now 4.8 percent or more of the balance), cut the withdrawal by 10 percent.
• Inflation freeze rule: in any year following a portfolio drop greater than 10 percent, freeze the inflation adjustment.

Variable spending rules trade higher worst-case survival rates for lower expected withdrawal levels in difficult sequences. A retiree who started in 2000 and used Guyton-Klinger rules would have cut withdrawals by approximately 20 percent over 2001-2003 and again in 2008-2009, reducing the cumulative withdrawal stream but preserving the portfolio for the recovery years. Studies show these rules can extend portfolio survival from 30 years (failure at 4 percent static) to 35 to 40 years (with Guyton-Klinger) in the worst historical sequences.

Defense 3: delayed Social Security to age 70

Each year of delaying Social Security past full retirement age adds 8 percent to the lifetime benefit, up to age 70. A worker with a $2,500/month benefit at FRA 67 receives $3,100/month at age 70 (24 percent more). The 24 percent increase compounds throughout the rest of retirement, permanently reducing portfolio withdrawal pressure.

Worked example: a 65-year-old couple with $1M and $50K combined Social Security at FRA. If they delay to 70, the SS benefit rises to $66K. Their portfolio withdrawal need drops from $40K to $24K to fund the same $90K total spending. A 30 percent equity drawdown in year 1 now affects a $24K withdrawal stream, not a $40K stream, reducing sequence-of-returns damage by approximately 40 percent.

The trade-off: delaying SS to 70 means living on portfolio withdrawals for years 65-70. Those bridge years may concentrate sequence-of-returns risk if equities crash early. The defense against the bridge-year risk is the 2-3 year cash buffer, which can fully fund bridge spending without touching equities. The bridge-year strategy is one of the most efficient retirement planning levers available because it converts portfolio assets (volatile) into guaranteed inflation-indexed income (Social Security delayed credits) at a 6 to 8 percent annual rate.

Defense 4: partial annuitization for the essential expense floor

A SPIA covering essential expenses removes the portion of retirement spending from portfolio risk entirely. If essential expenses are $60K and Social Security covers $48K, an annuity premium of approximately $150K can lock in the remaining $12K floor. The portfolio risk now applies only to discretionary spending.

For a retiree with $1M total, $150K to a SPIA reduces the portfolio-managed amount to $850K. The withdrawal rate from the portfolio also drops: discretionary spending of $24K from $850K is 2.8 percent, not 4 percent. Lower withdrawal rates are dramatically less vulnerable to sequence-of-returns risk; at 2.8 percent, the portfolio essentially never fails over 30 years even in the worst historical sequences.

Combined with delayed Social Security (Defense 3) and a cash buffer (Defense 1), partial annuitization can reduce the effective portfolio withdrawal rate to 1.5 to 2.5 percent, putting the portfolio in a regime where sequence-of-returns risk is functionally eliminated.

The Roth conversion silver lining in drawdowns

Drawdowns in the first 5 years of retirement, while painful for portfolio survival, are simultaneously the best time to execute Roth conversions. When equity values drop 30 percent, converting $50K of traditional IRA to Roth captures 30 percent more shares at the lower tax cost. When equities recover, those shares grow tax-free in the Roth.

Worked example: a 67-year-old has $200K of traditional IRA equities. After a 30 percent drawdown, the same shares are worth $140K. Converting $50K creates a Roth holding worth $50K with a tax cost of $11,000 (22 percent bracket). When equities recover to pre-drawdown levels, the $50K Roth holding is worth $71K, all tax-free forever. The same conversion before the drawdown would have created a smaller share count for the same $50K conversion amount.

This is one of the few times when drawdowns can be turned to the retiree's advantage. The discipline required: do not panic-sell during the drawdown; instead, opportunistically convert into Roth at the depressed values.

Decision checklist for the first 5 years

• Hold 2 to 3 years of spending in cash equivalents at retirement. Refill the buffer in years when equities perform well; do not deplete it to chase yield.
• Set variable spending rules in advance. Write down the Guyton-Klinger guardrails. Implement the spending cut without negotiation when the trigger fires.
• Delay Social Security to age 70 if you have other resources to fund the 65-70 bridge years. The 8 percent annual delayed retirement credit is one of the best guaranteed returns available.
• Consider partial annuitization if the income gap between Social Security and essential expenses is meaningful. A small SPIA can shrink the portfolio-managed portion enough to materially reduce sequence-of-returns vulnerability.
• Convert into Roth during drawdowns. When equities are down 20 to 30 percent, increase Roth conversion activity rather than reducing it. The math favors conversions at low prices.
• Do not sell equities to rebalance during drawdowns. The 3-bucket strategy refill rule says: in negative-return years, refill cash from bonds, not from equities. Stick with the rule even when market headlines are bad.
• Review the plan annually, not the portfolio. Looking at portfolio balances during a drawdown triggers behavioral failure. Looking at the plan (the rules) reinforces discipline.

Key takeaways

• Sequence-of-returns risk means the order of returns matters during decumulation even with identical average returns. A 30 percent drawdown in year 1 of retirement with 4 percent withdrawals can cut 30-year portfolio survival from 95 percent to 60 percent.
• The risk concentrates in the first 5 years of retirement because the portfolio is at peak dollar size, the withdrawal stream is at longest duration, and no recovery cushion has accumulated. Drawdowns later in retirement have far less impact.
• Historical failure clusters (1929-1932, 1937-1939, 1965-1973) shared bad equity returns combined with high inflation in years 1-5 of the retirement start. Modern challenging periods (2000-2003, 2007-2009, 2022) had similar sequence patterns but lower inflation.
• Defense 1: cash buffer of 2-3 years of spending. Reduces worst-case portfolio failure rates from approximately 30 percent (no buffer) to 10-15 percent (3-year buffer).
• Defense 2: variable spending rules (Guyton-Klinger guardrails) that cut withdrawals after large drawdowns and freeze inflation adjustments in negative years.
• Defense 3: delayed Social Security to age 70 (8 percent annual delayed retirement credit, up to 32 percent lifetime benefit increase). Permanently reduces portfolio withdrawal pressure.
• Defense 4: partial annuitization for the essential expense floor. Reduces the portfolio-managed portion of spending, lowering effective withdrawal rate to 1.5-2.5 percent where sequence-of-returns risk is functionally eliminated.
• Silver lining: drawdowns in the first 5 years are simultaneously the best time to execute Roth conversions. Convert at depressed values; benefit from tax-free recovery.