Sequence-of-Returns Risk

If your portfolio earns 7% annually for 30 years but crashes 40% in year 1 while you're withdrawing, you'll run out of money. Same return, different sequence, catastrophic outcome.

Key takeaways

• Sequence-of-returns risk: the order of returns matters enormously when withdrawals are happening simultaneously.
• A portfolio that averages 7% annual return can fail if you experience negative years early (drawing on depleted capital) and positive years late (no capital to compound).
• The same portfolio with the same average return succeeds if returns are positive early and negative late.
• This risk is largest for retirees withdrawing from portfolios; smallest for accumulators reinvesting returns.
• Sequence risk is why retirees should not hold 100% stocks, even with 30-year horizons.

The coin-flip demonstration

Imagine two investors, each with £400,000, each withdrawing £16,000 per year (4% initially), each expecting 7% annual returns over 30 years.

Investor A: Bad sequence (crash first)

• Year 1: Market falls 25%. Portfolio: £400,000 → £300,000. Withdrawal: £16,000. Remaining: £284,000. Loss so far: £116,000. Withdrawal rate: 5.6% of starting capital (now unsustainable).
• Year 2: Market rises 10%. Portfolio: £284,000 → £312,400. Withdrawal: £16,000. Remaining: £296,400. Now withdrawing 5.4% of current capital.
• Years 3–30: Markets average 8% (to make up for the crash). But early years of withdrawing from a depleted portfolio are damaging. By year 20, the capital is exhausted.

Investor B: Good sequence (gains first)

• Year 1: Market rises 12%. Portfolio: £400,000 → £448,000. Withdrawal: £16,000. Remaining: £432,000. Withdrawal rate: 4.0% (still sustainable).
• Year 2: Market falls 25%. Portfolio: £432,000 → £324,000. Withdrawal: £16,000. Remaining: £308,000. Even with the crash, the portfolio held strong in year 1, so the loss is less damaging.
• Years 3–30: Markets average 8%. With a larger capital base after year 1, compounding wins. By year 30, the portfolio has £600,000+ remaining.

Same average return (7% nominal, roughly 4.5–5% real after adjusting for the large year 1 returns/losses). Investor A runs out of money; Investor B has £600,000 left.

The sequence of returns determined success or failure.

Why sequence matters during withdrawal

When you're withdrawing, two things happen simultaneously: your portfolio is earning returns and you're taking money out.

If returns are positive and withdrawals are small relative to the portfolio, you're fine. If returns are negative and you must keep withdrawing, you're forced to sell at the worst time (at the bottom).

Here's the mechanics:

Non-withdrawing investor (age 35, accumulating)

• Year 1: Starts with £100,000. Market crashes 40%. Balance: £60,000. Profit/loss so far: -£40,000.
• Year 2–10: Markets return 8% annually. Balance grows. By year 10, the year 1 crash is a rounding error because the portfolio has compounded for nine years.
• End of year 10: Starts with £100,000, experiences a 40% crash early but ends with £200,000 due to subsequent compounding. The sequence didn't matter much because time allowed recovery.

Withdrawing investor (age 65, early retiree, withdrawing £30,000/year from £400,000)

• Year 1: Starts with £400,000. Market crashes 40%. Balance: £240,000. Withdrawal: £30,000. Remaining: £210,000. Capital has been cut in half, and withdrawal continues.
• Year 2–30: Markets return 8% annually. But the base is now £210,000, not £400,000. Compounding on a much smaller number means the portfolio never catches back up.
• End of year 10: Portfolio is depleted. The withdrawal didn't work because sequence forced selling at the worst time.

The non-withdrawing investor can recover from a crash because time and compounding work. The withdrawing investor cannot, because they're selling during the crash to fund living expenses.

Historical sequence risk: 1929 and 2008

The Great Depression (1929–1939) An investor who retired in 1929 with £100,000 and withdrew 4% annually faced:

• 1929: Crash 89% (1929–1932 nadir)
• 1930–1932: Further decline
• 1933–1938: Partial recovery
• 1939–1954: Full recovery to 1929 nominal levels (but 20+ years later, with £4,000/year withdrawn, the portfolio was largely depleted)

An investor retiring in 1939 (after the crash) with the same £100,000 and same 4% withdrawal would have ended the decade of the 1940s wealthy. Sequence determined success or failure by 30+ years.

The 2008 Financial Crisis A retiree who retired in early 2008 with £500,000, withdrawing £20,000 per year (4%), faced:

• 2008: Market crashes 48%. Portfolio: £260,000. Withdrawal: £20,000. Remaining: £240,000.
• 2009: Market crashes further. New withdrawal from £240,000.
• 2010–2013: Market recovers. But the portfolio, now at £240,000, is compounding from a much smaller base.
• By 2013: The portfolio is still positive but significantly behind where it would have been if the crash had come in year 20 instead of year 1.

A retiree who retired in early 2009 (after the crash) with £500,000, withdrawing the same 4%, would have experienced gains in every single year from 2009 onwards, and the portfolio would have grown significantly.

Sequence risk: the difference between retiring at the peak and retiring at the trough.

How to quantify sequence risk

Researchers (primarily Vanguard and Morningstar) have studied historical rolling periods to measure sequence risk:

For a retiree withdrawing 4% annually from a 60/40 portfolio over 30 years:

• Success rate if retiring in "good" markets (pre-crash): 95%+ chance of portfolio survival
• Success rate if retiring in "bad" markets (at a peak or crash): 80–90% chance of portfolio survival
• The difference: 5–15% failure rate depends on when you retire

For a 50/50 portfolio:

• Good markets: 90%+ success
• Bad markets: 75–85% success

For a 40/60 portfolio:

• Good markets: 85–90% success
• Bad markets: 70–80% success

For a 30/70 portfolio:

• Good/bad markets: 85%+ success (less sensitive to sequence because bonds buffer crashes)

The more equities you hold, the more sequence risk you face as a retiree.

Why retirees need bonds

This is the mathematical argument for why a 65-year-old should not hold 100% stocks, even with a 30-year horizon:

Not because of risk capacity or time horizon, but because sequence risk transforms a long horizon into a dangerous present. A crash in year 1 of retirement, with withdrawals underway, is far more dangerous than a crash in year 20 because of forced selling.

Bonds serve as a buffer. When equities crash 40%, bonds are often flat or up, so withdrawals can be taken from bonds without forced stock liquidation. By the time equities have recovered, the retiree's needs might be met from other sources (Social Security, pension), and the stock allocation can stabilize.

A 60/40 retiree in 2008 could:

• Take £20,000 from the 40% bond allocation (which was mostly stable)
• Let the equity portion recover undisturbed for a few years
• By 2012–2013, rebalance back to 60/40 with the recovered equities

A 100/0 retiree in 2008 was forced to sell equities at the bottom to fund spending.

The bucketing solution

The best way to manage sequence risk is bucketing:

1. Bucket 1 (Years 0–3): £60,000 in cash/short-term bonds. Covers three years of £20,000 withdrawals. Zero equity risk.
2. Bucket 2 (Years 3–10): £140,000 in bonds/balanced. Covers seven years of withdrawals (after bucket 1 is drawn). Moderate equity risk.
3. Bucket 3 (Years 10+): £300,000 in equities. Long-term growth. High equity allocation.

In 2008:

• Bucket 1 is untouched (crisis doesn't matter).
• Bucket 2 provides withdrawal in 2008.
• Bucket 3 crashes but is not touched.
• By 2011, equities have partially recovered, and bucket 2 (now recovered) refunds some money to bucket 1 for 2011–2013 withdrawals.
• By 2015, equities are fully recovered and bucket 3 is growing again.

Sequence risk is eliminated because you're not withdrawing from equities during the crash—you're withdrawing from bonds, which are stable or recovering.

This is why the first chapter of a retiree's life (years 0–5) should be conservative in absolute terms (bonds and cash), even if overall portfolio is 70/30.

The sequence risk flowchart

Related concepts

• Short, Medium, Long Horizon Defined
• Why Horizon Shapes Allocation
• Volatility vs Permanent Loss

Next

Sequence-of-returns risk is driven by one thing: the timing of crashes relative to when you need money. We've accepted crashes as inevitable and built portfolios to weather them. But there's a subtler risk: perceived risk, shaped by recent performance. After years of gains, investors feel safe and take more risk. After crashes, they panic and reduce risk. That's when recency bias becomes dangerous.