Sequence-of-Returns Risk: The Decade That Decides It All

Two retirees, identical portfolios, identical contributions, identical average annual return over their retirement. One ends up rich; the other runs out of money in their 80s. The difference is something the financial industry calls "sequence-of-returns risk" — and it might be the single most underappreciated concept in retirement planning.

This is Part 5 of 6 of Retirement Essentials. Once you understand sequence risk, you'll see why everything in Part 4 about glide paths and bond tents is structured the way it is. It's all designed around this one problem.

The Counterintuitive Maths

Picture two retirees, Diana and Eduardo. Each retires at 65 with €1,000,000. Each withdraws €40,000 per year (the 4% rule). Each holds the same 60/40 portfolio, with the same average return of 6% over 30 years.

The only difference: the order of their returns.

• Diana gets her bad years early (years 1–3: -15%, -20%, -10%, then recovery)
• Eduardo gets his bad years late (years 28–30: same drops, but at the end)

Average return for both: 6%. Same arithmetic mean. Same set of returns in the dataset.

What happens to their portfolios:

| Year | Diana (bad first) | Eduardo (bad last) | |---|---|---| | 1 | €801,000 | €1,019,300 | | 5 | €620,000 | €1,180,000 | | 10 | €540,000 | €1,420,000 | | 20 | €310,000 | €1,920,000 | | 30 | Portfolio exhausted around year 24 | €1,200,000 leftover |

Diana runs out of money. Eduardo dies with €1.2M unspent. Same average return, same withdrawals, opposite outcomes.

This is sequence-of-returns risk in its purest form. The order of returns matters enormously when you're withdrawing from a portfolio, even though it matters not at all when you're accumulating.

Why Order Matters During Withdrawal (But Not During Accumulation)

The asymmetry comes from withdrawals. During the accumulation phase, you're adding money — bad early years are great for you because you're buying cheap. During the withdrawal phase, you're selling — bad early years mean you're selling more shares to fund the same withdrawal, permanently shrinking the share count.

A 30% drawdown in year 1 of retirement means selling shares from a smaller pool to cover that year's spending. By the time the market recovers, you've sold many more shares than you would have at the higher prices. The recovery happens with fewer shares riding it. The portfolio never catches back up.

This is also why sequence risk doesn't apply to the accumulation phase. If the market drops 40% the year you turn 30, you're crying into your statement but actually thrilled — your monthly contributions are buying twice as many shares. By age 60, you'll likely have more money than if returns had been smooth.

Sequence risk is specifically a retirement-era problem.

The "Fragility Zone" — Roughly Age 60 to 75

Sequence risk isn't uniform across retirement. It's heavily front-loaded. The first 10 years matter far more than the last 10. There's a specific period — roughly the 5 years before retirement to 10 years after — where the portfolio is largest, the time horizon is shortest, and a market crash does the most damage.

This is the "fragility zone." If you can get through it without a major drawdown coincident with the start of withdrawals, the rest of retirement is comparatively safe. Markets recover. Time heals. But a crash in years 1–5 of retirement, combined with steady withdrawals, can do damage no later recovery can undo.

This is why our glide-path advice in Part 4 emphasised the "bond tent": deliberately over-protecting the fragility zone, then allowing risk back in once the danger has passed. You're not trying to protect against every crash; you're trying to protect against the one crash that can sink the whole plan.

The Real-World Examples That Made This Concept Famous

Two cohorts of retirees made sequence-of-returns risk vividly real:

1966 retirees

People who retired in 1966 entered one of the worst 15-year stretches in US market history. Real returns were essentially flat from 1966 to 1982 — the "lost decade and a half." Anyone who retired in 1966 with the standard advice of the time (and not enough cushion) often ran out of money in the 1980s, despite the bull market of 1982 onwards.

2000 retirees

People who retired right at the dot-com peak in 2000 faced the same problem on a steeper scale. Two crashes (2000–2002 and 2007–2009) within nine years of retirement. Anyone withdrawing 4% who didn't adapt their spending found themselves with materially less wealth a decade later than identical retirees who started in 2010.

In both cases, the average 30-year return was acceptable — but the sequence was brutal. Anyone who looked only at long-term averages would have walked off a cliff.

What This Means for Your Planning

Three practical implications:

1. Stress-test your plan against bad sequences, not just averages

The biggest mistake in retirement planning is using a simple "what if I get 6% per year for 30 years?" projection. Reality is bumpy. Use the Monte Carlo simulator instead — it runs your plan through thousands of different return sequences, including some genuinely bad ones, and tells you the probability of success rather than a single deterministic number.

A plan that's "fine on average" but fails 20% of the time is a plan that has a 1-in-5 chance of running out of money. That's worse than most people would accept on the way in, even if the headline number looks good.

2. Build a cash buffer for the fragility zone

The "bond tent" from Part 4 is one approach. Another is a dedicated cash bucket of 1–3 years of expenses, held separately from your equity portfolio. The purpose: during a market crash in early retirement, withdraw from cash, not from stocks. Wait for stocks to recover, then refill the cash bucket from rebalancing gains.

A standard implementation: 12–24 months of expenses in short-duration bonds or cash equivalents, refilled annually from portfolio rebalancing. We cover variants of this in detail in Part 6.

3. Be prepared to flex spending

Sequence risk research consistently shows that retirees who modestly flex their withdrawals during downturns (cutting 5–10% in years following a market drop) materially improve their portfolio survival rate. This is the basis for "guardrail" withdrawal strategies, also covered in Part 6.

This isn't extreme deprivation — it's the difference between drawing €40,000 versus €37,000 in a bad year. But that small flex, applied during the fragility zone, can be the difference between running out of money at 82 and dying with €500,000 unspent.

A Useful Mental Model: "Two Portfolios"

A way to think about your retirement wealth: split it mentally into two pieces.

• Survival portfolio — the amount needed to cover essential expenses through your expected lifespan. This piece should be very conservatively invested. Bonds, annuities, guaranteed income. Sequence risk is a threat to this piece.
• Aspiration portfolio — everything above that. Travel, gifting, legacy, large discretionary purchases. This piece can be invested more aggressively, because if it underperforms, you cut discretionary spending and life continues. Sequence risk is an inconvenience for this piece, not a survival threat.

This framing matters because the "right" allocation for the whole portfolio is the weighted average of the right allocation for each piece. A retiree whose essential expenses are mostly covered by state pension and a small annuity has very little survival-portfolio risk and can hold equities aggressively in the aspiration portfolio.

A retiree with no pensions and high spending needs has a much larger survival portfolio share and needs more conservatism overall.

What the Best Research Actually Says

A few empirical findings worth knowing:

• Cash buckets work in expectation but their advantage is small. They mostly help by reducing the behavioural temptation to sell stocks at the bottom — a real benefit even if the maths is marginal.
• The combination of low initial withdrawal rate + spending flexibility is the most robust defence against sequence risk. Better than complicated portfolio engineering.
• The "bond tent" works but the effect size is modest — perhaps 5–10% improvement in worst-case outcomes. Worth doing, not worth obsessing over.

The single largest lever, by a wide margin, remains starting with a conservative withdrawal rate (3.0–3.5%) and being willing to flex it.

Action Items for This Week

1. 1Run a Monte Carlo simulation with your real numbers. Use a 30-year horizon and your planned withdrawal rate. Look at the failure probability, not just the median outcome.
2. 2Stress-test a bad sequence. What happens if your portfolio drops 30% in your first year of retirement? How much can you afford to cut spending?
3. 3Plan your cash bucket. Decide whether you'll use a separate cash buffer in retirement and, if so, what size. 12–24 months is the standard range.
4. 4Write your "guardrail" rule. "If my portfolio drops X%, I will reduce spending by Y%." Having the rule pre-written makes the in-the-moment decision much easier.

Next week, in the final installment, we'll cover withdrawal strategies in concrete detail — bucket strategy, bond tents, dynamic withdrawal rules, and how to actually spend down a retirement portfolio without running out.