Advanced Finance

Sequence of Returns Risk: The Retirement Killer Nobody Talks About

5 min read

The Illusion of Average Returns

Most retirement planning uses average annual returns — plug in 7%, project 30 years, and arrive at a comfortable number. But averages are deeply misleading when you are withdrawing money rather than adding it. In decumulation, the order in which returns arrive matters dramatically more than the average.

How Sequence Risk Works

Imagine two retirees, each with $1 million and each withdrawing $40,000 per year (4% rule). Both earn an average annual return of 6% over 25 years, but the return patterns differ.

Retiree A earns: +15%, +18%, +12%, +8%, +5%, then averages 4% for the remaining years. By year 25, they have roughly $380,000 remaining — comfortable.

Retiree B earns: -20%, -15%, -5%, +2%, +10%, then averages 9% for the remaining years. The average is the same 6%, but Retiree B runs out of money in year 17.

Same average return, same withdrawal rate, radically different outcomes. Early losses force you to sell more shares at low prices to meet withdrawal needs, leaving fewer assets to participate in the eventual recovery. This is sequence risk.

The Mathematics of Irreversible Damage

When you add money (accumulation phase), bad early returns are actually beneficial — you buy cheap shares. When you withdraw money (decumulation), bad early returns compound in the wrong direction.

A 20% loss in year one of retirement requires a 25% gain just to get back to even. But during that recovery, you are still withdrawing. If you withdraw 4% of the original balance each year, a 20% loss plus a 4% withdrawal means your portfolio is down 24% before any recovery begins. The hole grows faster than you think.

Research by Wade Pfau and Michael Kitces shows that the first 5 to 10 years of retirement overwhelmingly determine whether a portfolio survives a 30-year retirement. If returns over that window are merely average or better, almost any reasonable withdrawal rate works. If they are poor — particularly if inflation is high simultaneously, as in the late 1960s — even a 4% withdrawal rate can fail.

SORR Mitigation Strategies

The Cash Wedge

Hold 2–3 years of spending needs in cash or short-term bonds. When markets drop, draw from cash instead of selling depressed assets. Refill the wedge when markets recover. This breaks the forced-selling-at-lows dynamic.

Dynamic Withdrawal Rules

Rather than blindly withdrawing 4% plus inflation, use guardrails. For example: if your portfolio falls 15% below its starting value, skip the inflation adjustment. If it falls 25%, cut spending by 10%. A small spending adjustment early prevents a catastrophic one later.

Bond Tents

Increase your bond allocation in the years immediately surrounding retirement (the "retirement red zone"), then gradually increase equity exposure later. This protects against sequence risk during the most vulnerable window while still providing long-term growth. A typical glidepath: 60/40 at age 60, shift to 40/60 by retirement at 65, then glide back to 60/40 by age 75.

Partial Annuities

Using a portion of assets to purchase an immediate income annuity converts market risk into longevity risk borne by an insurance company. A floor of guaranteed income covering essential expenses reduces the pressure on the remaining portfolio, making sequence risk less threatening. This is not an all-or-nothing decision — even annuitizing 20–30% of assets can meaningfully lower the probability of ruin.

Variable Spending

The most powerful lever is simply spending less after bad years and more after good ones. Retirees who can reduce discretionary spending by 15–20% during bear markets dramatically improve portfolio survival odds — far more than any asset allocation tweak can accomplish.

The Bottom Line

Sequence risk is the single most underappreciated risk in retirement planning. The tools to manage it exist — cash reserves, dynamic withdrawals, bond tents, annuities, and spending flexibility. What matters is having a plan before the rug gets pulled, because mid-crash panic is the worst time to design one.