Key takeaways
- Once withdrawals begin, a bad market in the early years does permanent damage, because you sell shares low that never get to recover.
- The same returns in a different order produce a different outcome; the average is not the story.
- The defenses are a bond allocation sized to several years of spending, flexible withdrawals, and a portfolio rebalanced rather than abandoned.
- Delaying Social Security and keeping a Roth reserve add flexibility exactly when it is needed.
Most people understand that markets go up and down and that a long-term investor rides it out. What changes at retirement is one detail: you're now selling. And selling during a decline does something that holding through a decline doesn't. It turns a temporary loss into a permanent one. That's sequence of returns risk, and it's why the first few years after retirement carry more weight than any other stretch of an investing life.
Same average, different outcome
Picture two retirees, each with $2 million, each withdrawing $80,000 a year adjusted for inflation, each earning the same set of annual returns over 25 years. One difference: the order is reversed.
Retiree A gets three bad years first, a 15% drop, then a 10% drop, then a flat year, followed by two decades of good returns. Retiree B gets the same three bad years at the very end, after twenty-two good ones.
The arithmetic average return is identical. Retiree B finishes with a comfortable surplus. Retiree A runs meaningfully short, and in a harsher version of the same example runs out of money. Why? A sold shares at depressed prices in years one through three to pay for groceries, and those shares were gone when the recovery came. B sold at high prices and had far more cushion when the decline finally showed up.
During the accumulation years, the order of returns doesn't matter at all; the ending balance is the same either way. It's withdrawal that creates the asymmetry.
Why the first years dominate
The damage from a bad early sequence compounds in two directions. The portfolio is smallest relative to the withdrawals it has to support, so each withdrawal is a bigger slice of what's left. And the recovery has to lift a smaller base over a longer period. A 30% decline in year one of a 30-year retirement, with withdrawals continuing, can cut the ending balance by half or more compared with the same decline in year fifteen.
This isn't a reason to panic. Most retirements don't begin with a crash. But the plan has to work if yours does, because you don't get to pick your start date's luck. Retirees who began in 2000 or 2008 learned this firsthand. Retirees who began in 2009 or 2020 (after March) never felt it.
The defenses
You can't eliminate sequence risk without giving up the returns that make retirement affordable. You can make it survivable, and there are several ways.
Bonds sized to your spending. Holding enough Treasury bonds, including inflation-protected Treasuries, to cover several years of planned withdrawals means a decline in stocks doesn't force stock sales to pay the bills. We size the bond portion of the portfolio that way, then manage the whole portfolio as one rebalanced allocation: each withdrawal comes from whatever is above its target, which in a bad year for stocks is the bonds. This is the sound core of the "bucket" idea, without the bookkeeping. We wrote about what's real and what's theater in the bucket strategy.
Flexible withdrawals. A plan that can cut spending by 10% in a bad year survives sequences that a rigid plan doesn't. Most retirees have discretionary spending that can flex, and knowing ahead of time what would get cut, and at what trigger, is what makes it possible to actually do it. Spending guardrails turn this into a rule.
Rebalancing, not retreating. The instinct in a decline is to sell stocks and wait. The evidence says rebalancing, meaning buying stocks when they've fallen to restore the target mix, is what repairs a portfolio after a bad sequence. The bonds are what make this possible. They're what you buy with.
An appropriate stock allocation. A retiree with 90% in stocks has more sequence risk than one with 60%. But a retiree with 20% in stocks has a different problem: a portfolio that may not keep up with thirty years of inflation. The allocation gets set by the spending need, with the bond portion sized to the withdrawals it has to cover, not by a feeling about the market. Some research supports a "rising equity glide path," starting retirement more conservative and adding stocks gradually as the early danger years pass.
Guaranteed income as a floor. Social Security, a pension, and in some cases a simple income annuity cover part of essential spending no matter what markets do. The bigger that floor, the less the portfolio has to produce in a bad year. Delaying Social Security to 70 raises the floor by 24% relative to claiming at 67, which is one reason the claiming decision belongs inside the sequence-risk conversation.
A Roth reserve. A withdrawal from a Roth IRA in a bad year costs no tax and doesn't raise Medicare premiums. Keeping some Roth balance for those years, instead of spending it first, adds a second kind of flexibility.
The years before retirement count too
Sequence risk starts before the retirement date. A 40% decline in the two years before you planned to retire has the same effect as one in the two years after, because it changes the balance the withdrawals will come from. So we start moving toward the retirement allocation three to five years before the planned date, moving money out of stocks in stages rather than all at once. It's also why "one more year" is a real option the plan should be able to price.
How we manage it
Every retirement income plan we build includes a bond allocation sized to the spending the household needs to cover through multiple years of a stock downturn, a rebalancing rule, a spending policy with defined triggers, and a stress test against bad early sequences using historical and simulated paths. We judge the plan on whether it survives the bad start, not on whether it does well with an average one. The rest of retirement income planning sits on top of that.
Frequently asked questions
How much of the portfolio should be in bonds?
Enough to cover the withdrawals the portfolio has to make, net of Social Security and pensions, through a long stock downturn. For most households we size it to about seven years of that spending, which is the longest stretch U.S. stocks have spent more than 20% below their prior peak after inflation. Households whose essential spending is already covered by guaranteed income can hold less, and those who can't cut spending at all hold more. More bonds reduce risk and reduce expected return, so the size follows the plan.
Does sequence risk apply if I only spend dividends?
Less directly, but dividends get cut in severe downturns, and a portfolio built to maximize dividend income usually gives up diversification and total return. A total-return approach with bonds sized to your spending handles the same problem more reliably.
Is an annuity the answer?
A simple, low-cost income annuity can be part of the answer for some retirees, because it converts a lump sum into a guaranteed floor. Complex indexed and variable annuities with high costs generally aren't. We have a separate guide on annuities from a fee-only view.
What if I retired into a bad market already?
The defenses still work: draw from the bonds, hold or rebalance the stocks, trim discretionary spending, and don't lock in losses. The first years are the most dangerous, and they're also when disciplined action matters most.
This article is general education, not individualized advice, and it is not tax, legal, or accounting advice. Figures reflect federal law for the year stated and can change. Valerio Wealth Management does not prepare or file tax returns. Speak with a qualified professional about your own situation before acting.
In this article
Want this applied to your own numbers? A 30-minute introductory call is the first step.
See if you’re a fit