Sequence of returns risk

In short

You have done everything right, saved for decades, and built a portfolio that should carry you through retirement. Then the market falls twenty or thirty percent in your first year. That is sequence of returns risk. Your withdrawals lock in the loss and leave you with less principal to ride any recovery. A retirement account that weathers a storm late feels very little pain; the same storm early can hollow out decades of security. You cannot control the market, but you can lower what you pull out in lean years, hold enough cash to avoid selling in a trough, and run the numbers again when volatility arrives. Knowing the danger exists gives you room to plan before it hits.

The whole of it

What it is

I once listened to two men at a coffee counter who had both retired in the same month with almost identical balances. Ten years later, one was comfortable and the other was rationing groceries. The market had done the same loop for both of them, but one had retired in 2007 and the other in 2009. The order of the gains and losses made all the difference.

Sequence of returns risk is the danger that poor market performance early in retirement will damage your portfolio beyond repair, even if average returns over the full period look perfectly fine. While you are saving, a bad year early and a bad year late produce the same ending balance because you keep adding money. While you are withdrawing, order matters. A fall in year one forces you to sell more shares to meet the same dollar need, and those shares are gone when the recovery comes. A fall in year twenty barely dents a plan because you have already spent most of what you needed and the remainder can wait.

The risk lives in the gap between arithmetic and real life. A portfolio that earns ten percent one year and loses eight percent the next has averaged a one percent gain, but that average hides the damage done when the loss comes first and you are pulling out cash at the same time.

How it works

You retire with five hundred thousand dollars and plan to withdraw thirty thousand a year, adjusted for inflation. If the market rises steadily, every withdrawal leaves the remainder to grow. If the market falls twenty percent in year one, you have four hundred thousand left before the withdrawal. You take out your thirty thousand. That leaves three hundred seventy thousand. When the market rebounds, it multiplies a smaller base. The percentage gain looks healthy on a chart, but the dollar recovery does not replace what you sold.

Imagine two thirty year retirements, both averaging a six percent annual return but in reverse order. Retirement A sees losses clustered in years one through five. Retirement B sees those same losses in years twenty six through thirty. Retirement B sails through; the early gains compound on the full principal, and by the time the downturn arrives, the retiree has already taken most of the planned income. Retirement A stumbles. Withdrawals in falling markets lock in losses, the portfolio shrinks faster than arithmetic predicts, and the retiree may run out of money while the historical average still looks generous.

The damage is not the volatility itself. Volatility during accumulation smooths out over time and can even help through dollar cost averaging. The damage is volatility during distribution, when every withdrawal is irreversible.

The numbers, and where to find yours

There is no official limit or threshold published by a tax agency because sequence of returns risk is a mathematical fact, not a regulatory event. The usual way to measure it is to model your planned withdrawal rate against historical market sequences and count how many starting years would have exhausted the portfolio before thirty years elapsed.

A common reference is the four percent rule, which emerged from research testing withdrawal rates against rolling historical periods. To find your own number, take your total retirement savings and multiply by your planned first year withdrawal rate. Then model that withdrawal, growing by the current figure, which the official source publishes each year each year, against a range of return sequences. Software such as FireCalc or the portfolio analysis tools at Vanguard and Fidelity let you test your plan against every historical thirty year window since 1926. If your plan survives ninety percent of sequences, you have a ten percent sequence risk. If it survives only seventy percent, your risk is thirty percent.

Your specific exposure depends on four things: your withdrawal rate, your asset allocation, the length of your retirement, and how willing you are to adjust spending when markets fall.

A worked example

Ellen retired in January with eight hundred thousand dollars, seventy percent in stock index funds and thirty percent in bonds. She planned to withdraw forty thousand dollars in year one, then increase that amount by three percent each year. Her withdrawal rate is five percent.

In her first year the stock market fell eighteen percent and bonds fell four percent. Her stocks dropped from five hundred sixty thousand to four hundred fifty nine thousand two hundred. Her bonds dropped from two hundred forty thousand to two hundred thirty thousand four hundred. Her total portfolio stood at six hundred eighty nine thousand six hundred before any withdrawal. She took out forty thousand, leaving six hundred forty nine thousand six hundred.

In year two the market rose twelve percent, lifting her portfolio to seven hundred twenty seven thousand five hundred fifty two. She withdrew forty one thousand two hundred, which is forty thousand increased by three percent. That left her with six hundred eighty six thousand three hundred fifty two.

If those same two years had occurred in reverse order, Ellen would have started year one with eight hundred thousand. A twelve percent gain would have brought her to eight hundred ninety six thousand. After withdrawing forty thousand, she would have had eight hundred fifty six thousand. In year two, the eighteen percent stock loss and four percent bond loss would have reduced that to seven hundred thirty seven thousand nine hundred ninety two. After the forty one thousand two hundred withdrawal, she would have had six hundred ninety six thousand seven hundred ninety two. That is ten thousand dollars more than the path where the loss came first. The returns were identical. The order carved out the difference.

Over thirty years, if Ellen faces several poor years early, she may deplete the portfolio by year twenty even if the average annual return matches the historical mean. If the same poor years arrive late, she finishes with money to spare.

Where it goes wrong

A friend of mine read about sequence risk and decided to hold everything in cash until the danger passed. He missed seven years of gains waiting for safety that never arrives, because you cannot know which year is dangerous until it is over. Overreacting to sequence risk is as harmful as ignoring it.

The most common mistake is assuming that a safe withdrawal rate is safe every year regardless of market behavior. A four percent rule is a starting point, not a lifetime contract. If you withdraw four percent like clockwork while your portfolio falls thirty percent, you are compounding the sequence damage. Flexibility is the countermeasure. Trim spending in down years, delay a large purchase, let inflation adjustments skip a year.

Another error is holding too little in stable assets. If your entire portfolio sits in stocks and you retire into a bear market, you have no cushion. A cash reserve covering two or three years of expenses lets you avoid selling equities at the bottom. Some planners call this a bucket strategy: one bucket of cash, one of bonds, one of stocks. You spend the cash, refill it from bonds in flat years, and refill it from stocks after gains.

A third mistake is testing your plan against average returns instead of actual sequences. An average tells you nothing about order. Monte Carlo simulations and historical backtests show you what happens when bad luck lands early.

Finally, many people believe sequence risk ends after the first five years. The highest danger is in the first decade, but the risk persists whenever you are net withdrawing. A second bear market in year twelve, if your portfolio has not fully recovered, can restart the damage.

Where it goes wrong on the other side is paralysis. You can model a thousand scenarios and never retire, convinced that the perfect moment will reveal itself. It will not. The remedy is a plan that bends, enough stable assets to weather three bad years, and the willingness to spend four percent in good times and three percent in bad.

Questions to answer before you leave this page

What is my withdrawal rate as a percentage of today's balance, and does my plan survive eighty percent of historical thirty year sequences starting in years like this one? How many years of expenses do I hold in cash or bonds, and is that enough to avoid selling stocks during a twenty or thirty percent decline? If the market falls fifteen percent in my first year of retirement, will I cut my withdrawal, delay a large expense, or sell anyway? Have I tested my plan against real return sequences, or only against an average I found in an article? Do I know the difference between a withdrawal rate that worked historically and one that will work in my specific future, and have I left room to adjust when that future refuses to cooperate?