Stock markets have been more volatile than anyone would like, and the advice for retirees has arrived in matching shape. Hold up to two years of portfolio withdrawals in cash. Cut stock exposure to roughly 30% as retirement begins, then let it climb back toward 60% over the next three decades. Diversify away from a cap-weighted index in which technology and communications make up more than 40% of the market. That is the package Barron's laid out for retirees in the current environment, as syndicated by Mint from the reporting of Elizabeth O'Brien. Each recommendation is reasonable, and each is a response to volatility. The shared premise deserves a closer look, because the number at the center of the conversation, the volatility figure itself, is built to be blind to the risk that breaks retirements.

Volatility, as it is measured and quoted, is symmetric. A 15% decline and a 15% gain contribute the same amount to the statistic; they count as equal events. For an investor in the saving years, buying on a schedule and never touching the money, that symmetry is roughly fair. For a retiree, it is not. A retiree does not hold a portfolio; a retiree spends one, and the spending changes everything.

The number treats every direction alike

The purpose of a volatility figure is to describe how much a portfolio moves. The trouble is that for someone making withdrawals, a move down and a move up are not equivalents. A down year that coincides with a withdrawal forces the sale of shares at a low price, and that sale is permanent: the shares are gone, the low is locked in, and no later rally can restore what was sold. An up year brings no comparable benefit; it simply means the retiree sells fewer shares at a higher price. Morningstar's illustration makes the whole argument in miniature: with a withdrawal taken each year, the same three returns leave a smaller balance when the decline comes first than when it comes last. Same numbers, different order, different outcomes. The volatility of the two sequences is identical.

Wade Pfau's simulations make the same point with a wider lens. Holding the underlying portfolio fixed, he found that the spread of possible outcomes widens as a saver moves from holding a lump sum to building a portfolio to drawing one down, with the standard deviation of outcomes climbing while the portfolio's own volatility never changed. The portfolio was identical. The danger grew because withdrawals were added to the mix, and the risk that grew is invisible to the number every article quotes. Pfau's analysis, published in Forbes, puts the specific movement at roughly 1.9% for a lump sum rising to about 2.5% in retirement, on an identical portfolio.

The order is the risk, and it never shows in the math

What separates a retirement that works from one that runs out of money is the order in which returns arrive, interacting with the schedule of withdrawals. Morningstar's Amy Arnott has noted that serious negative sequences, two or more consecutive down years, have occurred only four times since 1926: 1929-32, 1939-41, 1973-74, and 2000-02. Christine Benz of Morningstar describes the underlying hazard plainly: encountering a very bad market at the beginning of retirement leaves less of the portfolio in place to recover when markets rebound. The rarity of such sequences is cold comfort, because a single one is enough to set a retirement permanently back.

The arithmetic is stark. A retiree who entered 2000 with a million dollars in an all-equity portfolio and withdrew $40,000 a year, adjusted for inflation, saw the actual order of the following two decades leave the portfolio near $600,000. Run the same returns in the opposite order, and the portfolio is left near $2 million. Same average return, same volatility, outcomes an ocean apart. The difference between those two retirements is not captured by any risk statistic a retirement portfolio is typically measured with. It lives in the sequence.

Cash and the glide path are spending tools, not volatility tools

This is why the defensive advice, sensible as it is, is usually aimed at the wrong enemy. The two-year cash buffer is not a volatility-reduction tool. Holding two years of withdrawals in cash barely moves the standard deviation of an overall portfolio. It is a spending tool: its job is to keep withdrawals from being forced at bad prices, so the retiree draws from a pool that does not fluctuate while the fluctuating part of the portfolio recovers. The same logic applies to the glide path. Cutting stocks to 30% at retirement and rising toward 60% later is a statement about which years' spending is exposed to the market. It protects the early years, when sequence risk is highest, and accepts more market risk later, when the portfolio has been drawn down enough that a crash does less dollar damage. Both tools work on the interaction between withdrawals and returns. Neither registers in the volatility figure the conversation is organized around.

The real instruments never appear in the chart

Once the object is the spending policy rather than the portfolio, the tools that matter come into view, and none of them is a volatility statistic. The first is spending flexibility. A plan with room to cut discretionary spending in bad years produces a better distribution of outcomes than one with fixed spending, because flexible spending does not force sales at the worst moment. Morningstar research also suggests most retirees naturally reduce their real spending over time, with David Blanchett estimating real spending declines on the order of 20% to 30% by age 85, which means a plan that assumes a flat spending line adds phantom risk of its own.

The second is guaranteed income. Social Security, pensions, and immediate annuities are the only instruments that convert market risk into something else: they pay a contractual amount that does not depend on the market's path. A retiree with enough guaranteed income to cover essential spending has removed the worst tail of their own distribution, whatever the volatility number says. The third is the rule for which dollars get spent in which conditions, the bucket approach, which is not an allocation strategy at all but a withdrawal rule wearing an allocation costume.

None of these appears in the risk measures that dominate retirement writing. You can read a dozen articles about volatility without ever seeing the withdrawal rate, which is the single most powerful lever a retiree controls. The withdrawal rate and its flexibility, not the standard deviation of the portfolio, are what set the odds of running out of money. Morningstar's own research, which now suggests a 3.3% safe withdrawal rate for a 50/50 portfolio aiming to last 30 years, is an argument about spending, published in the language of markets.

Reading the market commentary differently

The practical consequence is a change in how to read the commentary. A headline about volatility is a description of the market. It is not a description of any particular retirement, because the same market, measured by the same volatility, produces a wide range of retirement outcomes depending on what the retiree does with it. The honest question is not "how much will my portfolio move?" It is "what will I spend when it moves?" The first is the number that gets quoted. The second is the number that determines the outcome.

One piece of the standard volatility advice deserves a special word, because it is aimed at substance rather than at the metric. The warning that a cap-weighted index now carries a heavy concentration of technology stocks, and that newly listed megacap companies will keep flowing into index funds, is not about how much the market moves. It is about whether the diversified portfolio the investor believes they hold is diversified at all. That warning holds regardless of the volatility figure, and acting on it makes sense. But it is a warning about composition, and it should not be mistaken for a comment about volatility.

The volatility number is not wrong. It measures what it says it measures: how much the market moves. The mistake is in thinking that is the same thing as how much the retirement is at risk. Retirement risk is a joint product of the market's path and the spending plan that runs on top of it, and only the first of those is ever measured or quoted. The tools that protect a retiree, the cash buffer used as a spending rule, the flexibility to spend less in bad years, and the guaranteed income that does not depend on the market at all, are invisible to the statistic the discussion is built around. The retiree who sees that will stop asking how bad the volatility is, and start asking what their spending does when it arrives. That is the question the numbers answer, even though no volatility chart will ever show it.

Primary sources

  1. Barron's, as syndicated by Mint in the reporting of Elizabeth O'Brien, for the framing that stock volatility is the central problem for retirement portfolios and for the specific guidance reported there, including holding up to two years of withdrawals in cash, cutting stocks to roughly 30% at retirement and rising toward 60% by year 30, diversifying beyond the S&P 500, and the worked example in which a retiree starting with $1 million and taking $50,000 in inflation-adjusted withdrawals who suffers 15% losses in the first two years runs out of money after about 18 years, while the same losses in years 10 and 11 leave about $400,000.
  2. Morningstar for the demonstration that the order of returns determines outcomes once withdrawals begin, for Amy Arnott's observation that serious negative sequences occurred only four times since 1926, for Christine Benz's description of sequence risk and the bucket approach, for David Blanchett's finding that real retirement spending declines roughly 20% to 30% by age 85, and for Morningstar's research suggesting a 3.3% safe withdrawal rate for a 50/50 portfolio.
  3. Wade Pfau's Forbes column for the Monte Carlo finding that the standard deviation of outcomes widens, from about 1.9% to about 2.5%, when an identical portfolio moves from lump-sum holding to retirement withdrawal.