The bond tent: allocation for sequence-of-returns risk
Two retirees can earn the exact same average return over thirty years and end up in completely different places — one comfortable, one broke. The difference is the order the returns arrive in. This is sequence-of-returns risk, and the “bond tent” is one of the best-known ways to defend against it.
Why the first years are the dangerous ones
While you are saving, a crash early in your career barely matters — there is little money at stake, and every paycheck buys shares on sale. Once you start withdrawing, the math flips. Selling shares into a falling market converts a temporary paper loss into a permanent one: the shares you sold to fund this year’s spending are not there to recover when the market does. A bad decade at the start of retirement can sink a plan that the same decade, arriving twenty years later, would barely dent.
Consider someone who retired at the start of 2008 heavily in stocks. U.S. stocks fell about 37% that year (the S&P 500 with dividends returned −36.55% in the historical series this simulator samples from). Add a 4% withdrawal on top and the portfolio ended the year down roughly 40% — needing a two-thirds gain just to get back to even, while withdrawals kept coming. Markets did recover, but a portfolio that is simultaneously shrinking from withdrawals recovers far more slowly than the index does.
The bond tent: a rising equity glidepath
Conventional wisdom says stock exposure should drift steadily down as you age. Research by Wade Pfau and Michael Kitces found something more interesting: in historical and simulated retirements, plans often did better when equity exposure was reduced into retirement and then rose again during retirement — for example, gliding down to something like 30–40% stocks at the retirement date, then drifting back up toward 60–70% over the following ten to fifteen years.
Plotted over a lifetime, bond allocation rises into the retirement date and falls after it — a tent shape, hence the name. The logic: the years just before and just after retirement (Kitces calls this the “retirement red zone”) are when your portfolio is biggest and your remaining flexibility smallest, so a crash hurts most. Holding more bonds through exactly that window means bad early years are absorbed by the bond side while you spend from it. If the early years turn out fine, the danger has largely passed — the portfolio is comfortably ahead of schedule — and re-raising stock exposure lets the survivors’ plans keep growing through a long retirement and fight inflation.
What it costs
The bond tent is insurance, and insurance has a premium. Bonds have crashes of their own — 2022 hurt both sides of the tent at once — and a heavier bond allocation held for decades gives up real growth. In the many scenarios where markets do well early in retirement, a conservative allocation drags returns, and the median outcome is usually somewhat lower than staying stock-heavy throughout. The trade is fewer catastrophic left-tail outcomes in exchange for a more modest middle. Whether that trade appeals to you depends on how much margin your plan has — which is exactly the kind of question a Monte Carlo simulation is built to answer.
Try it in Deorbit Plan
Open the Market assumptions panel. The “glide to at retirement” control models the front half of the tent — stocks drift from today’s mix to a chosen retirement mix over the years you pick. Expand “Post-retirement glide (bond tent)” to add the rising back half: a second glide from the retirement mix up to a new target over a chosen number of years. Load your current plan as scenario A, add the tent as scenario B, and compare success rates and the spread of outcomes — especially the worst-decile paths, where the tent earns its keep.
Educational content only — not financial, tax, or investment advice.
Read this article in the interactive simulator →