Retirement planning conversations tend to focus on accumulation — how much you save, which accounts you use, when you hit your number. That framing makes sense during the working years, but it quietly sets people up for the harder problem: what happens on the other side of that number. Decumulation — the art of drawing down a portfolio in a way that doesn’t run out before you do — is a fundamentally different discipline than accumulation, and the rules that made you successful getting here don’t automatically apply once you’ve arrived.
The distinction matters more in 2026 than it has in a long time, for reasons that are specific to this moment. Equity valuations remain elevated after a three-year bull run, the Federal Reserve is holding rates steady with no clear path to cuts, inflation from tariffs and energy prices is keeping the cost-of-living adjustment from working the way retirees expect, and the legislative landscape — from SECURE 2.0’s RMD changes to the permanent rate structure under the OBBBA — has rewritten enough of the rules that planning assumptions from even three years ago may need to be revisited. The retirees who navigate this environment well are not necessarily the ones with the most money. They’re the ones who understand the specific decisions the current environment demands of them.
This piece covers the four decisions that matter most for retirement income in 2026: how to think about withdrawal rates in a world of elevated valuations and frozen rates; why the order in which you draw from different accounts is as important as the amounts; what the Roth conversion window actually is and how to use it before it closes; and how to construct a genuine income floor that doesn’t depend entirely on portfolio performance. None of it is novel in theory. In practice, most people don’t do any of it until the urgency is already past.
The Withdrawal Rate Question Has a Better Answer Than 4%
The 4% rule — withdraw 4% of your portfolio in year one, adjust for inflation annually, and your money should last 30 years — has been the default reference point for retirement income planning for three decades. Morningstar’s most recent retirement income research puts the current safe starting rate at 3.9% for a 30-year horizon with a 30–50% equity weighting. That’s close to 4%, but the methodology note matters: the higher the equity weighting above 50%, the more sequence-of-returns risk you’re carrying, and the lower the starting rate that Morningstar’s modeling supports. A retiree who enters 2026 with 80% in equities after the bull run of the past three years is carrying more sequence risk than the 4% rule was calibrated for.
Sequence-of-returns risk is the most underappreciated mechanic in retirement finance. The basic problem: a 30% market decline in year two of retirement does permanent damage that a 30% decline in year twenty does not, because in year two you’re selling shares at depressed prices to fund living expenses and permanently reducing the base from which your remaining portfolio compounds. Schwab’s research illustrates this cleanly — two investors with identical $1 million starting portfolios and identical long-run returns end up in radically different financial positions depending on whether the drawdowns come early or late. The investor who hits the bad years first runs out of money far sooner, regardless of what follows.
The practical response is not to abandon equities or shrink the withdrawal rate to 3%. Morningstar found that flexible withdrawal strategies — adjusting spending downward after poor performance years rather than maintaining a fixed inflation-adjusted draw — can push the sustainable starting rate as high as 5.7% when combined with delayed Social Security. The key word is flexible. A retirement income plan that has no mechanism to throttle spending when the portfolio takes an early hit is structurally fragile, regardless of the starting withdrawal rate. Building in an explicit guardrails framework — rules for when to reduce withdrawals and when to restore them — transforms the 4% rule from a passive assumption into an active tool.
The Account Withdrawal Sequence: Where the Real Tax Leverage Lives
Most retirees think about how much to withdraw from their portfolio. Fewer think carefully about which account to draw from first — and that sequencing decision is where a meaningful amount of lifetime tax savings either gets captured or left on the table. The conventional wisdom is to draw down taxable accounts first, then tax-deferred accounts, and save Roth accounts for last. That order generally holds, but 2026’s specific tax and RMD landscape creates enough nuance that the default often isn’t optimal.
The RMD rules under SECURE 2.0 require withdrawals from traditional IRAs and most employer-sponsored plans starting at age 73 for those born between 1951 and 1959. Roth 401(k)s are now treated the same as Roth IRAs — no lifetime RMDs for the original owner, a significant planning win. But here’s the problem that catches retirees off guard: a large traditional IRA balance that has been growing tax-deferred for 30 or 40 years can generate RMDs large enough to push Social Security income into the 85%-taxable threshold, bump Medicare premiums through IRMAA surcharges, and move the retiree into a higher bracket than they were in during their working years. The tax bill at 75 can be larger than the tax bill at 55 if the conversion work wasn’t done in the interim years.
The withdrawal sequencing decision also intersects with the timing of Social Security. Up to 85% of Social Security benefits become taxable once provisional income — adjusted gross income plus tax-exempt interest plus 50% of Social Security — exceeds $44,000 for married filers or $34,000 for single filers. For retirees who delay Social Security to 70 for the larger benefit (generally the right call for the higher earner in a married couple, with break-even around age 78 to 82 on a gross basis), the years between retirement and age 70 represent a window of potentially low taxable income that the withdrawal sequence should be designed to exploit.
The Roth Conversion Window: Use It or Lose It
The gap between the year you retire and the year your RMDs begin is the most tax-efficient window most people will see for the rest of their lives. Earned income is gone. Social Security may not have started yet. RMDs haven’t kicked in. That combination drops many retirees into the 22% or even 12% bracket temporarily — lower than they paid during their working years and almost certainly lower than they’ll pay once RMDs force large distributions from a still-growing traditional IRA balance. Converting traditional IRA assets to Roth during this window means paying tax at today’s lower rate instead of tomorrow’s higher one.
The OBBBA permanently extended the TCJA tax brackets, which eliminates the planning uncertainty that hung over Roth conversions for years. For married couples filing jointly in 2026, the 22% bracket runs to $201,050, and the 24% bracket extends to $383,900. For a retired couple with modest Social Security income and no wages, there may be $100,000 or more of conversion space available at 22% before hitting the next bracket — room that would be worth using aggressively if the alternative is paying 24% or 32% on forced RMD distributions from a larger balance a decade from now. The math of converting at 22% now versus withdrawing at 32% later is not subtle.
The IRMAA trap deserves explicit attention in conversion planning. Medicare Part B and Part D premiums are determined by income two years prior, meaning a large Roth conversion in 2026 affects 2028 Medicare costs. The 2026 IRMAA income thresholds start at $212,000 for married filers — for many retirees, aggressive conversions can be sized to stay meaningfully below that threshold while still making substantial progress on reducing the future traditional IRA balance. For retirees with very large pre-tax balances, the goal isn’t necessarily to eliminate the RMD entirely — it’s to reduce it to a size that doesn’t create a tax cascade across Social Security, Medicare, and bracket exposure simultaneously.
Building an Income Floor That Doesn't Depend on the Market
The structural vulnerability of a pure portfolio-withdrawal retirement income strategy is that every dollar of spending depends on markets cooperating. For retirees with flexible spending, adequate savings, and a long time horizon, that dependency is manageable — sequence risk is real but survivable with the right guardrails. For retirees who are less flexible, closer to the bone on savings, or simply more vulnerable to the psychological stress of watching a bear market coincide with monthly withdrawals, building a guaranteed income floor changes the risk profile of the entire plan.
Social Security delayed to 70 is the best annuity available to most Americans — inflation-adjusted, backed by the federal government, and paying roughly 24–32% more than full retirement age benefit. For a married couple where the higher earner has a $3,000 FRA benefit, the difference between claiming at 62 and waiting to 70 is approximately $1,800 per month in benefit, indexed for inflation for the remainder of both spouses’ lives. No commercial annuity offers that combination at reasonable cost. The portfolio’s job, during the gap between retirement and Social Security onset, is simply to bridge the income need — a role that even a conservative allocation of high-quality bonds and short-duration munis can handle without forcing equity sales.
For retirees who want additional guaranteed income beyond Social Security, the current interest rate environment has improved the economics of certain income annuities in ways that weren’t true during the zero-rate era. Single Premium Immediate Annuities and Deferred Income Annuities are priced from prevailing interest rates, and higher rates translate directly into higher payout factors. An annuity purchased locks today in today’s rate environment for the duration of the contract — which, for a retiree converting a portion of a traditional IRA into guaranteed lifetime income, is also a pre-tax Roth conversion of sorts, replacing a taxable RMD with a stream of taxable annuity income spread over many years rather than forced into a few concentrated distribution years.
The Plan That Works Regardless of What the Market Does
Pull all four pieces together, and the retirement income framework for 2026 has a clear internal logic. Start with a flexible withdrawal strategy that builds in explicit rules for throttling spending when portfolio performance disappoints — the guardrails approach, not a fixed inflation-adjusted draw. Sequence account withdrawals to exploit the tax window between retirement and RMD onset, running Roth conversions in years when the bracket math is most favorable. Delay Social Security to 70 for the higher earner in most married households, and use the portfolio bridge years strategically. And build a guaranteed income floor sufficient to cover non-discretionary expenses, so the equity portfolio is funding lifestyle rather than survival.
None of this requires perfect market timing. None of it requires an unusual level of financial sophistication. What it requires is a plan that addresses the mechanics of decumulation specifically — not an accumulation plan with the savings direction reversed. The retirees who will struggle in 2026 and beyond are not primarily the ones who saved too little, though that matters. They’re the ones who hit their number, stopped planning, and assumed the rest would take care of itself. It rarely does.
