The hard part of retirement isn't growing your savings. It's turning them into a paycheck that never stops.

Income flooring: cover the essentials, then invest freely

The most useful thing guaranteed income does is not the income itself. It is what covering your fixed expenses permits the rest of your portfolio to do.

Start by separating two kinds of spending

Retirement spending divides fairly cleanly into expenses that must be paid and expenses that can flex.

Essential covers housing, property taxes, utilities, food, insurance premiums, healthcare, and transportation. These arrive whether or not the market cooperated last quarter.

Discretionary covers travel, dining, gifts to family, hobbies, a second home. These can be deferred in a difficult year without any real damage.

The distinction matters because the two categories deserve fundamentally different funding. Variable resources should fund flexible expenses. Fixed obligations should be funded by something that does not vary.

Building the floor

The construction is straightforward and worth doing on paper before doing it with money.

1. Total your essential annual expenses. Actual figures, not estimates — most people are surprised in one direction or the other.
2. Subtract guaranteed income you already have. Social Security first, then any pension. For many households this covers a meaningful share of the floor already.
3. Measure the gap. Whatever remains is the number the strategy is actually about.
4. Fill it with contractual income sized to the gap — not to the whole portfolio, and not to a round number that sounds impressive.

Note that step four often requires far less capital than people assume, because steps two and three have already done most of the work. Social Security is, for most retirees, the largest guaranteed income asset they will ever own.

The effect that matters is second-order

Here is where the strategy earns its keep, and it has little to do with the income itself.

Sequence-of-returns risk exists because retirees are forced to sell assets to live, and market declines do not consult anyone’s schedule. Selling into weakness converts a temporary paper loss into a permanent one, because those shares are gone and cannot participate in the recovery.

Once essential spending is covered by income that arrives regardless of market conditions, that forced-seller dynamic disappears for the portion it covers. The remaining portfolio is no longer survival money. It is genuinely long-horizon money, and it can be invested accordingly — and left alone through a bad year, because nothing compels you to touch it.

Retirees without a floor frequently end up in the opposite position: holding more cash and bonds than their time horizon warrants, because they need the whole portfolio to feel safe. The floor is what lets the rest of it stop pretending to be safe and go do its job.

The behavioral half

A consistent finding in retirement research is that retirees with guaranteed income spend more comfortably, and report higher satisfaction, than retirees holding equivalent assets without a guaranteed floor.

The reason is not complicated. Drawing down a balance feels like depletion, and it triggers a reasonable instinct to spend less. Receiving a payment feels like income, and income feels spendable.

A great many people spend their retirement underspending — declining trips, postponing the kitchen, saying no to things they could easily afford — out of a fear the numbers never justified. A floor does not make anyone wealthier. It makes the money they already have easier to actually use.

What to watch

• Inflation. A level payment buys measurably less after twenty years. Options include increasing payout designs, laddering start dates so income steps up over time, or leaving the discretionary portfolio to handle rising costs.
• Don’t oversize the floor. Covering essentials is the objective. Covering everything commits capital that would serve you better staying flexible.
• Diversify the promise. If the floor is large, spreading it across more than one issuing carrier spreads the credit exposure behind it.
• Coordinate with Social Security. Delaying to 70 increases a benefit that is already inflation-adjusted and federally backed. Frequently the most efficient way to raise the floor is to enlarge the piece you already own.

Sequence-of-returns risk, explained without the math

Two retirees can earn the identical average return over thirty years and end up in completely different places. The difference is the order the returns arrived in.

While you are saving, the order of your returns barely matters. A bad year early and a bad year late produce the same ending balance, because nothing is leaving the account. The arithmetic is commutative.

Once you begin withdrawing, that stops being true, and the change is dramatic.

Why withdrawals change everything

Imagine two retirees. Both start with the same balance, both withdraw the same amount each year, and both earn the identical average annual return across thirty years. The only difference is that one experiences their worst years at the beginning of retirement and their best years at the end. For the other, the order is reversed.

The retiree who hits bad years first can run out of money. The one who hits them last may finish with more than they started with. Same average. Same withdrawals. Entirely different outcomes.

The reason is that a withdrawal during a decline forces you to sell more shares to raise the same dollars. Those shares are permanently gone. When the recovery arrives — and historically it does — there is less left to participate in it. A temporary paper loss has been converted into a permanent one.

The years that matter most

Research on this risk consistently points to the same window: roughly the five years before and the five years after the date you stop working. Losses in that decade do disproportionate damage, because the balance is at its largest and the withdrawal schedule is just beginning.

This is uncomfortable, because it is also the decade in which you have the least ability to respond. You cannot work another twenty years to make it up.

What actually reduces it

• Guaranteed income covering essential expenses. If the mandatory bills are paid by something that arrives regardless of market conditions, you are not a forced seller. The portfolio can stay invested and recover.
• A cash or short-bond reserve. Holding one to three years of spending in stable assets lets you draw from those during a decline rather than from equities. Effective, though it requires discipline to refill.
• Flexible spending. Reducing discretionary withdrawals in a bad year genuinely helps. It also asks retirees to cut back at precisely the moment they feel least secure, which is why plans that depend on it often fail in practice.
• Delaying Social Security. A larger inflation-adjusted federal benefit is the cheapest guaranteed income most people can buy, and it reduces how much the portfolio has to produce.

The point most people miss

Sequence risk is not an argument for holding more cash. Holding more cash lowers your expected return every year in exchange for protection you may need in only a few of them.

It is an argument for segmenting: funding the non-negotiable spending with something contractual, and letting the rest of the portfolio hold its long-term allocation without being raided during downturns. That is the entire logic behind income flooring.

The Social Security delay bridge

Delaying to 70 buys the best guaranteed income available to a retiree. The obstacle is funding the gap years — which is a solvable problem.

Social Security benefits increase by roughly eight percent for each year you delay claiming past full retirement age, up to age 70. That increase is permanent, adjusted for inflation annually, and backed by the federal government.

It is difficult to overstate how good that is. There is no commercially available product that offers an inflation-adjusted, federally guaranteed lifetime income increase on those terms. For most retirees, the largest guaranteed income asset they will ever own is the one they already have, and enlarging it is often the single most efficient planning move available.

So why doesn't everyone delay?

Because you have to eat between 65 and 70. Most people claim early not because they analyzed it but because they needed income and Social Security was sitting there.

That is a funding problem, not a math problem, and funding problems have solutions.

Bridging the gap

A period-certain annuity pays a defined amount for a defined number of years and then stops. Purchased with a portion of retirement assets, it can fund exactly the years you are delaying — income arriving monthly through the bridge period, ending as the larger Social Security benefit begins.

The effect is a swap: a fixed number of years of contractual income now, in exchange for a permanently larger, inflation-adjusted, government-backed benefit for the rest of two lifetimes.

Where it does not work

• Health. Delaying is a bet on longevity. Serious health issues can reverse the conclusion entirely.
• Insufficient bridge assets. If funding five years of spending would leave the portfolio too thin, the strategy is asking too much.
• Tax interactions. Drawing more heavily from qualified accounts during the bridge raises taxable income in those years. Sometimes that is a bonus — it doubles as a drawdown strategy that reduces future RMDs. Sometimes it pushes across an IRMAA threshold. It needs modeling rather than assuming.

What to work out before deciding

1. Your actual benefit at 62, at full retirement age, and at 70 — from your Social Security statement, not from memory.
2. What five years of essential spending costs, net of any other income.
3. Which assets would fund the bridge, and what that does to your tax picture in those years.
4. For couples: run the higher earner and lower earner decisions separately. They are genuinely different questions.

Annuity laddering: three ladders, three different problems

Laddering is usually described as one technique. It is really three, and they solve completely different things.

Committing an entire allocation to a single contract, on a single day, with a single insurance company, concentrates three separate risks that do not need to be concentrated. Each has its own ladder.

Ladder one: staggered start dates

Rather than turning on all your income at once, you arrange for it to begin in stages — some now, some in five years, some in ten.

The problem it solves is inflation. A level payment buys measurably less after twenty years. Income that steps up as later contracts activate rises against your costs instead of falling behind them. It also means later contracts spend more time deferring, which generally produces larger payments when they do begin.

Ladder two: staggered purchase dates

Annuity payouts and crediting rates depend heavily on interest rates on the day you buy. Purchasing everything in one transaction means one day’s rate environment determines your outcome permanently.

The problem it solves is timing. Spreading purchases across quarters or years averages your exposure to the rate environment, in the same way periodic investing averages exposure to market prices. If rates rise after your first purchase, later ones capture the improvement.

Ladder three: multiple carriers

Every guarantee in an annuity contract rests on the claims-paying ability of the issuing insurance company. One contract means one balance sheet standing behind all of your guaranteed income.

The problem it solves is concentration. Splitting a large allocation across two or three well-rated carriers spreads that exposure. For a modest allocation this may not justify the added complexity. For a large one it usually does.

Combining them

The three ladders are independent, and a well-built plan often uses more than one at once — for instance, purchases spread across two years, with start dates staggered five years apart, split between two carriers. That is six decisions, not one, and each should have a reason attached to it.

How much guaranteed income is enough?

The most common mistakes are committing far too much and committing nothing at all. Here is how the number is actually derived.

There is no percentage that answers this. The right amount of guaranteed income is not a share of your portfolio — it is a function of your expenses, and it is calculated from the bottom up.

The derivation

1. Total your essential annual spending. Housing, property taxes, utilities, food, insurance, healthcare, transportation. Use twelve months of actual statements, not estimates. Most people are wrong by a meaningful margin in one direction or the other.
2. Subtract guaranteed income you already have. Social Security, any pension, any existing contractual income.
3. The remainder is the gap. That is the only number the strategy is about.
4. Fund the gap — not the portfolio, not a round figure, and not what a product illustration happens to produce.

Step two frequently does most of the work. For a couple with two Social Security benefits, essential expenses may already be largely covered, and the gap turns out to be far smaller than the conversation started out assuming.

Two ways to get it wrong

Committing too much. Capital committed to guaranteed income is capital that is no longer flexible, no longer available for a large unexpected expense, and no longer positioned for growth. Covering discretionary spending with contractual income buys certainty for expenses that did not need it.

Committing nothing. The alternative is that every dollar of retirement spending depends on markets and on withdrawal-rate assumptions. That works until it does not, and the failure mode arrives at the age when you have the least capacity to respond.

Adjustments worth considering

• Inflation. Consider sizing the floor slightly above today’s essentials, or laddering start dates so income rises over time.
• The survivor. Model household expenses after one spouse dies. Some costs fall; many do not. Income frequently falls further, because one Social Security benefit ends.
• Healthcare and long-term care. These are the largest genuine unknowns in retirement spending, and they argue for keeping liquid reserves rather than for a larger floor.
• Legacy intent. If leaving assets matters, that changes which payout options make sense — not necessarily the size of the floor.

Joint and survivor decisions for married couples

Household income almost always falls at the first death. Household expenses rarely fall as far. Planning for that gap is a decision made years in advance.

Retirement planning for couples tends to model one household with two people in it. The arithmetic changes at the first death, and it changes in a direction most plans never examine.

What actually happens

Social Security falls. The household keeps the larger of the two benefits and loses the smaller. For a couple with similar earnings records, that can approach a one-third reduction in guaranteed income.

Any pension may fall or stop, depending on the survivor election made at retirement — a decision usually irrevocable.

Filing status changes. The survivor generally files as a single taxpayer beginning the year after the death, with roughly half the bracket width, on a portfolio and required distributions that have not shrunk. The same income produces a higher tax bill.

Medicare thresholds tighten. IRMAA brackets for a single filer are lower than for a couple, so the same distribution can now trigger a surcharge that it did not before.

Expenses do not halve. Housing, property taxes, insurance, and maintenance are largely unchanged. Some households face higher costs, because work a spouse handled is now purchased.

The decisions that address it

Social Security claiming

The higher earner’s benefit is what the survivor keeps. Delaying that benefit toward 70 raises income for as long as either spouse lives, not just the higher earner’s own lifetime. It is effectively survivor protection purchased with patience, and it is frequently the most valuable single decision available to a couple.

Correspondingly, claiming the lower earner’s benefit earlier is often perfectly sensible, since that benefit ends at the first death regardless.

The joint and survivor election

Annuities and pensions with lifetime income offer a choice about what continues to the survivor — commonly 100%, 75%, two-thirds, or 50% of the original payment. A higher continuation percentage means a lower payment while both are alive.

The instinct is to take the larger current payment. Weigh it against the survivor’s complete picture: what Social Security they will keep, what other income exists, and how many years they may live alone. Women statistically outlive men, and the survivor period is frequently longer than couples anticipate.

Life insurance as a coordinating tool

Where a pension election maximized current income at the cost of survivor continuation, life insurance on the pensioner can replace what the survivor loses. This requires insurability and a durable premium plan, and the comparison should be run honestly against simply electing a survivor option.

The planning conversation nobody wants to have

Model the survivor’s situation explicitly. Not as an afterthought — as its own scenario, with its own income statement, its own tax calculation, and its own answer to whether it works.

It is an uncomfortable exercise. It is also the difference between a survivor discovering a problem in a year when they have the least capacity to solve it, and a couple addressing it a decade earlier when every option is still open.

Flattening the RMD curve

Required distributions don’t just continue in retirement. They grow — as a percentage — every single year. Here is why that happens, what it costs, and the four levers that work best when they’re pulled early.

Why the number keeps climbing

An RMD is calculated by dividing your prior year-end balance by a life expectancy divisor published by the IRS. The balance changes with markets. The divisor does something more predictable: it shrinks every year, without exception.

Read that column on the right again. Even if the account balance never grows by a dollar, the mandatory withdrawal more than doubles as a percentage between 73 and 90. And of course most accounts do grow, which means the dollar figure can rise considerably faster than the percentage alone suggests.

This is the mechanic behind what advisors call RMD creep: a tax bill that increases on a schedule, whether or not the retiree needs or wants the money.

What RMD creep actually costs

The obvious cost is income tax on money you were not planning to spend. The less obvious costs are often larger, because required distributions raise your income, and a surprising number of things in the tax code are keyed to income rather than to wealth.

• Bracket compression. Distributions stack on top of Social Security, pensions, and investment income. A retiree comfortably in one bracket at 73 can find themselves two brackets higher at 85 without a single change in lifestyle.
• Medicare surcharges. IRMAA surcharges on Part B and Part D premiums are triggered by income thresholds, and they are assessed on a two-year lookback. A large distribution today raises your premiums two years from now.
• Taxation of Social Security. The share of your Social Security benefit that becomes taxable rises with income. RMDs are one of the most common reasons a retiree crosses those thresholds.
• The survivor’s problem. When one spouse dies, the survivor generally files as a single taxpayer, with roughly half the bracket width, on an account that has not shrunk. This is one of the most predictable and least anticipated tax events in retirement.

The window most people leave empty

Consider a typical sequence. Someone retires at 65. Social Security is delayed to 70. Required distributions begin at 73. That leaves a five-to-eight-year stretch with unusually low taxable income — often the lowest since early in their career.

This is the cheapest tax real estate a retiree will ever own. Most of it goes unoccupied, for a reason that is entirely understandable: taking money out of a retirement account when you do not need it feels like the opposite of prudence.

But the money is going to come out. The only open question is when, at what rate, and stacked on top of what other income. Deferral is not avoidance — it is a decision to pay later, at a rate nobody can promise you.

Lever one: deliberate bracket filling

The simplest approach is to take voluntary distributions before they are required, sized to fill the remaining room in your current bracket without crossing into the next one.

Each dollar taken now is a dollar not in the balance that drives future RMDs, and it comes out at a rate you selected rather than one the calendar selected for you. The arithmetic is straightforward. The discipline is not, which is why it helps to run the projection and see the two paths side by side.

Lever two: Roth conversions

A Roth conversion uses the same mechanic with a substantially better destination. Rather than moving money to a taxable account, you move it to an account where future growth and future distributions are generally tax-free, and where no required minimum distributions apply during your lifetime.

That last point does the real work. Every dollar converted is a dollar permanently removed from the RMD calculation, forever, along with all of its future growth.

Lever three: the QLAC carve-out

A Qualified Longevity Annuity Contract is a fixed deferred income annuity purchased inside an IRA or 401(k). Its distinguishing feature is regulatory: the premium is excluded from the account balance used to calculate required minimum distributions.

Under SECURE 2.0 the old 25%-of-balance restriction was eliminated and replaced with a flat, inflation-indexed dollar limit — $210,000 per person for 2026, per IRS Notice 2025-67. It is a lifetime, per-person cap that applies across every account you own, so it cannot be repeated account by account.

Two things happen at once. Taxable distributions fall during the deferral period, and the deferred contract turns into guaranteed lifetime income beginning on a date you choose, as late as age 85 — which is to say, covering precisely the years most likely to exhaust a portfolio.

Two constraints worth knowing up front. It must be a fixed contract; indexed and variable products do not qualify. And the sequencing matters more than people expect: the premium has to leave the account balance before a year’s RMD is calculated in order to reduce it, so a QLAC funded at 75 has already missed two years of benefit it could have captured at 72.

Lever four: qualified charitable distributions

For anyone who is charitably inclined and at least 70½, the QCD is the most efficient charitable tool in the code. Funds transfer directly from an IRA to an eligible public charity, and the amount is excluded from gross income entirely — it never appears on the return at all.

The 2026 limit is $111,000 per person, also under IRS Notice 2025-67, with a separate one-time $55,000 allowance for funding a charitable remainder trust or charitable gift annuity.

The reason this beats writing a check is structural. A deduction requires itemizing and reduces taxable income; an exclusion never enters income in the first place, so it also lowers the figure that drives IRMAA surcharges and Social Security taxation. It works for non-itemizers, and it counts toward satisfying the RMD.

Two mechanical traps: the transfer must go directly from custodian to charity — withdrawing and then writing a personal check disqualifies it — and donor-advised funds and private foundations are not eligible recipients.

Where annuities actually fit — and where they don’t

This deserves a clear statement, because it is frequently misrepresented in the marketplace.

The real contribution an annuity makes to this strategy is different, and in practice more valuable. Accelerating distributions out of a qualified account only helps if the money lands somewhere useful. Converting a portion of it into guaranteed lifetime income means you are not simply moving a balance from one account to another and hoping — you are exchanging a future stream of forced, unpredictable taxable income for a contractual stream whose timing and amount you chose deliberately.

It also makes the rest of the strategy easier to actually execute. Roth conversions are far less daunting when essential spending is already covered by income that arrives regardless of what the conversion did to this year’s balance.

Sequencing: the part that gets missed

These four levers are not alternatives. They work together, and the order matters more than the selection.

1. Project first. Model RMDs to age 90 under current assumptions before deciding anything. Most people have never seen this number, and it is usually the moment the strategy becomes obvious.
2. Use the low-income window aggressively. Between retirement and 73, fill brackets and convert. This window does not reopen.
3. Fund a QLAC before RMDs begin, ideally in the late sixties or at 72, so the carve-out applies from the first calculation forward.
4. Layer QCDs from 70½ if charitable giving is already part of the picture. There is no reason to give from a checking account while an IRA is sitting there.
5. Revisit annually. Brackets, thresholds, and limits are indexed and change every year. So does the balance.

The QLAC carve-out, in detail

A narrow provision that does one thing well: it removes a defined amount from the balance the IRS uses to calculate your required distributions.

A Qualified Longevity Annuity Contract is a fixed deferred income annuity purchased inside an IRA or employer plan. What distinguishes it from any other annuity held in a qualified account is regulatory: the premium is excluded from the account balance used to compute required minimum distributions.

The 2026 numbers

SECURE 2.0 eliminated the old restriction limiting QLAC premiums to 25% of the account balance, and replaced it with a flat, inflation-indexed dollar cap. For 2026 that figure is $210,000 per person, under IRS Notice 2025-67.

Two things about that limit catch people out. It is a lifetime cap, not annual. And it is per person across all accounts — funding $130,000 from an IRA and $80,000 from a 401(k) exhausts it. A married couple can each use their own limit from their own accounts.

What qualifies

• The contract must be a fixed annuity. Indexed and variable products do not qualify.
• Income must begin no later than the first day of the month following the insured’s 85th birthday.
• The contract must state that it is intended to be a QLAC.
• Return-of-premium death benefits are permitted, which addresses the most common objection — though electing one reduces the income the contract produces.

What it accomplishes

Two effects run at the same time. Taxable required distributions fall for every year the contract is in deferral, because the premium is not in the calculation base. And the contract becomes guaranteed lifetime income beginning on a date you choose — covering precisely the later years most likely to exhaust a portfolio.

Honest limits

It is a bounded carve-out, not a solution to the RMD problem. Against a $2 million IRA, $210,000 shelters about ten percent of the balance. The dollar reduction in annual distributions is real and can be meaningful, but a large-balance retiree should not expect it to do the whole job. It works best layered with Roth conversions and, where applicable, qualified charitable distributions.

It is also illiquid by design. QLAC premiums are generally not accessible during deferral. This is money you are deliberately setting aside for age 85, and it should be sized on that basis.

Roth conversions in the low-income window

The years between retirement and 73 are the cheapest tax real estate most retirees will ever own. Most of it goes unused.

Consider the common sequence. Someone retires at 65. Social Security is delayed to 70. Required distributions begin at 73. That leaves a stretch of five to eight years with unusually low taxable income — often the lowest since early in their career.

Then it closes, permanently, and income steps up on a schedule that keeps rising.

What a conversion actually does

You move money from a traditional IRA to a Roth IRA and pay ordinary income tax on the amount converted in the year you convert. In exchange, that money and all of its future growth generally come out tax-free, and — the part that matters most here — it is permanently removed from the balance that drives required minimum distributions.

Roth IRAs have no lifetime RMDs for the original owner. Every dollar converted is a dollar the IRS will never force out of your account.

Why the window is the whole point

Converting is only attractive if the rate you pay now is lower than the rate you would pay later. In the low-income window you frequently control your bracket almost entirely, because you decide how much to convert.

Once Social Security and RMDs are both running, that control largely disappears. The income arrives whether you want it or not, and conversions on top of it are taxed at the margin.

The three things that go wrong

Paying the tax from the converted money. This substantially weakens the outcome, since less lands in the Roth and the amount used for tax never compounds. Before 59½ it can also trigger the additional 10% tax. Fund the tax from outside dollars.

Crossing an IRMAA threshold. Medicare surcharges are keyed to income with a two-year lookback, and they operate as cliffs rather than gradual phase-ins. One dollar over a threshold raises premiums for a full year. Converting slightly less across more years frequently costs less overall.

Ignoring the five-year rules. Each conversion carries its own five-year clock for penalty-free access to the converted amount, and there is a separate five-year rule for earnings. This rarely constrains someone converting in their sixties with no near-term need for the money — but it should be checked rather than assumed.

Situations that make conversions more attractive

• Heirs in higher brackets than you. Under the ten-year rule, adult children usually empty an inherited IRA during their peak earning years. Converting at your rate instead of theirs can be a large transfer of value.
• A surviving spouse ahead. The survivor generally files single, with roughly half the bracket width, on an account that has not shrunk.
• A year with unusually low income — a gap year, a business loss, a large deductible expense.
• Substantial time before RMDs. The longer the converted money compounds tax-free, the better the arithmetic.

Where guaranteed income helps

Conversions ask you to voluntarily pay tax and reduce your balance during years when nothing is forcing you to. That is uncomfortable, and discomfort is why many people who should convert do not.

When essential expenses are already covered by contractual income, the conversion stops being a question about whether you can afford to part with the money and becomes purely a question of tax arbitrage — which is what it always should have been.

Qualified charitable distributions: why an exclusion beats a deduction

For anyone over 70½ who gives to charity, this is the most efficient tool in the code — and it is still badly underused.

A qualified charitable distribution transfers funds directly from an IRA to an eligible public charity. The amount is excluded from gross income entirely. It never appears on your return as income at all.

For 2026 the limit is $111,000 per person, under IRS Notice 2025-67, with a separate one-time $55,000 allowance for funding a charitable remainder trust or charitable gift annuity. Each spouse has their own limit from their own IRA.

Why exclusion is better than deduction

Writing a check to charity produces a deduction, which requires itemizing and reduces taxable income. A QCD produces an exclusion, which means the money never enters income in the first place.

That difference matters more than it sounds, because a great many things in the tax code are keyed to income rather than to taxable income:

• IRMAA. Medicare Part B and Part D surcharges are triggered by income thresholds. A QCD keeps the amount out of that calculation; a deduction does not.
• Social Security taxation. The share of your benefit that becomes taxable rises with income. Same logic.
• Itemizing. Most retirees now take the standard deduction, which means a charitable deduction produces no benefit at all. A QCD works regardless.
• It counts toward your RMD while never being taxed — satisfying the requirement without the income.

The timing detail worth knowing

The first dollars distributed from an IRA in a year count toward that year’s RMD. If you want a QCD to satisfy the required distribution, it has to happen before you take any other withdrawals — not afterward as a correction.

How non-qualified annuities are taxed

After-tax money in an annuity follows rules that differ sharply depending on whether you withdraw or annuitize — and the difference can be substantial.

A non-qualified annuity is funded with money that has already been taxed. That creates a contract with two components: your original principal, which has been taxed, and accumulated earnings, which have not.

How those components come out to you depends entirely on which method you use to access the money.

While it grows

Nothing is taxed. There is no annual 1099 for interest, dividends, or gains inside the contract, and reallocating among the options inside it does not create a taxable event.

Compare that to a taxable brokerage account, which generates a tax bill every year whether or not you spent a dollar. That annual leak compounds against you over decades, and its absence is the core structural advantage here. There is also no contribution limit and no income phase-out, which makes non-qualified annuities one of the few deferral options remaining for someone who has already maxed every qualified plan available.

Method one: withdrawals

Withdrawals follow a last-in, first-out rule. Earnings come out first and are taxed as ordinary income — not at capital gains rates. Only after all earnings have been withdrawn does your original after-tax principal come back to you tax-free.

The practical consequence is that early withdrawals from a contract with substantial gains are fully taxable, which surprises people who assume some portion is a tax-free return of their own money.

Method two: annuitization

If you convert the contract into a payment stream instead, treatment improves considerably. An exclusion ratio applies: each payment is divided into a tax-free return of principal and a taxable earnings portion, based on the relationship between what you paid and what you are expected to receive.

Rather than front-loading all the tax, it spreads evenly across the payment period. For many retirees this produces meaningfully lower taxable income in the early years than an equivalent withdrawal strategy would.

Other rules that apply

• Before 59½. The taxable portion may face an additional 10% federal tax, with limited exceptions.
• No step-up at death. Unlike appreciated stock or real estate, an annuity does not receive a step-up in basis. A beneficiary owes ordinary income tax on the gain as it is distributed — though many can spread distributions over time rather than taking a lump sum.
• No RMDs. Non-qualified annuities are not subject to required minimum distributions during the owner’s lifetime. The money can stay deferred indefinitely.
• 1035 exchanges. One annuity may generally be exchanged for another without current tax, carrying its cost basis forward.

IRMAA: the surcharge nobody budgets for

Medicare premiums are keyed to your income from two years ago, and they move in cliffs rather than slopes. That combination catches a lot of retirees.

IRMAA stands for the income-related monthly adjustment amount. It is a surcharge added to Medicare Part B and Part D premiums for beneficiaries whose income exceeds published thresholds.

Two features make it a planning problem rather than merely a cost.

Feature one: the two-year lookback

Your premium this year is determined by the tax return you filed two years ago. A large distribution, a Roth conversion, or a property sale today shows up as higher Medicare premiums two years from now — frequently long after the reason for it has been forgotten.

Feature two: it is a cliff, not a ramp

Most of the tax code phases in gradually. IRMAA does not. Cross a threshold by a single dollar and you pay the full surcharge for that entire tier, for twelve months, on both Part B and Part D, for each spouse enrolled.

This is why a conversion or distribution plan that ignores IRMAA can be genuinely expensive at the margin. The last few thousand dollars of income in a year can cost several times what the marginal tax rate alone would suggest.

What raises the figure IRMAA uses

• Required minimum distributions and other IRA withdrawals
• Roth conversion amounts
• Capital gains, including from a home sale above the exclusion
• Taxable interest and dividends, plus tax-exempt interest, which is added back
• The taxable portion of Social Security benefits

What helps

• Qualified charitable distributions — excluded from income entirely, so they never enter the calculation.
• Sizing conversions to stay under a threshold rather than converting a round number. Converting slightly less across more years frequently costs less in total.
• Roth withdrawals, which do not count as income.
• Appealing a life-changing event. If income dropped because of retirement, the death of a spouse, divorce, or loss of a pension, Form SSA-44 lets you request that Social Security use current income instead of the two-year-old return. This is regularly overlooked in the first years of retirement, when income typically falls sharply.

Thresholds and surcharge amounts are adjusted annually, so any planning near a line should use the current year’s published figures rather than last year’s.

The inherited IRA problem, and the policy built to solve it

The SECURE Act compressed the inherited IRA from a lifetime to ten years, and it lands on your children during their highest-earning decade. Here is what that costs, and how a survivorship policy can be built to absorb it.

What changed, and why it matters more than people realize

Before 2020, a non-spouse beneficiary who inherited an IRA could generally stretch distributions across their own life expectancy. A 55-year-old inheriting from a parent could take small annual withdrawals for thirty years, most of them absorbed in lower brackets, with the balance compounding tax-deferred the entire time.

The SECURE Act eliminated that for most non-spouse beneficiaries. The rule now is ten years: the account must be fully distributed by the end of the tenth year following the year of death.

Under the final regulations issued in 2024, there is an additional wrinkle. If the original owner died on or after their required beginning date, the beneficiary must also take annual distributions in years one through nine, with whatever remains cleared out in year ten. The account cannot simply be left alone and emptied at the end.

Why ten years is so much more expensive than thirty

The compression is the problem, and it is worth being concrete about who absorbs it.

A couple who die in their eighties typically leave the account to children in their fifties or early sixties. That is the highest-earning decade of most careers. The inherited distributions do not arrive in a vacuum — they stack directly on top of a salary that is already near its lifetime peak.

Money that under the old rules might have come out gradually at modest rates now comes out in ten concentrated slices at the top of someone’s bracket. And that is before state income tax, which is not trivial in New York, New Jersey, California, or Minnesota.

There is a second effect that often surprises families. The inherited distributions raise the beneficiary’s income for other purposes too: phase-outs, surtaxes, and in some cases their own Medicare premiums if they are already enrolled. The headline rate understates the real cost.

The two-part response

Once the problem is stated properly, the response splits cleanly into two halves that address different portions of the same account.

Part one is to shrink the taxable asset while you are alive — bracket filling, Roth conversions, the QLAC carve-out, and qualified charitable distributions. That work is covered in detail in Flattening the RMD Curve, and it should come first, because every dollar handled there is a dollar that never becomes an inherited-IRA problem at all.

Part two is to fund the tax on whatever remains. Almost no one converts an entire IRA; the rates on the last conversion dollars stop making sense long before the account is empty. Something will be left, and it will be taxable to your heirs.

Why survivorship life insurance fits this particular problem

Second-to-die insurance — also called survivorship life — insures two people under one policy and pays a death benefit when the second of them dies. For most other purposes that is a strange design. For this one it is almost exactly right, for three reasons.

The timing matches the liability

When the first spouse dies, the survivor can generally roll the IRA into their own name. No ten-year clock, no acceleration, no tax event. The problem simply does not exist yet.

It exists at the second death, when the account passes to the children and the ten-year rule engages. A survivorship policy pays at precisely that moment. Insurance that paid at the first death would arrive years early, for a bill that had not yet come due.

The cost structure is favorable

Because the insurer is not obligated until both insureds have died, a survivorship policy generally provides more death benefit per premium dollar than an individual policy on either spouse. Underwriting also tends to be more accommodating — couples where one spouse has health issues that would make individual coverage expensive or unavailable can often still obtain survivorship coverage on reasonable terms.

The proceeds arrive income-tax-free

Life insurance death benefits are generally received by beneficiaries free of federal income tax under Internal Revenue Code Section 101(a). That is what makes the offset work: the asset arriving is untaxed, and the liability it is meant to cover is an income tax bill.

Funding it from money you are already forced to take

This is the part of the strategy that tends to make it click for people.

A couple with a large qualified balance is required to take distributions they frequently do not need. That money is taxed on arrival and then usually parked in a taxable account, where it generates more taxable income, gets added to the estate, and continues compounding the very problem it came from.

Redirecting some portion of that unneeded distribution into premium changes what it becomes. Money the IRS forces out of the account gets converted into an asset that arrives income-tax-free at exactly the moment the tax bill lands. The distribution was going to happen regardless. This determines where it ends up.

What this strategy does — stated precisely

It is worth being exact here, because this strategy is frequently described in a way that is not quite accurate, and the imprecise version can create expectations a policy cannot meet.

What has to be true for it to work

This strategy fits a specific situation well and other situations poorly. The honest version of the conversation covers both.

• Insurability. Both lives are underwritten. Survivorship underwriting is more forgiving than individual coverage, but it is not unlimited, and health drives the cost of the entire strategy.
• Premium durability. This is a commitment measured in decades. A policy that lapses in year eighteen may deliver nothing and can create a taxable event on the way out. The funding plan has to be sustainable through a market downturn, a long-term care event, and the death of the first spouse.
• Sizing requires an estimate. The right death benefit depends on the projected account balance at the second death and your children’s expected brackets a decade or more from now. That is a modeled figure, not a known one, and it should be revisited periodically.
• Ownership needs legal input. If the estate is large enough to face federal or state estate tax, the policy should generally be owned by an irrevocable trust so that the death benefit is not pulled into the taxable estate. That is attorney work, and it needs to be set up correctly before the policy is issued rather than fixed afterward.

When something else is the better answer

Three situations where a different approach usually wins, and a good advisor should say so.

When the money is needed. If the required distributions are funding your actual retirement, they are not available for premium. The strategy assumes surplus, and manufacturing surplus that is not there is how good planning turns into bad planning.

When there is charitable intent. A qualified charity is a tax-exempt beneficiary. Naming a charity directly as the IRA beneficiary and leaving other assets — which receive a step-up in basis — to your children is frequently more efficient than any insurance solution, and it costs nothing to implement. If charitable giving is already part of your plan, this deserves examination first.

When Roth conversions can substantially finish the job. For a moderate balance and a long runway, systematic conversion may reduce the taxable account enough that the remaining exposure does not justify a decades-long premium commitment. The two strategies are complementary, and the right mix is arithmetic rather than doctrine.

How the analysis should actually be done

1. Project the account to the second death, including required distributions along the way, to estimate what will actually be left.
2. Estimate the tax on that balance under a ten-year distribution to your beneficiaries, using a defensible assumption about their brackets and their state of residence.
3. Reduce the exposure first through conversions, QLAC positioning, and charitable planning — then measure what remains.
4. Price survivorship coverage for the residual figure, and compare total projected premiums against the tax the benefit is intended to offset.
5. Confirm ownership and beneficiary structure with your attorney before applying, particularly if state estate tax is in play.

If the numbers work, they work clearly. If they do not, that is worth knowing before a policy is issued rather than after.

The ten-year rule, beneficiary by beneficiary

Who has to empty an inherited retirement account within ten years, who does not, and what the 2024 final regulations added.

Before 2020, a non-spouse beneficiary could generally stretch distributions from an inherited IRA across their own life expectancy. A 55-year-old inheriting from a parent could take small annual withdrawals for thirty years, most absorbed in lower brackets, while the balance kept compounding.

The SECURE Act ended that for most beneficiaries. The rule now is ten years: the account must be fully distributed by December 31 of the tenth year following the year of death.

What the 2024 final regulations added

For several years there was genuine confusion about whether beneficiaries had to take anything during the ten-year period or could simply empty the account at the end. The final regulations settled it, and the answer depends on when the original owner died.

• If the owner died on or after their required beginning date, the beneficiary must take annual distributions in years one through nine, with the remaining balance cleared out by the end of year ten.
• If the owner died before their required beginning date, no annual distributions are required — the account simply has to be empty by the end of year ten.

The first case is the more common one for retirees who have already started taking RMDs, and it removes the flexibility to defer everything to the final year.

Who is exempt

The regulations define a category called eligible designated beneficiaries who are not subject to the ten-year rule:

• A surviving spouse. Generally the strongest position — a spouse can typically roll the account into their own IRA and treat it as their own, with no acceleration at all.
• A minor child of the account owner. Life-expectancy distributions apply until the child reaches majority, at which point the ten-year clock begins. Note that this covers the owner’s own children, not grandchildren.
• A disabled or chronically ill individual, as defined in the regulations.
• A beneficiary not more than ten years younger than the owner — often a sibling, or a partner of similar age.

Adult children, the most common beneficiaries of large IRAs, are generally not in this group.

Why the compression is expensive

A couple who die in their eighties typically leave the account to children in their fifties or early sixties — the highest-earning decade of most careers. The distributions stack on top of a salary already near its peak, in ten concentrated slices rather than thirty gradual ones.

State income tax compounds it. In New York, New Jersey, California, or Minnesota, the combined federal and state rate on those distributions can be substantially higher than what the original owner would ever have paid.

What can be done about it

1. Reduce the taxable balance during your lifetime through conversions, the QLAC carve-out, and charitable distributions.
2. Consider which assets go to whom. A qualified charity is a tax-exempt beneficiary; appreciated taxable assets receive a step-up in basis for your children. Splitting along those lines is often more efficient than dividing everything evenly.
3. Fund the residual tax with life insurance sized to the projected liability.
4. Review beneficiary designations and any trust language against the current rules.

Naming a charity as your IRA beneficiary

If charitable giving is already part of your plan, which asset you leave to whom may matter more than how much.

Most people divide their estate by percentage and leave the asset selection to chance. For anyone with both charitable intent and a large qualified account, that is usually the expensive way to do it.

The asymmetry

Consider two assets of equal value: $500,000 in a traditional IRA, and $500,000 in a taxable brokerage account holding long-held appreciated stock.

To your children, those are not equal at all. The IRA is fully taxable as ordinary income as it comes out, compressed into ten years during their peak earning years. The brokerage account receives a step-up in cost basis at your death — the embedded gain simply disappears for income tax purposes.

To a qualified charity, they are equal, because a charity is tax-exempt and pays nothing on either one.

The implication

If you intend to leave something to charity, funding that bequest with qualified money and leaving taxable assets to your children generally produces more total value for both — without changing the size of anyone’s share.

The charity receives the full amount, untaxed. Your children receive assets with a stepped-up basis, and the money that would have gone to income tax stays in the family.

This costs nothing to implement. It is a beneficiary designation form.

Practical points

• Name the charity directly on the IRA beneficiary form. Directing a bequest through your will and funding it from IRA money instead creates income to the estate.
• Percentages or separate accounts. A charity and individuals can be named on the same account, though splitting into separate IRAs sometimes simplifies administration and avoids complications with distribution timing.
• Confirm the legal name and tax ID of the organization. Charities merge and rename, and a designation naming an entity that no longer exists creates problems at exactly the wrong moment.
• QCDs during life do similar work. If you are over 70½ and giving annually, giving from the IRA rather than a checking account reduces the balance while producing a better tax result today.

When this is worth examining before anything else

If charitable giving is already part of your plan and you are also considering life insurance to offset the tax on an inherited IRA, run the beneficiary-designation analysis first. It may absorb a meaningful portion of the qualified balance at zero cost, leaving a smaller residual to insure — and a smaller policy is a smaller premium commitment.

Beneficiary designations override your will

The most common estate planning failure has nothing to do with the estate plan. It is a form somebody filled out in 1998 and never looked at again.

Retirement accounts, annuities, and life insurance policies pass by beneficiary designation. They do not pass under your will. If the two disagree, the designation wins.

This surprises people who have just paid an attorney to draft a careful estate plan. The plan governs the assets that flow through the estate. It does not reach a 401(k) whose beneficiary form still names a spouse from a previous marriage.

How the failures actually happen

• A divorce that was never reflected on the form. State law sometimes revokes an ex-spouse designation and sometimes does not, and federal law governing employer plans can preempt state law entirely. The reliable fix is updating the form.
• A deceased beneficiary with no contingent named. The account may default to the estate, which forfeits favorable distribution treatment and drags the asset into probate.
• A job change. A rollover into a new IRA does not carry the old designation with it. The new account starts blank or defaults to whatever the custodian’s standard is.
• Naming the estate as beneficiary. Almost always the worst option for a retirement account. It generally eliminates the beneficiary’s ability to use the more favorable distribution rules and exposes the asset to probate and creditors.
• Naming a minor directly. Minors cannot legally control an inherited account, which typically means a court-appointed guardianship — expensive, public, and easily avoided with proper planning.
• Outdated trust language. Trusts drafted before 2020 may not work well under the current ten-year rules for inherited retirement accounts.

Events that should trigger a review immediately

Marriage. Divorce. A birth. A death. A job change or plan rollover. Opening any new account. A change in your estate plan. A disability or special-needs situation arising in the family.

None of these updates a beneficiary form automatically. Every one of them can invalidate the reasoning behind the last update.

Death benefit options, compared honestly

Every option that protects your beneficiaries reduces your income. Here is what each one costs and who it actually suits.

When lifetime income begins, you choose what happens to the money if you die early. Every choice sits on the same spectrum: more protection for your heirs, less income for you.

There is no right answer. There is only the answer that matches what these particular dollars are for.

Life only

Payments continue for your lifetime and stop at your death. Nothing goes to beneficiaries.

The highest payment of any option, sometimes by a wide margin, because the insurer has no residual obligation.

Suits: someone whose legacy objective is covered by other assets, or a single person maximizing income with no beneficiary concern. It is also the option most often rejected reflexively when it was in fact the right choice — if other assets handle the inheritance, paying for a death benefit here is paying twice.

Life with period certain

Payments continue for life, with a guaranteed minimum number of years — commonly ten or twenty. Die in year four of a twenty-year certain period, and your beneficiaries receive the remaining sixteen years of payments.

Suits: someone who wants protection against dying shortly after purchase without giving up much income. A ten-year certain period typically costs modestly against life only; twenty costs more.

Cash refund

Payments continue for life. If you die before total payments equal your premium, the difference is paid to your beneficiaries as a lump sum.

Suits: someone whose main objection is the possibility of the insurer keeping the money. It guarantees you or your heirs receive at least what you paid in, and it removes that objection cleanly. The cost is a permanently lower payment.

Installment refund

The same guarantee, paid as continuing installments rather than a lump sum. Slightly higher income than cash refund, since the insurer keeps the money longer.

Joint and survivor

Payments continue across two lifetimes, often at a reduced percentage after the first death — 100%, 75%, two-thirds, or 50%.

Suits: nearly every married couple relying on this income for living expenses. The reduction while both are alive is real, and it purchases protection against the moment household income drops.

How to choose

1. What are these dollars for? If they exist to produce income you will spend, protective options may be buying something you do not need.
2. What else will your heirs receive? If a house, a taxable account, or life insurance already handles the inheritance, this contract may not need to.
3. What does the protection actually cost? Get the income figures side by side. Sometimes ten-year certain costs very little; sometimes a refund option costs a great deal. The numbers vary by carrier, age, and rate environment.
4. Are you married? If this income covers essential expenses, the joint-and-survivor question is generally not optional.

One structural note: annuities pass by beneficiary designation, outside probate, which means whatever your heirs receive typically reaches them privately and without court involvement. Confirm the beneficiary form is current, since that form — not your will — controls the outcome.

The four families of annuity, and the job each one does

Hundreds of product names on the market, and only a handful of underlying structures. Learn to sort any contract and most of the confusion disappears.

Annuity marketing generates product names at a remarkable rate. Underneath them sit a small number of structures, and almost every contract you will be shown belongs to one of four families. Identify the family and you know what questions to ask.

One: immediate income annuities (SPIA)

You hand over a lump sum. The insurer begins paying you, typically the following month, for life or for a defined number of years.

This is the purest form of the product — a pension you purchase. It produces the most guaranteed income per dollar available anywhere, because part of what funds your payment is the pooling of longevity risk across many contract owners rather than only your own money and its return.

There is no account value, no statement showing a balance, and generally no access to the principal. What you have is a payment obligation. Refund and period-certain options can guarantee a minimum total payout to your beneficiaries, and each reduces the payment somewhat.

Best suited to: covering an income gap now, with money you have deliberately decided is no longer needed for anything else.

Two: deferred income annuities (DIA)

The same structure, purchased today, with income beginning on a future date you select. Waiting produces a substantially larger payment, because the insurer holds the money longer and expects to pay for fewer years.

The QLAC is a regulated variety of DIA purchased inside an IRA or employer plan, with the additional feature that the premium is excluded from the balance used to compute required minimum distributions.

Best suited to: someone in their sixties who wants to guarantee income for their eighties and beyond, particularly when the RMD carve-out is also useful.

Three: fixed and fixed indexed annuities (MYGA / FIA)

These are built for accumulation first, with income available later — often through an optional rider. Principal is protected and interest is credited, but the two varieties credit it very differently.

A multi-year guaranteed annuity declares a fixed interest rate for a set term, functioning much like a bank certificate of deposit with tax deferral and without FDIC insurance. Simple and easy to compare across carriers.

A fixed indexed annuity credits interest linked to the movement of an index. You do not own the index and receive no dividends from it. In a year the index declines, the credited interest is zero rather than negative — principal is not reduced by index losses. The upside is limited by caps, participation rates, or spreads, and those limits are how the protection gets funded.

Best suited to: money that needs to grow but cannot afford to lose ground, on a horizon that comfortably exceeds the surrender period.

Four: variable annuities (VA)

Your money goes into sub-accounts that are invested in the market. The account value rises and falls with them. There is no principal protection in the base contract; guarantees come from optional riders purchased at additional cost.

Variable annuities are registered securities and must be sold with a prospectus. They also carry the most layered cost structure — a mortality and expense risk charge, administrative fees, the operating expenses of the underlying sub-accounts, and any rider charges on top.

Best suited to: someone who wants market participation inside a tax-deferred wrapper and values a specific guarantee enough to pay for it.

Sorting any contract in four questions

1. When does income start — now, on a chosen date, or only if I elect it later?
2. Can the account value go down because of market movement?
3. Is there an account value at all, or only a payment obligation?
4. Was I handed a prospectus? If yes, it is a registered security — variable or RILA.

Caps, participation rates, and spreads

Indexed contracts rarely show a fee on your statement. That does not mean there is no cost — it means the cost lives somewhere you have to know to look.

A fixed indexed annuity protects your principal from index declines and credits interest when the index rises. Someone has to pay for that protection, and it is not the insurance company donating it.

The cost appears not as a deduction but as a limit on how much of the index’s movement reaches your account. There are three mechanisms, and a contract may use more than one at the same time.

The cap rate

A ceiling on credited interest for the period, regardless of how far the index rose. With a cap of 7%, an index gain of 20% credits 7%. An index gain of 4% credits 4%.

Caps are the most common mechanism and the easiest to compare across contracts — which also makes them the number most heavily marketed.

The participation rate

The percentage of the index’s movement used in the calculation. At a 60% participation rate, a 10% index gain is treated as 6%.

Participation rates are frequently paired with uncapped strategies, which sound generous. An uncapped strategy at 40% participation may credit less than a capped strategy at 100% participation, depending on how the index actually moves. Neither number means anything by itself.

The spread, margin, or asset fee

An amount subtracted from the index movement before interest is credited. With a 2% spread, a 7% index gain credits 5%, and a 2% gain credits nothing.

Three names for the same idea, which is worth knowing so it is recognizable whatever the contract calls it.

Two more things that reduce the number

Dividends are excluded. Index-linked crediting almost always tracks price movement only. The dividend component of an index’s total return — historically a meaningful share of it — is not credited. This is rarely stated prominently and it matters when comparing to a market return.

The crediting method changes the outcome. Annual point-to-point compares the index on two dates a year apart. Monthly sum adds up monthly changes, usually with a cap on the upside of each month but no floor on the downside — which can produce zero in a volatile year that finished higher. Monthly average smooths the path. The same index and the same cap can produce very different results under different methods.

How to compare two contracts properly

1. Which index, and which specific version of it?
2. Which crediting method — annual point-to-point, monthly sum, monthly average?
3. Which limiting mechanisms apply, and are any of them stacked?
4. What are the current rates, and what are the contractual guaranteed minimums?
5. Are these rates locked for a term, or resettable annually?
6. Does electing an income rider change any of them? It frequently does.

Account value versus benefit base

Two numbers on one statement, and only one of them is money you can actually withdraw. This is the most consequential misunderstanding in the annuity world.

If you own an annuity with an income rider, your statement probably shows two figures. They can differ substantially, and the larger one is usually the one that is not real in the way people assume.

The account value

Your actual money. Premium paid, plus credited interest or investment performance, minus withdrawals and charges.

This is what you would receive if you surrendered the contract, and it is generally what passes to your beneficiaries. It is the number that answers “how much do I have?”

The benefit base

A calculation figure. It exists to determine two things: the guaranteed amount you may withdraw each year for life, and the annual fee you pay for the rider.

Benefit bases often grow at an attractive contractual rate during deferral — sometimes described as a roll-up. That growth is genuinely valuable, because it increases the lifetime income the contract will eventually guarantee.

What it generally is not is money you can withdraw as a lump sum, surrender for, or leave to your heirs.

Why this causes so much trouble

Consider a contract where the benefit base has grown well above the account value. The owner sees the larger number, understandably concludes they have that much money, and plans accordingly. Then they need a lump sum, request a surrender, and receive the account value instead — often considerably less.

Nothing improper has happened. The contract worked exactly as written. But the owner spent years believing something that was never true, and by then it is too late to have planned differently.

Details worth confirming in your own contract

• Is the roll-up simple or compound? Both are quoted the same way and they diverge substantially over ten or fifteen years.
• How long does the roll-up run? Frequently capped at ten years, or ending when withdrawals begin, whichever comes first.
• Which value is the rider fee charged against? If it is the benefit base — common — then the fee grows as the base grows, and it is deducted from your account value. In a flat market this can shrink your real money while the marketing number climbs.
• Does an excess withdrawal reduce the benefit base? Almost always, and often more than proportionally. Taking more than the guaranteed amount in one year can permanently reduce guaranteed income.
• What do beneficiaries receive? Usually the account value or a stated death benefit — rarely the benefit base.

The question to ask

When any figure is presented to you, in a statement or an illustration or a meeting, ask directly: is that my account value or my benefit base? Anyone competent will answer immediately. It is a fair question and it should never be an uncomfortable one.

Reading a surrender charge schedule

A row of declining percentages, and everything you need to know about how long your money is genuinely committed.

Every deferred annuity contract contains a table that looks something like this: 9, 8, 7, 6, 5, 4, 3, 2, 1, 0. Those are the percentages charged if you withdraw more than the contract permits, in each year of the surrender period.

It is the single most useful table in the document, because it tells you exactly how long the money is committed and what leaving early would cost.

How to read it

The first number is year one, and each subsequent number applies to the following contract year — measured from your purchase date, not the calendar. A ten-entry schedule means a ten-year commitment.

The charge applies only to amounts withdrawn above the free withdrawal allowance, not to the whole contract. Taking out slightly more than permitted does not incur a charge on the entire balance.

The free withdrawal provision

Most contracts allow a defined amount each year without any charge — commonly around ten percent of the value, though terms vary considerably.

Two things to confirm: whether it is available in the first contract year, since some contracts do not allow it until year two, and whether unused allowance carries forward, which it usually does not.

The waivers — the most overlooked part of the contract

Most contracts waive the surrender charge entirely in defined circumstances. Common triggers include confinement to a nursing home or extended care facility, diagnosis of a terminal illness, disability, and in some contracts unemployment.

These are real liquidity features and they go almost entirely undiscussed at the point of sale. Ask specifically which waivers your contract includes and what conditions trigger them. Terms differ meaningfully between carriers, and for someone weighing a long surrender period, a strong waiver package can change the decision.

What the schedule is actually telling you

Surrender charges exist for a structural reason. To fund long-dated guarantees, an insurer invests in long-dated assets. The schedule is what permits that commitment, and therefore what permits the contract to offer terms a fully liquid account cannot.

Which means a longer schedule is not automatically worse — it often buys a better rate or a stronger guarantee. It is a trade, and the right response is to size the contract so that you never need to trigger a charge, rather than to fear the table itself.

Should you 1035 exchange an older contract?

Tax-free does not mean free. What has to genuinely improve before restarting a surrender period makes sense.

Section 1035 of the tax code permits one annuity contract to be exchanged for another without triggering current income tax. Cost basis carries forward, and no taxable event occurs.

That is a useful provision, and it is also the single most heavily promoted transaction in the annuity business — because an exchange generates a new sale. Both things are true, and the second is why the analysis deserves care.

What an exchange costs even when it is tax-free

• A new surrender period. The most significant cost. A client eight years into a ten-year schedule who exchanges may restart at year one of a new ten-year schedule — committing money for another decade.
• Surrender charges on the way out, if the existing contract is still within its period.
• Loss of legacy guarantees. This is the one most often missed. Contracts written in higher-rate eras sometimes carry guaranteed minimum interest rates, roll-up rates, or annuitization factors that no carrier offers today. Those provisions can be worth a great deal and are gone the moment the contract is surrendered.
• New underwriting on any riders, and new pricing based on your current age and health.

When an exchange is genuinely justified

There are real cases, and they share a common feature: something material improves that cannot be obtained any other way.

• The existing contract is out of its surrender period and current rates or caps are meaningfully better.
• A needed feature simply does not exist in the old contract — a lifetime income guarantee, a long-term care benefit, a death benefit structure the plan requires.
• The issuing carrier’s financial strength has deteriorated materially.
• The current contract carries high ongoing fees for riders you have decided you will not use.
• Consolidating several small contracts genuinely simplifies administration and pricing.

The protections that exist

Annuity replacements are regulated in every state, and for good reason. You should expect to complete a replacement disclosure form, and your existing carrier will generally be notified and given an opportunity to contact you.

Treat this paperwork as protection rather than friction. If anyone frames the replacement forms as a formality to be rushed through, that is worth noticing.

Five questions before agreeing to an exchange

1. What specifically does the new contract do that my current one cannot?
2. What guarantees am I giving up — including guaranteed minimum rates written years ago?
3. What is the new surrender schedule, and what is left on my current one?
4. What is the total compensation on this transaction, compared with leaving the contract in place?
5. Can I see both contracts compared on guaranteed values, not illustrated ones?

If the answer to the first question is vague, or if the case rests mainly on the new illustration looking better, the exchange probably is not warranted.

How to read financial strength ratings

Every guarantee in your contract rests on the issuing company. The ratings are public, free, and take about five minutes to check.

An annuity guarantee is a promise from an insurance company, backed by that company’s ability to pay claims. No government agency insures it the way the FDIC insures a bank deposit. Which company issues your contract is therefore not a detail — it is part of what you are buying.

Four independent agencies rate insurers on financial strength, and all four publish their ratings free.

The four agencies

• AM Best — focused specifically on insurance and the most widely cited in the industry. Scale runs A++, A+, A, A-, B++, B+, and downward.
• Standard & Poor’s — AAA, AA+, AA, AA-, A+, A, A-, BBB+, and downward.
• Moody’s — Aaa, Aa1, Aa2, Aa3, A1, A2, A3, Baa1, and downward.
• Fitch — a scale similar to S&P.

What to look at beyond the letter

• The outlook. Agencies attach a stable, positive, or negative outlook indicating the likely direction. A negative outlook on a strong rating is worth noticing.
• Ratings from more than one agency. Broad agreement is reassuring. Meaningful disagreement is worth understanding.
• The trend. A company upgraded over the past decade and one downgraded to the same level are not in the same position.
• Which entity is rated. Insurance groups contain many subsidiaries. The rating that matters belongs to the specific company issuing your contract, which is not always the one whose name appears on the brochure.

How to check

All four agencies allow free public searches, though some require a no-cost registration. Search the exact legal name of the issuing company as it appears on your contract or application — not the marketing name.

For a long-dated obligation, this is a reasonable five minutes to spend, and it is one of the few pieces of due diligence you can complete entirely on your own.

Keeping it in proportion

Ratings are opinions about the likelihood of meeting obligations. They are not guarantees, they change over time, and highly rated companies have failed before.

They remain the best independent, publicly available measure of the strength behind a promise you may rely on for thirty years. Use them as one input rather than the only one — alongside the company’s size, its history in the annuity market, and how long it has been issuing the kind of contract you are considering.