The word "annuity" covers two products that have almost nothing to do with each other.
An income annuity is a trade. You hand over a sum, and an insurer pays you a fixed amount every month for as long as you live. If you live to 100 they keep paying. If you die at 68 they stop. That is the whole product.
An accumulation contract is a savings and investment wrapper with an insurance company's name on it — money grows inside it under a formula, with charges, guarantees, riders and a schedule of penalties for taking it out early. It may eventually be converted into income. Usually it is not.
They are sold by the same people under the same word, and confusing them is the single most expensive mistake in this subject. The rest of this chapter treats them separately.
Start with the honest steelman, because it is unusually strong.
There is one financial risk that no portfolio, savings rate or investment strategy can remove: you do not know how long you will live. It is not a market risk. You cannot diversify away your own lifespan, and you cannot hold enough for the worst case without massively underspending in every other case.
An income annuity is the only retail product that transfers it. The mechanism is real economics, not marketing: everyone in the pool pays in, the people who die early fund the people who live long, and that transfer — the mortality credit — is a genuine gain that cannot exist outside a risk pool. No fund can replicate it, because no fund can pay you with money belonging to people who died.
The pricing evidence is good and it is favorable. Because annuities have no claims, they have no loss ratio; the equivalent measure is the money's worth ratio — the expected present value of the payments against the premium. For a 65-year-old on current pricing, a single-premium immediate annuity returns roughly 87 cents per premium dollar valued with general-population mortality, and close to a dollar valued with the mortality of a typical annuity buyer.1
The gap between those two numbers is not profit. It is adverse selection — healthy people buy annuities, and the price reflects it. Which means the product is close to fairly priced for exactly the person who should be buying one.
Set that against how many do. Only 3.5% of households headed by someone aged 65 to 74 own an annuity, and 2.8% of those aged 75 to 84.2
Most people under-buy the one product that solves the one problem they cannot solve any other way. That, not mis-selling, is the biggest annuity problem in the country.
Now the other half. Total US annuity sales ran $464.1 billion in 2025.3
Single-premium immediate annuities — the product just described, the one with the good valuation evidence and the unique mechanism — were about 3.1% of that. Fixed indexed and registered index-linked contracts together were about 45%.3
The product with the strongest case is a rounding error. The products with the most complicated formulas are nearly half the market.
That inversion is the fact to hold onto. It is not evidence that indexed contracts are frauds — some suit some people. It is evidence that what gets sold is not determined by what the evidence supports.
An income annuity can be compared on one number. Give three insurers the same age, sex, state and sum, and each returns a monthly income. Higher wins. It takes an afternoon.
That comparison is worth real money: across a national sample of eight carriers the spread between best and worst was about 4%, rising to about 12% between the best and worst state-specific quotes, and over 17% on deferred contracts.1 Shopping is worth more the more deferred the product.
Now try comparing two indexed contracts. One has a cap on the index return, another a participation rate, another a spread. There are surrender schedules of different lengths, bonus credits with conditions, and optional riders with their own fees and their own benefit bases that are not the same thing as your money. Several of the terms can be changed by the issuer after you buy.
There is no single number. There cannot be. And that is the point:
If you cannot compare two quotes on one number, the difference between them is not the product — it is the price you cannot see.
Two things, both sourced, and worth keeping in proportion.
The sales-practice history is real. A regulatory sweep of free-meal investment seminars ran 110 examinations between April 2006 and June 2007. It found 23% involved possibly unsuitable recommendations and 13% showed indications of possible fraud. Only 4% were clean. 57% — 63 of the 110 — used exaggerated or misleading advertising claims. The products most often involved were variable annuities, REITs and equity-indexed annuities. One ad promoted a 38% return with no risk in a way implying a government guarantee; in another case a representative placed about 80% of a customer's net worth into variable annuities.4
That examination is now old, and it should be read as history rather than as a description of today. What gives it weight is its scope — a coordinated national sweep of 110 firms and branch offices — and it explains why the rules that followed exist.
The current record is about exchanges. Four enforcement actions on the supervision of annuity exchanges landed inside six months:5
customers moved into more expensive variable annuities, at an average incremental cost of $8,718.86 each. The finding was a failure to supervise, not a finding about any individual recommendation.
over 22 exchanges affecting 14 customers by one representative.
Four cases in six months is not proof that exchanges are usually wrong. It is proof the practice is under active scrutiny, which tells you where to look.
And on incentives, a Senate committee report in September 2024 catalogued sales incentives offered by at least 29 companies — six days in Playa Mujeres for $200,000 of production, a Danube cruise, a week in Australia, a five-night Venice trip for a carrier's top 40 producers.6 Treat that as an advocacy document rather than a neutral regulator finding; it is evidence about what is offered, not a measurement of harm.
The state of the rules matters and changes, so here is where it stands in August 2026.
Annuity sales are covered by a best-interest standard in most states. The model regulation requires a producer to act in the consumer's best interest, exercise reasonable diligence, care and skill, and not put their financial interest ahead of yours.7 New York runs its own equivalent regulation instead, and California moved to a best-interest standard for annuity transactions from 1 January 2025.8
Life insurance sales are not covered by that standard. The model regulation's own scope section applies it to "any sale or recommendation of an annuity".7 This is a scope fact worth knowing, not an accusation.
The federal fiduciary rule is dead. The Retirement Security Rule never took effect; the courts entered final judgments in March 2026 and the Department published notice of the vacatur that month. The 1975 five-part test governs again.9 Anyone telling you a federal fiduciary duty applies to an annuity sale is describing a rule that was struck down.
You can ask what they are paid. The model regulation entitles you to a disclosure of the producer's cash compensation on request.7 Nobody volunteers it. The question is allowed and it is the single most informative thing you can ask.
There is a free-look window to cancel after buying. In California it is 30 days for buyers aged 60 or older.10 Windows vary by state and product — find yours before you sign, not after.
If the insurer fails, state guaranty associations cover annuity benefits. The model standard is $250,000 of present value, and states vary, some higher.11 Buying more than the covered amount from one insurer is a decision, not a detail.
The moment most likely to produce a bad annuity purchase is not old age. It is a lump sum arriving alongside grief or an ending — a death benefit, a retirement rollout, a settlement. Roughly five million savers a year move money out of workplace plans, and $779 billion was rolled into IRAs in 2022.12 A government economic analysis estimated that conflicted advice on fixed indexed annuities alone might cost savers up to $5 billion a year.12
There is a simple defense and it costs nothing: no irreversible financial decision within some months of a bereavement. Money in a savings account is not going anywhere. A surrender schedule is.
The question an annuity actually answers is whether your guaranteed income covers your committed costs, and how large the gap is. Plenee already holds both halves — what comes in reliably and what has to go out every month regardless. That gap is the honest size of any income annuity you might need, and it is a number rather than a feeling.
Plenee will not tell you to buy one. It can tell you whether the problem the product solves is one you have.
Two products share one word. The income version solves the only financial risk you cannot diversify away, prices close to fair for the people who should buy it, and is about 3% of the market. The accumulation version is nearly half the market and cannot be compared on any single number, which is the fact to reason from. Annuities have no loss ratio, so use the money's worth ratio instead. Ask what the seller is paid — you are entitled to know. Check whether a guaranteed benefit is actually your money. And make no irreversible decision in the months after a death.
Plenee Academy provides financial information and education, not personalized financial advice. Plenee Co. is not a registered investment adviser, broker-dealer, or financial planner. Legal Disclosures & Notices →