Revisiting Annuity Income

There are moments in markets when two things that normally compete for attention become interesting at the same time. We happen to be in one of those moments now. The stock market has had a strong year, with the S&P 500 reaching another record closing high on October 6, 2026, while long-term U.S. Treasury yields remained above 5% as of October 7.[4][5] Neither of those facts tells us what happens next, and I would be suspicious of anyone who claims otherwise with great confidence. Markets have been humiliating confident people for centuries. But the combination does create an unusually interesting question for people who are already retired or close to it: if a portion of your portfolio has appreciated substantially, and today’s interest-rate environment can support higher lifetime-income payouts from certain fixed annuities than lower-rate environments, is this a reasonable time to review whether part of that money should be assigned a different job?

That question has been on my mind because of a paper published recently by Joshua Gotbaum and Olivia S. Mitchell through the Brookings Institution’s Retirement Security Project. Their argument is that our retirement system still does a surprisingly poor job of helping people turn retirement assets into lifetime income. Only about 10% of defined-contribution plans offer lifetime-income payout options, according to research cited in the paper, and the authors propose making partial annuitization a default inside retirement plans. Their rough framework would convert somewhere around 20% to 25% of eligible assets above a threshold into lifetime income, while still giving the retiree the ability to change the amount or decline it altogether.[1] That is a policy proposal, not a recommendation that any particular retiree annuitize 20% to 25% of assets.

I am not particularly interested in whether 23% happens to be the magical percentage or whether Washington ever adopts their proposal. I am interested in the problem they are trying to solve, because I see a version of it sitting across the table from me all the time.

A person can have a substantial investment portfolio and still not feel as financially secure as someone receiving a large pension every month. That sounds odd until you think about how the money actually arrives. A pension shows up in the checking account. Social Security does the same. Nobody wakes up on the first of the month, sees the pension deposit and thinks, “Well, there goes another piece of my principal.” It feels like income because that is how it behaves. An investment portfolio behaves differently. If you need $15,000, something has to be sold or cash has to be transferred, and however routine that transaction may be, there is a little subtraction problem taking place in plain sight. The statement used to say one number. Now it says a slightly smaller number. Human beings notice these things.

This is where an income annuity can play an interesting role, not because annuities are universally good investments or because everybody should own one, but because certain annuities can perform a job that a portfolio cannot perform in quite the same way. Certain fixed annuity contracts can provide contractually guaranteed lifetime payments for as long as someone lives, subject to the terms of the contract and the financial strength and claims-paying ability of the issuing insurer. A portfolio can be carefully designed to support a lifetime of withdrawals, and we can model those withdrawals six ways from Sunday, but it cannot promise the date of death. An insurance company can pool that uncertainty across thousands of people. That is fundamentally what you are buying.

It is important to be precise here because “annuity” has become one of those financial words that describes so many different products that saying you like or dislike annuities is a little like saying you like or dislike vehicles. A pickup truck and a school bus have some obvious similarities, but I probably would not recommend them for the same trip. When I am talking about annuities in the context of this article, I am primarily interested in fixed insurance contracts designed to create lifetime income, such as fixed immediate income annuities, fixed deferred income annuities, or fixed annuity contracts with lifetime-income guarantees. References here to guaranteed, dependable, steady, or reliable income refer to contractual guarantees of fixed insurance products, not to securities or investment-account values. A fixed annuity that simply credits interest for several years can certainly have a role in a plan, but owning one does not automatically solve the lifetime-income problem.

Today’s rate environment makes that conversation more interesting. The amount of lifetime income an insurer can offer depends on several factors, including the purchaser’s age, whether the income covers one life or two, the contract features selected and prevailing interest rates. Generally speaking, higher interest rates support higher income-annuity payouts because insurers can earn more on the assets backing those contractual obligations. Fidelity made this point explicitly in research published this summer: as rates rise, income-annuity payouts generally improve; if rates fall, the opposite can occur.[2] That does not mean today’s rate is necessarily the best rate we will ever see. Nobody knows that. It simply means that the broad rate environment is different when long-term Treasury yields are above 5% than it was when rates spent years hovering at extraordinarily low levels. Actual annuity quotes can vary materially by insurer, purchaser characteristics, contract features and the date of purchase.

And that becomes particularly interesting when the other side of the balance sheet has also been cooperative. Stocks have had a strong run, and the S&P 500 reached a record closing high on October 6, 2026.[5] Again, that is not an argument that stocks are about to fall. Record highs have a funny habit of being followed by other record highs. But retirement planning should not require us to predict the date of the next correction before we are allowed to rebalance risk.

If a client owns substantially more equities today because the market has appreciated, and at the same time higher prevailing rates can support higher lifetime-income payouts than lower-rate environments, I think it is perfectly reasonable to ask whether the current allocation still matches the household’s income needs, liquidity requirements, risk tolerance, tax situation and legacy goals. We are not talking about fleeing the stock market. We are asking whether money that previously had one job, growth, might appropriately have a different job within the plan, including producing fixed lifetime income under an insurance contract.

That distinction matters because the goal should not be to annuitize a portfolio. It should be to solve a problem.

Suppose a retired household has regular living expenses that comfortably exceed Social Security and whatever pension income they receive. That difference is the income gap. There is nothing inherently wrong with filling the entire gap through portfolio withdrawals, and in many cases that will remain the best solution. But if every mortgage payment, utility bill, insurance premium, grocery run and property-tax payment ultimately requires a withdrawal from investments, the retiree remains unusually aware of the market’s mood. A bad twelve months in stocks is no longer merely an investment experience. It is happening at the same time the portfolio is being asked to write the paycheck.

Now imagine that some portion of that gap is filled with lifetime income instead. The portfolio does not disappear. Most of the assets may remain invested, available for travel, large purchases, future healthcare costs, family assistance, charitable giving and eventually an estate. But the amount the portfolio must produce every month has been reduced. That changes both the mathematics and, perhaps just as importantly, the psychology of the plan.

This is one reason the Brookings research caught my attention. The authors cite work suggesting that retirees with lifetime income experience less anxiety about running out of money and may be more comfortable spending what they have. They also cite research estimating that, under modeled assumptions, partial annuitization could allow baby boomers collectively to spend approximately $27 billion more per year in retirement.[1] I would not get too hung up on whether the exact number eventually proves to be $27 billion. The more interesting point is why spending could increase at all. The households did not suddenly become richer. The form of part of their wealth changed, and their behavior changed with it.

That is useful information for planners because some of the wealthiest retirees I know are not particularly good at spending money.

They are not afraid that dinner is going to bankrupt them. What bothers them is the slow, visible erosion of an account they spent years mentally identifying as security. A large portfolio balance feels safe. A somewhat smaller one can feel strangely less safe, even when the financial plan could barely distinguish between the two. Put dependable income underneath part of someone’s lifestyle, however, and the experience can change. The question becomes less about whether every vacation requires permission from the portfolio and more about whether the income architecture and remaining assets continue to support the life they planned.

None of this makes an annuity free money, and this is where the conversation has to remain balanced. An income-annuity payout should not be confused with an investment yield. Depending on the structure, some of what is being paid back may represent a return of the purchaser’s own capital, and the payout reflects longevity assumptions in addition to interest rates. In many immediate annuities, committing the premium also means giving up direct access to that capital. Adding survivor benefits, refund provisions, inflation adjustments or other protections can reduce the initial payment. Fixed payments can lose purchasing power over time if inflation remains high. Other annuity designs preserve more liquidity but may introduce additional costs, restrictions or complexity. Insurance products may also be subject to fees, surrender charges and holding periods that vary by contract, and they are not FDIC insured. Every contractual guarantee is ultimately dependent on the financial strength and claims-paying ability of the insurance company issuing it.[3]

Those are not footnotes to wave away. They are the trade.

You are exchanging something in order to receive something else. Liquidity may be exchanged for longevity protection. Some upside may be exchanged for predictability. A portion of an estate may be exchanged for a higher lifetime payment, depending on how the contract is structured. The decision becomes useful only when the thing you are receiving solves a problem that actually matters to you.

That is why I have little interest in treating annuities as an all-or-nothing philosophy. Brookings’ proposal is specifically built around partial annuitization, and I think that framework is useful to consider even outside the policy debate. A person with substantial wealth probably does not need every dollar guaranteed. In fact, doing so could destroy much of the flexibility that made having substantial wealth useful in the first place. The better question is whether there is a slice of the balance sheet where certainty is more valuable than additional upside.

For one person, the answer may be no. Social Security and pension income may already cover everything they expect to spend regularly, leaving the investment portfolio free to do what portfolios do best. For another family, there may be a meaningful gap between dependable income and the amount required to maintain their lifestyle, and filling part of that gap could make the rest of the plan work better. In that situation, the annuity is not being purchased because somebody has a heroic opinion about interest rates or thinks the stock market has reached its final high. It is being considered because a known retirement need may be addressed through a fixed insurance contract, and higher prevailing rates can support higher lifetime-income payouts per dollar of premium than lower-rate environments. Actual pricing varies by insurer and contract, and future rates are unknown.

That is a much more useful way to think about the current moment.

If markets continue higher from here, wonderful. We still own assets that participate.

If markets eventually pull back, which markets have occasionally been known to do without first asking permission, scheduled payments from a fixed lifetime-income guarantee are not determined by day-to-day stock-market moves, although the guarantee remains subject to the contract terms and the financial strength and claims-paying ability of the issuing insurer.

And if somebody lives to 98, the central risk we were trying to insure was not whether the market averaged 7% instead of 8%. It was the possibility that life lasted considerably longer than the spreadsheet’s neat middle estimate.

The older I get as a planner, the less interested I am in finding one product or one portfolio that supposedly solves retirement. Retirement income is a collection of different problems, and they do not all need the same tool. Investments are extraordinarily useful for growth, flexibility and legacy planning. Cash is useful for reasons that become painfully obvious at exactly the wrong moments. Social Security is an unusually valuable source of lifetime income for many households. And in the right circumstances, a fixed annuity can take a specific portion of the uncertainty off the table and replace it with a contractually guaranteed payment designed to continue for life, subject to the contract terms and the insurer’s claims-paying ability.

That last part is worth more than the spreadsheet sometimes gives it credit for.