Most of what gets written about annuities is written by someone selling them or someone who hates them. Neither is helpful when you have actual money and an actual life to plan for.
I have spent thirty years sitting across from people trying to figure out whether annuities belong in their retirement plan. The honest answer is: sometimes. Not as often as the brochure says. Not as rarely as the column in the Wall Street Journal says. The decision depends on what you’re actually trying to solve, what else you own, and what you are willing to live with.
What an annuity actually is
An annuity is a contract with an insurance company. You give them money — once or over time — and in exchange they promise to pay you money back, on terms specified up front. That’s it. Everything else is variation on that one idea.
There are simple annuities and complicated annuities. Simple ones (a single-premium immediate annuity, or “SPIA”) trade a lump sum for income for life. Complicated ones (variable annuities, fixed-indexed annuities) add features — market participation, principal protection, optional riders — and, because complexity costs money, they cost more.
The complexity is not inherently bad. Sometimes it solves a real problem. But complexity is never free, and a complicated annuity that solves no problem you actually have is an expensive way to feel safe.
The two real problems annuities solve
Longevity risk. The risk that you live longer than your money was planned to last. A SPIA is the cleanest way to handle this — at the cost of giving up principal, you can buy a paycheck for life. For someone with a healthy life expectancy and modest other guaranteed income, this can be the most efficient income tool there is.
Behavioral risk. The risk that you, faced with a forty-percent drawdown in year three of retirement, do something you’ll regret. A portion of guaranteed income — Social Security plus an annuity — can lower the emotional stakes of the rest of the portfolio enough that you stop trying to manage it through fear. That has real economic value, even if it doesn’t show up in a Monte Carlo.
The annuity isn’t the solution. The certainty is the solution. The annuity is one way to buy it.
Where annuities don’t earn their place
If you already have enough guaranteed income to cover essential expenses — pension plus Social Security plus rental income, say — you may not need any more. A second layer of certainty costs the same as the first but solves less. The math is unforgiving on this.
If your portfolio is large enough that the failure rate at a reasonable withdrawal level is already small, an annuity may just be expensive insurance against an event that wasn’t going to happen.
If liquidity matters more than guaranteed income — because of business interests, large unfunded liabilities, or the desire to make a significant gift to children — locking up principal in an annuity is the wrong tool.
If you don’t understand the contract, you don’t own it yet. You can buy an annuity that promises something you can’t articulate; you can also buy one with a surrender period that quietly removes options for the next decade. Both are common. Neither is okay.
How I think about it in practice
Three questions, in order:
-
What gap are we trying to fill? Income to cover essentials, longevity protection, or behavioral protection? Each implies different products — and sometimes no product at all.
-
Have we maximized what’s already free? Social Security claiming strategy alone can produce more guaranteed lifetime income than most people realize. We exhaust this first.
-
If a product makes sense, what’s the simplest version that solves it? A SPIA is usually simpler than a variable annuity with a guaranteed lifetime withdrawal benefit. Simpler is usually better. Where complexity is needed, we account for what it costs.
The shape of a good decision
A good decision about an annuity looks like this: you can describe, in your own words, what problem it solves; you can describe what it costs; you have compared it to the alternative of doing nothing; and you have compared it to the alternative of solving the problem some other way. If you can do all four, the decision is yours to make.
If you can’t, the decision isn’t ready yet. That’s almost always the right place to stop.
— Bruce