The Case Against Annuities Doesn’t Hold Up. Here’s the Research

Ask someone why they have not purchased an annuity and you will likely hear one of a handful of familiar objections: the fees are too high, they do not want to give up control of their money, they want to leave something for their heirs, or they already have Social Security. These objections feel intuitive. They circulate widely. And according to a new working paper from the National Bureau of Economic Research, most of them do not hold up.

The paper examines what economists call the “annuity puzzle” — the persistent gap between what lifecycle models suggest people should do and what they actually do. Those models consistently find that converting a meaningful portion of retirement savings into guaranteed lifetime income is rational for most people. Yet most people do not do it. The NBER paper asks why, and its answer is not flattering to the conventional objections.

The Fee Objection

The most common objection to annuities is cost. Annuities carry fees, the argument goes, and those fees eat into returns. This framing makes a category error: it compares an annuity to an investment product when an annuity is fundamentally an insurance product.

The relevant comparison is not between an annuity and an index fund. It is between an annuity and the cost of self-insuring against longevity risk — the risk of living longer than your savings last. Self-insuring against that risk requires holding significantly more assets than most people accumulate, because you have to fund a worst-case longevity scenario on your own. An annuity pools that risk across a large population, which allows it to provide guaranteed income at a cost that is almost always lower than what individual self-insurance would require.

The fee objection also tends to conflate different types of annuities. Simple income annuities — products that convert a lump sum into a guaranteed monthly payment — are straightforward and relatively low-cost. The complexity and expense that critics often cite applies to a much narrower subset of annuity products, not to the category as a whole.

The Liquidity Objection

The second common objection is liquidity. Annuities require converting savings into an income stream, and that conversion is largely irreversible. For people who want to preserve access to their money, this feels like a significant trade-off.

But this objection rests on a misunderstanding of what liquidity is actually needed for in retirement. Essential monthly expenses — housing, food, healthcare, utilities — do not require a liquid portfolio. They require reliable income. Annuities are specifically designed to provide that income, which means they are solving the right problem.

The fear of giving up liquidity also tends to overestimate how often retirees actually need access to large lump sums. Most large unexpected expenses in retirement — healthcare costs, home repairs, family needs — can be planned for through a combination of a smaller liquid reserve and reliable income that does not depend on market performance. Partial annuitization addresses this directly: convert enough savings to cover essential expenses with guaranteed income, and keep the remainder in a flexible account for discretionary spending and unexpected needs.

The Bequest Objection

The bequest objection is the most legitimate of the common arguments — and also the most overstated. Many people want to leave assets to their heirs, and an annuity that ends at death leaves nothing behind. For individuals with strong bequest intentions, this is a real consideration.

But the objection applies most forcefully to full annuitization — converting all savings into an annuity — which almost no one recommends. Partial annuitization, which covers essential expenses while keeping the remainder of the portfolio intact, preserves meaningful bequest potential while still providing the income security that an annuity delivers.

There is also a counterintuitive dynamic worth noting. Retirees who have guaranteed income covering their essential expenses tend to spend less from their investment portfolio, because they do not need to draw it down to meet monthly needs. The result is that annuitization often leaves more in the portfolio over time, not less — which is better for heirs, not worse.

The “I Already Have Social Security” Objection

A fourth objection holds that Social Security already provides enough guaranteed income, making an annuity redundant. This argument misunderstands both what Social Security provides and what most retirees actually need.

Social Security replaces a fraction of pre-retirement income for most people — typically somewhere between 30% and 50% for middle-income earners, and less for higher earners. For the majority of retirees, that replacement rate leaves a significant gap between Social Security income and the income needed to cover essential expenses. That gap is precisely what annuities are designed to fill.

The argument that Social Security is sufficient also ignores the uncertainty around its long-term benefit levels. The NBER paper’s broader research context includes widespread concern among near-retirees about whether Social Security will continue to provide full benefits. Relying on Social Security as a complete guaranteed income solution is a planning assumption that carries its own risk.

What the Research Actually Finds

Having examined and challenged each of the major objections, the NBER paper turns to what it finds is the real explanation for low annuity demand: institutional design failures, not product failures.

Financial literacy — knowing more about finance, compound interest, and longevity risk — does correlate with better retirement planning and a greater understanding of annuities. But it has only a weak correlation with actual annuity demand. Educating people more about annuities does not reliably lead to more people buying them. The paper concludes that literacy should be understood as a complement to good institutional design, not a substitute for it.

What does move the needle? Defaults. Framing. Guidance at the point of decision. When lifetime income is the default option in a retirement plan, participation rises significantly — without any change in financial literacy. When the conversation is framed around solving a specific problem — how to make sure income lasts as long as you do — rather than presenting a complex financial product with trade-offs to evaluate, engagement improves dramatically.

Common Objection	What the Research Shows
Fees are too high	Wrong comparison — annuities are insurance, not investments; self-insuring longevity risk costs more
Loss of liquidity	Essential expenses need income, not a liquid portfolio; partial annuitization preserves flexibility
Want to leave money to heirs	Partial annuitization preserves portfolio; guaranteed income often leaves more behind, not less
Already have Social Security	Social Security replaces 30–50% of income for most; annuities fill the gap that remains
Low financial literacy	Weakly correlated with demand; defaults and framing matter far more than education alone

The Framing Problem

The paper’s findings point to a conclusion that is both practical and important. The annuity puzzle — the gap between what people should do and what they actually do — is not primarily a product problem. It is a framing and design problem.

An annuity presented as “giving up control of your money in exchange for a monthly check” produces hesitation. The same product presented as “creating a guaranteed paycheck that covers your essential expenses for the rest of your life, no matter how long you live” produces a very different response. The product is identical. The conversation is not.

This matters for near-retirees who are evaluating their options. The discomfort many people feel about annuities is not a signal that the products are unsuitable. It is a signal that the conversation about them has not been structured well. When the conversation starts with the problem — how do I make sure I do not run out of income — rather than the product, the decision tends to become clearer.

What This Means for Retirement Planning

The NBER research does not make the case that annuities are right for everyone. There are genuine situations where the trade-offs do not favor annuitization — households with very high Social Security replacement rates, individuals with serious health conditions that reduce life expectancy, or those with specific bequest goals that cannot be met through partial annuitization.

But those situations are narrower than the conventional objections suggest. For the majority of near-retirees who face a meaningful gap between Social Security income and essential expenses, who are living longer than prior generations, and who no longer have access to a pension, the case for guaranteed income is strong — and the research supports it. The objections that have kept many people from engaging seriously with annuities do not hold up under scrutiny. What holds up is the underlying problem they are designed to solve.

Source: PSCA, covering research from the National Bureau of Economic Research. Read the original article.