How do annuities work and when are they useful?

The short answer

An annuity is a financial contract in which an insurer commits to pay a stream of income — typically monthly — for a defined period or for life, in exchange for an upfront lump sum or a series of premiums. The simplest version is the immediate single-premium life annuity (SPIA): pay X today, receive Y monthly until death.

The economic case for annuities is that they pool longevity risk across many retirees. A single retiree must plan to the upper percentile of life expectancy to avoid running out of money; an insurance pool can plan to the average, distributing the savings between long-lived and short-lived contract holders. The result is what economists call mortality credits — additional yield beyond what the same lump sum could safely generate self-managed.

Yet annuitization is rare. The annuity puzzle, named by Yaari (1965) and reinforced by Modigliani’s 1986 Nobel lecture, captures the persistent gap between theoretical efficiency and observed retiree behavior. Voluntary annuitization rates remain below 5% of US retirement assets despite decades of advocacy.

→ New to retirement income concepts? Financial education hub

What the data shows

The reference numbers come from LIMRA Secure Retirement Institute (2023-2024), academic papers on annuitization decisions (Brown 2007, Benartzi-Previtero-Thaler 2011), and the SEC investor education materials.

The reference numbers (LIMRA 2023-2024 and academic literature):

• Voluntary annuitization rate in US retirement assets — under 5% across most studies
• Mortality credit benefit — for a 65-year-old, an immediate life annuity typically delivers monthly payments roughly 30-50% higher than a 4% withdrawal from the same lump sum, depending on prevailing rates
• Defined-benefit pension prevalence in US private sector — under 15% of workers have access (BLS 2024), down from over 50% in 1980
• French Sécurité Sociale plus AGIRC-ARRCO acts effectively as a near-universal annuity, eliminating most voluntary annuitization need for full-career employees
• Bequest motive in surveys — over 60% of US retirees cite leaving an inheritance as important, partly explaining annuitization reluctance

The exception that complicates the picture: defined-benefit pensions and public pay-as-you-go systems are de facto annuities. Households with full-career French Sécurité Sociale or US Social Security entitlements are already heavily annuitized at the household level, which reduces the marginal value of additional voluntary annuitization.

→ Dataset: US personal savings rate dataset

Why it happens — the macro mechanism

The economic puzzle of annuity under-utilization combines welfare theory, behavioral economics, and product-design realities.

The first channel is the theoretical efficiency of risk pooling. Yaari (1965) showed that under standard utility assumptions, a risk-averse retiree without bequest motive should annuitize close to 100% of retirement wealth. The mortality credit — the financial benefit of pooling longevity risk — represents real welfare gain that cannot be replicated through self-management.

The second channel is the behavioral and informational frictions. Brown (2007) and Benartzi-Previtero-Thaler (2011) documented multiple frictions. The mental framing of an annuity decision (giving up a lump sum permanently) triggers loss aversion. The illusion of control bias makes self-management appear safer than insurer-managed pooling. Distrust of insurers and concerns about counterparty risk are persistent. Sales practices in the US insurance industry (commissions, complex riders, marketing pressure) have damaged the reputation of the product class.

A brief transition: this is why the academic literature has shifted toward partial annuitization (covering basic expenses while leaving discretionary spending in marketable assets) rather than full annuitization.

The third channel is the bequest motive and family structure. Annuities terminate at death, eliminating the lump sum from the estate. Households with strong bequest preferences face a real tradeoff between longevity insurance and intergenerational transfer. The asymmetry is real: rational welfare analysis can favor annuitization at the individual level while household preferences favor non-annuitization.

Synthesis by regime: in countries with generous public pay-as-you-go systems (France, Germany), the marginal annuitization need is small because public benefits already cover basic expenses for life. In countries with weak public benefits and limited employer-provided defined-benefit plans (US for younger cohorts), the theoretical case for voluntary annuitization is strongest, yet take-up remains low because of behavioral frictions. The transition parameter is the share of basic expenses already covered by mandatory life-contingent income — above 70%, additional annuitization is largely redundant; below 40%, the unaddressed longevity exposure is material.

Annuities solve a problem most retirees would rather not solve. The math says yes; the psychology says no — and the psychology usually wins.

→ Framework: Asset allocation and resilient portfolios

What it means for different economic actors

Retirees with weak public pension coverage face the largest theoretical case for annuitization, yet the behavioral frictions are also strongest in this group because the lump sum represents a larger share of total wealth.

Retirees with strong public benefits (full-career French and German employees, US Social Security maximizers) are already heavily annuitized at the household level. Additional annuitization adds limited welfare benefit and may not justify the loss of liquidity and bequest potential.

Single retirees without heirs face the cleanest case for annuitization because the bequest tradeoff is absent. The literature suggests this group should be the most natural annuitization audience, though take-up remains low even here.

A common error is to evaluate annuities purely on rate-of-return comparison with self-managed portfolios. The correct framing includes the mortality credit and the elimination of longevity-tail risk, which conventional return comparison misses entirely.

Practical observation

Go deeper

Related questions

• What is longevity risk?
• Why is Social Security claiming age critical?
• Why does the 4% rule face challenges?
• What is the retirement income gap?

Frequently asked questions

Why is annuitization so rare despite theoretical efficiency?

The annuity puzzle has multiple drivers. Loss aversion makes the upfront lump sum payment feel like a permanent loss. Illusion of control favors self-managed portfolios over insurer-managed pools. Bequest motives favor leaving assets to heirs. Counterparty risk concerns target insurer solvency. Sales practices in some markets, particularly the US, have damaged trust in the product class. Together, these frictions consistently outweigh the mortality credit in retiree decision-making, even when the mortality credit is substantial.

What are the main types of annuities?

Single-premium immediate annuities (SPIAs) are the simplest: pay a lump sum, receive lifetime income immediately. Deferred income annuities (DIAs) are paid for now but begin payments later, providing longevity insurance for advanced ages. Variable annuities link payments to investment performance, often with riders that complicate the structure. Fixed indexed annuities reference market indices but with caps and floors. The pure economic case for longevity insurance is strongest with SPIAs and DIAs; more complex products often dilute the insurance value with investment elements that do not pool risk efficiently.

How do French and US annuity markets differ?

The structural difference is that French households are already heavily annuitized through Sécurité Sociale and AGIRC-ARRCO, reducing voluntary annuity demand. The French market exists primarily for high-net-worth households seeking tax-efficient supplementary income. US households face larger uncovered longevity risk because of the decline in defined-benefit pensions, but voluntary annuity take-up remains low despite this. The structural takeaway is that public-pension generosity is the dominant determinant of voluntary annuity demand, with behavioral frictions explaining why take-up is below theoretical optimum even where the case is strongest.

Last updated — 4 June 2026