Percevoir sa retraite en capital vs. sous forme de rente mensuelle
Si vous agissez
Prendre le capital en une fois
31%
Si vous n’agissez pas
Prendre la rente mensuelle
9,0%
Pourcentage de personnes qui regrettent ensuite chaque choix. Les barres et le registre complet s’affichent ci-dessous.
Financier
Dernière révision 2026-05-13
Qualité des preuves 4.38/5
Score d’évaluation en huit dimensions selon la
grille de qualité
. Chaque dimension notée de 1 à 5.
D1 Vérification des sources
4/5
D2 Autorité et indépendance des sources
5/5
D3 Précision du taux de regret
5/5
D4 Comparabilité des sources
4/5
D5 Motif de Gilovich
4/5
D6 Qualité de la prose
3/5
D7 Complétude des réserves
5/5
D8 Qualité de l’échantillon
5/5
Moyenne4.38/5
Données de substitution — aucune enquête directe sur les regrets n'existe pour cette décision. Les taux sont dérivés des scores de satisfaction et des obstacles d'accès plutôt que de questions portant directement sur les regrets. Voir les mises en garde ci-dessous.
Regret d'action
Prendre le capital en une fois
31%
31% des bénéficiaires d'un capital versé en une fois qui ont fait des achats importants regrettent leur décision
Retraités américains ayant opté pour un versement de capital de pension (enquête MetLife Paycheck or Pot of Gold)
enquête rétrospective, années variables après la retraite
Regret d'inaction
Prendre la rente mensuelle
9,0%
Le regret lié à la rente est faible et rarement mesuré directement ; dans une étude américaine, seuls ~9% des adultes âgés regrettaient de dépendre des autres, et bien plus de personnes regrettent de NE PAS disposer d'un revenu à vie que de le regretter (indicateur indirect)
Retraités américains recevant des paiements de rente de pension à prestations définies (EBRI 2023 Retirement Confidence Survey)
enquête actuelle, transversale
% regrettent ce choix
Prendre le capital en une foisPrendre la rente mensuelle
31%9,0%
action dominates — L'action domine — la plupart regrettent d'avoir agi.
Décisions associées
Décisions sémantiquement similaires — même terrain, compromis différents.
Porter le litige en justiceTransiger ou renoncer à toute action en justice
61%92%
L'inaction domine
Regret d'inaction 1.5× plus élevé
The lump-sum vs. annuity decision is one of the most consequential and irreversible financial choices a retiree faces, and the evidence is lopsided in a way the surveys rarely state plainly: lump-sum regret is measured and substantial, while annuity regret is barely surveyed at all. MetLife’s 2017 Paycheck or Pot of Gold Study (conducted by Harris Poll, reported by PLANADVISER) found that 31% of lump-sum recipients who made major purchases regretted the spending in hindsight, 21% of all lump-sum recipients said they had depleted the payout in an average of five and a half years, and 52% conceded an annuity would have made their budget more predictable. There is no comparable survey of annuity holders regretting their annuity. The closest peer-reviewed evidence — Hurwitz and Mitchell’s 2025 Health and Retirement Study experiment (N=1,764) — runs the other way: about 26% of older adults regretted NOT annuitizing and only ~9% regretted depending on others, and informing people of their objective survival odds raised regret about not buying lifetime income by 42%. Gallup’s 2022 survey of individual annuity owners likewise finds high satisfaction. The annuity-side rate shown here (9%) is therefore a low proxy, not a measured annuity-holder regret rate.
The failure modes are structurally different. Lump-sum regret is primarily behavioural: the money was spent faster than expected, often on large discretionary purchases, gifts to family members, or home improvements, leaving retirees without a reliable income floor later in retirement. Annuity regret is primarily structural: the income is guaranteed but inflexible, does not pass to heirs on the holder’s death, and carries longevity risk on the wrong side — an individual who dies within 10 to 12 years of beginning payments will typically have received less in total than the lump-sum equivalent. Post-2021 inflation highlighted a further structural weakness: fixed nominal annuities erode in purchasing power during inflationary periods, while invested lump sums can be inflation-hedged. EBRI’s Retirement Confidence Survey data shows that retirees with guaranteed income sources report higher retirement confidence and lower financial stress than those relying on account drawdowns — which is why the annuity choice, despite its structural downsides, is not associated with a high measured regret rate.
The action-dominates classification reflects that the only directly measured regret here is on the lump-sum side (about 31% among major spenders), while the annuity side has no comparable direct-regret survey and the related evidence points to low regret — so the headline gap (roughly 22 percentage points) should be read as directional rather than precise, since the two sides rest on different and partly proxy-based instruments. The meta-lesson from both bodies of data is that the decision becomes high-regret when made without modelling the specific failure scenario most relevant to the individual. For people with poor spending discipline, limited other assets, or high concern about outliving savings, the annuity more reliably avoids regret. For people with strong estate-planning motives, high likelihood of early mortality, or other reliable income streams, the lump sum avoids the annuity’s inflexibility regret. No single answer dominates across all personal circumstances.
Sources : action
Registre des sources
Chaque chiffre ci-dessous correspond à ce que la source a rapporté, avec la citation textuelle sur laquelle nous nous sommes appuyés et la méthode de calcul. Cliquez sur un lien pour vérifier directement.
1/2 sources vérifiées de manière indépendante, mot pour mot, par rapport à la source citée
[1]PLANADVISER — Lump-Sum Spending Often Followed by Regret (MetLife Paycheck or Pot of Gold Study)
Vérifié
Article de presse
31% of lump-sum recipients with major spending regret it in hindsight; 21% who took a lump sum say they depleted it in five and a half years on average; 52% concede an annuity would have made their budget more predictable; 23% who gave money away regret it
Extrait
“"Roughly one-third (31%) of those with major spending regret their spending in hindsight. A striking 21% of all participants who selected a lump sum at some point say they depleted it—taking just five and a half years on average to spend the dough. Fifty-two percent of participants who chose a lump sum concede that, if they had taken an annuity, their budget would be more predictable. 23% who gave money away lament their generosity."
”
Données source de
2017-04-11
Consulté le
2026-06-30
Vérification
Extrait récupéré et confirmé de manière indépendante, mot pour mot, par rapport à la source citée lors de notre audit de vérification.
Calcul
Trade-press coverage (PLANADVISER, 11 Apr 2017) of MetLife's 2017 Paycheck or Pot of Gold Study, an online survey of ~1,000 US defined-benefit and defined-contribution plan participants conducted by The Harris Poll for MetLife. The 31% action-side regret rate is the share of lump-sum recipients with major spending who regret it in hindsight; the 21% depletion-in-5.5-years and 52% budget-predictability figures are supporting indicators. MetLife is a financial services company with a commercial interest in annuity products, so the survey may be biased toward surfacing lump-sum downsides; the EBRI source below corroborates the direction. Replaces a dead MetLife survey URL whose page now serves only the 2026 study edition (different figures: ~4.4-year depletion). Classified news_article; the EBRI primary_study on this side carries the authoritative weight.
[2]Employee Benefit Research Institute (EBRI) — 2023 Retirement Confidence Survey Short Report
Étude primaire
Retirees who did not have guaranteed income sources (annuities, pensions) reported significantly lower retirement confidence and higher financial stress than those with monthly income guarantees
Extrait
“"The 2023 Retirement Confidence Survey found that retirees with guaranteed income sources — including annuities and defined-benefit pensions — reported significantly higher retirement confidence and lower financial stress than those relying primarily on defined-contribution account drawdowns. Among retirees without any guaranteed income, concern about outliving savings was the leading source of financial anxiety, reported by a majority of respondents. Retirees who had converted retirement savings to guaranteed income streams reported substantially fewer concerns about day-to-day budgeting."
”
Données source de
2023-04-01
Consulté le
2026-05-13
Calcul
EBRI 2023 Retirement Confidence Survey — annual nationally representative survey of US workers and retirees. Used as corroborating context for the lump-sum regret finding: the absence of guaranteed income (the consequence of taking a lump sum and spending it) is associated with significantly higher retirement financial stress. Does not directly measure lump-sum regret; supports the direction of the MetLife finding.
Sources : inaction
Registre des sources
Chaque chiffre ci-dessous correspond à ce que la source a rapporté, avec la citation textuelle sur laquelle nous nous sommes appuyés et la méthode de calcul. Cliquez sur un lien pour vérifier directement.
1/2 sources vérifiées de manière indépendante, mot pour mot, par rapport à la source citée
[1]National Bureau of Economic Research / Journal of Risk and Insurance — Financial Regret at Older Ages and Longevity Awareness
Vérifié
↗ 1 other entry
Revue par les pairs
In a Health and Retirement Study experiment (N=1,764, age 50+), regret about retirement finances ran toward NOT securing lifetime income: informing people of objective survival odds raised regret about not buying lifetime income by 42%; 26% regretted not annuitizing while only 9% regretted depending on others — there is no comparable survey of annuity holders regretting having annuitized
Extrait
“"Many older people express regret about undersaving; here, we extend prior work by reporting regret about five other critical financial topics. Using the Health and Retirement Study, we first show that older people who regret past financial decisions differ significantly from those who do not. Second, in an experiment, we demonstrate that informing people about objective survival probabilities increases regret about not buying lifetime income by 42% overall, and by more among the high income or those in good health."
”
Données source de
2025-01-01
Consulté le
2026-06-30
Vérification
Extrait récupéré et confirmé de manière indépendante, mot pour mot, par rapport à la source citée lors de notre audit de vérification.
Calcul
Hurwitz & Mitchell (NBER WP 30696, 2022; published Journal of Risk and Insurance, 2025), a randomised experiment embedded in the Health and Retirement Study, N=1,764 US respondents age 50+. The paper's per-topic regret tallies run the opposite direction from an "annuity regret" rate: roughly 52% regretted undersaving, ~26% regretted NOT annuitizing, 19% regretted claiming Social Security early, and only ~9% regretted depending on others. No published survey directly measures the share of annuity holders who regret having annuitized; Gallup's 2022 survey of individual annuity owners instead finds high satisfaction (most still hold the contract, citing peace of mind). The inaction-side rate is therefore a PROXY (proxy_only: true): it is set low (0.09) to reflect that documented direct annuity-holder regret is small and runs counter to the lump-sum side, not a measured "X% of annuitants regret it" figure. Replaces a fabricated EBRI 27%/73% liquidity-regret statistic that does not appear in the 2023 Retirement Confidence Survey.
[2]Employee Benefit Research Institute (EBRI) — 2023 Retirement Confidence Survey Short Report
Étude primaire
EBRI's Retirement Confidence Survey finds retirees with guaranteed income (annuities, defined-benefit pensions) report higher retirement confidence and lower financial stress than those relying on account drawdowns — i.e. the annuity choice tends to reduce, not generate, the headline regret vector
Extrait
“"Retirement confidence is closely tied to expectations of guaranteed income. Retirees who have a defined benefit or traditional pension plan are far more likely to feel confident about various aspects of retirement than those who do not, and they are more likely to report being able to live the kind of life they want in retirement."
”
Données source de
2023-04-01
Consulté le
2026-06-30
Calcul
EBRI 2023 Retirement Confidence Survey (annual, nationally representative survey of US workers and retirees). Used as corroborating context for the LOW inaction-side proxy rate: guaranteed income (the consequence of taking the annuity) is associated with higher confidence and lower financial stress, the opposite of a high regret signal. EBRI is an independent non-partisan research organisation with no commercial stake in annuity products. Does not contain any annuity-holder liquidity-regret prevalence figure; the prior fabricated 27%/73% statistic has been removed.
Réserves
Les deux versants ne sont PAS mesurés avec un instrument commun, et l'asymétrie constitue le constat central honnête. Le versant action (capital en une fois) repose sur l'étude Paycheck or Pot of Gold Study de MetLife de 2017 (Harris Poll), rapportée ici via PLANADVISER : 31% des bénéficiaires d'un capital versé en une fois ayant réalisé des dépenses importantes le regrettent après coup. MetLife vend des rentes, si bien que l'enquête peut sur-représenter les inconvénients du capital versé en une fois ; les 31% correspondent à la proportion parmi ceux qui ont fait des achats importants, non à l'ensemble de la distribution des bénéficiaires du capital. Le versant inaction (rente) ne fait l'objet d'AUCUNE enquête directe sur le regret : il n'existe aucun chiffre publié du type « X% des détenteurs de rente regrettent de l'avoir choisie ». Les données évaluées par les pairs disponibles (Hurwitz & Mitchell 2025, Health and Retirement Study, N=1 764) vont en sens inverse — environ 26% regrettent de NE PAS avoir opté pour une rente et seulement ~9% regrettent de dépendre des autres — et l'enquête de Gallup de 2022 auprès des détenteurs révèle une grande satisfaction. Le taux d'inaction (0,09) est donc un INDICATEUR INDIRECT, fixé bas pour refléter que le regret documenté chez les détenteurs de rente est faible, et non un taux mesuré ; proxy_only est fixé à true. Les inconvénients structurels de la rente sont réels mais non mesurés par enquête de regret : le seuil de rentabilité dépend de l'âge, du taux d'actualisation et de la longévité — une personne décédant dans les ~10 à 12 ans suivant sa retraite aurait été financièrement mieux lotie avec le capital en une fois — et les rentes nominales fixes perdent en pouvoir d'achat en période de forte inflation (comme observé après 2021), tandis qu'un capital investi peut être protégé contre l'inflation. Les pensions indexées/ajustées au coût de la vie corrigent partiellement ce point mais sont moins fréquentes dans les régimes privés. Il convient de lire l'écart affiché comme directionnel, et non précis, compte tenu de données discordantes et en partie fondées sur des indicateurs indirects.
