The 85 to 89 Cliff.
The single greatest financial risk for a woman entering retirement is not a heart attack. It is outliving her capital in a drawdown model calibrated on male death patterns.
The arithmetic mismatch
Pension-fund actuarial tables, defined-benefit reserve models, and defined-contribution decumulation products are overwhelmingly indexed to a stylised lifecycle curve. In that curve, income peaks in the 50s, tapers through the 60s, and principal is drawn down through a low-utilisation 70s phase that ends in an acute terminal event. Most wealth-management drawdown software assumes the principal is substantially exhausted by the mid-80s.
This curve is a male curve. It describes men's spending and mortality patterns with reasonable accuracy. It describes women's lifecycle curve poorly and in some respects catastrophically.
The female curve does not terminate in an acute event in the early 70s. Women outlive men by roughly 5 to 7 years in OECD countries. They spend approximately 25 percent more of those years in poor health (WHO; McKinsey Health Institute). And the dominant female morbidity pattern is chronic, compounding, and multi-organ, driven by post-menopausal cardiovascular, osteoporotic, cognitive, and autoimmune decline. Peak healthcare utilisation therefore defers by approximately a decade, concentrating at ages 85 to 89. The pension architecture meets the woman at precisely the moment it has assumed she is gone.
The two curves, drawn
The illustration below is stylised (it aggregates several OECD per-capita spend curves into two indicative trajectories) but the shape is correct and it is the shape that matters. The male curve peaks early and falls sharply. The female curve defers and then compounds.
Indicative values drawn from OECD Health at a Glance 2023 and WHO Global Health Observatory. Country-level figures vary; the directional gap (longer life + longer morbid tail for women) is consistent across OECD members. McKinsey Health Institute 2024 estimates the healthspan deficit at roughly 25 percent more morbid years for women.
The shaded zone in the chart is the pension industry's single largest unpriced liability. It is not hidden. It is the most visible part of the lifecycle, and for the female sub-population it is where the curve is still climbing while the product has already declared itself spent.
Why this keeps happening
Three structural reasons, each tractable.
- Actuarial tables blend. Most retail decumulation software runs on unisex or weighted-blend mortality tables because regulatory frameworks in some jurisdictions require unisex pricing. The blend suppresses the female late-life tail because the male curve terminates earlier and carries equal weight in the average.
- Morbidity is modelled as a binary. Standard plans ask whether the insured is alive or dead. The 25-percent-more-of-life-in-poor-health finding only matters if the plan models utilisation state, not just survival state. Few retail products do.
- Caregiver transitions are invisible in the balance-sheet. For the middle-class female retiree, the single largest late-life cost shock is the transition from informal family caregiver to paid care. Current decumulation products do not surface this transition as a priced event. See the Sandwich Generation essay for the $115 billion national-scale estimate of what the informal-caregiver economy is absorbing, and who loses access to it when they become the patient.
What a female-calibrated drawdown product looks like
The proposed product architecture is not speculative. It assembles existing building blocks that the industry already sells separately.
- A stepped-care annuity base. Standard lifetime income annuity. Reserves priced on a sex-specific mortality table, not a blend. Benefit level steps up at three pre-defined inflection points: documented menopause onset (year 0), first osteoporotic or cardiovascular risk flag (approximately year 15 to 20), and first cognitive-decline marker (variable). The step schedule is priced transparently at issue.
- A caregiver-transition rider. When informal family care becomes insufficient (a documentable trigger: admission for a fall-related injury, hospital discharge without a home-health plan, or formal dementia staging) the rider routes a defined portion of the annuity into paid caregiver infrastructure. This converts the hidden tail risk into a scheduled, bounded cost.
- A cardiovascular surveillance premium rebate. Women enrolled in the surveillance pathway described in the preeclampsia CVD essay present a measurably different claim-frequency profile. The product prices the rebate, not the surcharge.
- A post-menopausal longevity rider. Optional, underwritten, expressly priced as an extension of principal into the 85-to-99 utilisation window. Positioned not as catastrophic-care insurance but as the plausible-case extension every female retiree actually faces.
Each component uses existing reserve mathematics. What is new is the recognition that the late-life window is the central product, not an edge case.
Who this is for
Pension-fund trustees and sponsor CFOs carrying defined-benefit obligations that are disproportionately female in their payout distribution (public-sector plans, teacher and nurse pensions, healthcare and education trust funds) are carrying the largest version of the mismatch described here. The reserving implications are material and under-disclosed.
Wealth managers and retirement-product designers selling into the mass-affluent female market face a product line whose most significant consumer, the retired 70-year-old woman who is statistically the sole financial decision-maker of her household for the next 15 to 20 years, is not the customer the current drawdown model was built for.
Insurers writing life and long-term-care hybrid products have the clearest path to act. The architecture above is a product brief, not a research agenda.
The financial industry has spent the better part of a decade instrumenting wealth management around longevity as a single aggregated variable. Longevity is a female variable. The cliff lives, and it is not invisible. It simply has not yet been named as the product people are actually retiring into.
Related reading: The Unpriced Liability in Portfolios for the wealth-management sector framing; The Lifetime Gender Health Tax for the cumulative consumer burden; Pricing the Pathway for the reinsurance-treaty mechanics.