Longevity Risk in Pension Plans: What It Is and How to Manage It

By Jonas Osman Abdelghafour — Actuary & Risk Expert

People living longer is one of the great successes of the modern era. For a defined benefit pension plan, it's also a liability. Longevity risk — the risk that members live longer than the plan assumed when it set aside money for their pensions — is one of the least visible but most persistent threats to pension funding. Unlike a market crash, it never announces itself. It shows up slowly, one valuation at a time, as actuaries update mortality assumptions and liabilities drift upward. What follows is a plain-English tour of where longevity risk comes from, why it's harder to manage than investment risk, and the main tools plans use to deal with it.

A pension promise is a stream of payments that lasts as long as the member (and often a surviving spouse) lives. To put a value on that promise today, an actuary needs two big ingredients: a discount rate and a mortality assumption. The mortality assumption answers the question, "How long will these payments run?" That assumption has two layers, and both can be wrong. Baseline mortality describes how likely members are to die at each age today. Standard mortality tables give a starting point, but every plan's membership is different — higher earners tend to live longer, certain industries carry heavier mortality, regional differences matter. If a plan uses an off-the-shelf table without adjusting for its own population, its baseline can be off from day one. Mortality improvement describes how death rates will change in the future, and this is the harder problem. Over the twentieth century, mortality at older ages improved decade after decade, but not smoothly and not predictably. Improvements driven by cardiovascular medicine slowed; improvements from other sources emerged; pandemic years disrupted the trend entirely. Projecting improvement forty years forward is genuinely uncertain, and small changes compound. Adding a single year of life expectancy at retirement typically increases pension liabilities by several percent — a large number when applied to a multi-billion liability. Longevity risk, then, isn't one risk but two: the risk of mis-estimating today's mortality, and the risk that the future trend surprises everyone.

Trustees and sponsors are used to thinking about investment risk. Longevity risk behaves differently in three important ways. First, it's a one-way street in practice. Investment markets go up and down; longevity assumptions, over long horizons, have mostly gone one direction — toward longer lives and higher liabilities. Periods where assumptions are weakened do occur, but the structural bet a pension plan holds is against continued improvement. Second, it's slow and systemic. A plan doesn't discover longevity losses in a quarterly statement. It discovers them at triennial valuations, when the actuary updates tables and the funding position quietly deteriorates. Because the same medical and social forces act on everyone, longevity shocks hit all plans at once — there is no diversification across members of the kind that protects an insurer writing many independent risks. Individual lifetimes diversify; the trend does not. Third, it interacts with other assumptions. Longer lives mean payments further into the future, which makes liabilities more sensitive to discount rates and inflation. A plan that has hedged its interest rate exposure based on current mortality is under-hedged if lifespans extend.

The starting point for managing the exposure is an experience study: comparing the plan's actual deaths against the assumed table over several years. For large plans this supports a credible plan-specific adjustment; for smaller plans, actuaries blend plan experience with industry or postcode-based models, giving weight to each in proportion to its credibility. For the trend, most actuaries use projection models that extrapolate historical improvement with expert-judgment overlays, typically expressed as an assumed long-term improvement rate. The honest position is that no model predicts medical breakthroughs or societal shifts; the model's job is to make the assumption explicit and testable rather than to be right. Stress testing translates the uncertainty into money. A useful, plain-English metric for a trustee board: "If life expectancy at 65 turns out one year higher than assumed, our deficit grows by X." That single sentence often communicates more than a full stochastic analysis.

Once a plan understands its exposure, it has a spectrum of options, roughly in order of increasing completeness and cost. The simplest response is to fund against a prudent mortality assumption. This doesn't transfer the risk, but it builds a cushion and avoids the pattern of repeated bad news at each valuation. Beyond that, a longevity swap lets the plan pay a fixed series of payments based on expected mortality and receive payments based on actual mortality — if members live longer than expected, the counterparty covers the difference. The plan keeps its assets and investment strategy; only the longevity risk moves. Swaps suit large plans that want to retain investment control, though they involve long-dated counterparty relationships and collateral arrangements that need careful governance. A buy-in is the next step up: the plan buys a bulk annuity from an insurer covering some or all members, and the policy is held as a plan asset with payments that match the covered benefits, removing longevity, investment, and inflation risk on that block. Members' benefits and the trustee relationship are unchanged. A buy-out is the full transfer — the insurer takes over the obligation entirely and members become policyholders of the insurer. This is the endgame for many closed defined benefit plans, and it's the most expensive option because the insurer prices in its own prudence, capital costs, and margin. The right choice depends on plan maturity, funding level, sponsor strength, and pricing conditions in the bulk annuity market, which move with insurer capacity and reinsurance appetite. Timing and preparation matter: clean member data and codified benefits routinely improve pricing more than clever negotiation does.

Longevity risk is the risk that runs quietly in the background of every defined benefit promise. It cannot be diversified away, it has historically moved in one direction, and it compounds with interest rate and inflation exposure. Managing it starts with measurement — an honest experience study and an explicit improvement assumption — and proceeds through a well-established toolkit, from prudent funding through longevity swaps to full buy-out. Plans that treat longevity as a risk to be measured and priced, rather than an assumption to be updated reluctantly every three years, consistently make better de-risking decisions.

---

Jonas Osman Abdelghafour is an actuary and risk expert who helps pension plans, insurers, and corporates quantify and manage financial risk — from mortality assumption reviews to de-risking strategy. If your board needs an independent view on longevity exposure or a transaction-readiness assessment, reach out at contact@jonasosman.org.