Why Financial Plan Longevity Assumptions Need to Go Past 95: A Case Study

Walt and Bev came in for their routine fall annual review. Both early 60s, married forty-plus years, adult kids independent. Walt is three years from retiring out of a long single-employer career. Bev left the workforce earlier, by choice, which means her Social Security benefit is much smaller than Walt’s. About $1.4 million investable — his managed IRA, his principal-protected annuities with lifetime-income riders, a joint taxable account, a small Roth on her side, and a house that’s nearly paid off. Roughly 85% of the investable money is titled to Walt. Their plan on file was built four years ago, ran to age 95 for both of them (software default), came back on-track, and nobody had reopened the horizon since.

They didn’t bring a question about longevity. Walt’s version, compressed: we were on track two years ago, are we still on track? Bev said one thing, offhand, that turned out to be the whole conversation. Ninety-five feels like plenty. I’m not sure I want to be around that long anyway.

The framing move came first.

Before any projection re-run, I named the thing the age-95 horizon had quietly been deciding for them. Ninety-five isn’t when you die. It’s when the software stops looking. For a couple, the planning horizon that matters isn’t the average of the two of you — it’s the survivor’s. The plan’s last decade belongs to the survivor, alone. Bev’s “I’m not sure I want to be around that long” wasn’t resistance. It was fear speaking as indifference, because she had never seen a version of those years that looked safe. The question we actually re-opened: if Walt dies first, what does Bev’s life run on, for how long, and who’s holding the pen if she can’t?

The re-run itself was simple — same plan, extended to age 100 for both, plus a survivor case with first death in Walt’s late 70s and Bev living to 100. The projection outputs aren’t the story. What the re-run exposed was structural, and it shows up on one page: the household’s survivor income map.

Income stream	Whose life it rides on	Survives first death
Walt’s Social Security (~$34K/yr)	Walt	Kept by Bev as the higher benefit
Bev’s Social Security (~$17K/yr)	Bev	Lost — the smaller check drops
Walt’s pension (~$12K/yr)	Walt, single-life (locked)	Lost entirely
Walt’s annuity rider income (~$24K/yr)	Walt, single-life	Lost entirely
Total guaranteed income	~$87K/yr while both alive	~$34K/yr for Bev as survivor

Guaranteed income falls by more than 60% at the first death. Bev’s spending doesn’t fall anywhere near that far. The longest stretch of the plan sits on the spouse with the least control over it, and the care decision — long-term care quotes were run two years ago and no decision was ever made — most likely lands in her solo years, when there’s one income and no spouse to coordinate the response.

The right shape was four moves.

Step 1 — Move the horizon line, and draw it for the survivor.

Action: reset the plan horizon to age 100 for both spouses, and make the survivor case — first death at a plausible age, survivor to 100 — a standing annual-review item.

Why: a horizon set years ago by software default quietly decides how much risk the plan is willing to ignore. For a couple, the planning horizon is the joint last-survivor horizon, which runs longer than either single life.

Trade-off: the plan looks less comfortable on paper. Some of what felt like “on-track” was being bought with a horizon that was never chosen.

Step 2 — Map whose income dies with whom, before buying anything new.

Action: build the survivor income map (above). For each guaranteed income stream, tag joint, single-life on Walt, or single-life on Bev. For each single-life stream, record whether the joint-or-single choice is still available or already locked. The pension election is locked. The annuity rider contracts decide whether the choice is made at issue or at activation — contract-specific, verified per policy, not assumed. Where the choice is still open, the default consideration is joint-life at activation. Where it’s locked single-life, the gap becomes a sizing question for permanent life insurance on Walt owned to benefit Bev, or for repositioning over time.

Why: the household’s biggest longevity risk wasn’t running out of money. It was the income cliff at the first death, landing on the spouse most likely to live longest. You can’t see that cliff in a joint projection. You see it only on the map.

Trade-off: joint-life payouts are lower than single-life. The household gives up some income while both are alive to protect income after one is gone.

Step 3 — Rebuild the estate plan for the long middle, not just the end. 

Action: treat the six-year-old revocable trust and pour-over wills as written for the wrong timeline. Coordinate with the estate attorney on durable financial powers of attorney with authority broad enough to act on retirement accounts and annuities (activating income, changing elections, filing care claims), with the carriers confirmed to honor them; successor trustee provisions that trigger on incapacity, not only on death, with a clear incapacity standard; funding the trust (the joint taxable account was sitting outside it); beneficiary-designation synchronization across every annuity and IRA, checked against what the trust actually says, with attorney and CPA review on the spouse-versus-trust choice for qualified money; and rebalancing control so Bev isn’t the passenger on 85% of the household’s money.

Why: when the horizon stretches, incapacity moves from a tail risk toward the planning median. An estate plan written to transfer assets at death is silent on the ten or fifteen years when someone may be alive but unable to act. Here, the person most likely to need those documents is the one whose name is on almost nothing.

Trade-off: legal fees and a document refresh the household didn’t budget for. Retitling and funding take paperwork and time. Broad POA powers require real trust in the named agent.

Step 4 — Decide the care question, including who decides. 

Action: move long-term care from “quotes on file” to a decision of record. The household picks one of three paths — insure (traditional or hybrid), partially insure, or deliberately self-fund from a named, earmarked pool. Whichever path, the plan names who makes the care decisions and which money pays if Bev needs care after Walt is gone. That ties straight back to the Step 3 documents.

Why: a longer horizon raises the odds a care event lands, and in this household it most likely lands on the survivor alone. The undecided quote was the most expensive choice on the table, because it was a default nobody had chosen.

Trade-off: insurance means premiums, health-dependent underwriting, and rate risk — life insurance and long-term care are medically underwritten, annuities are not. Self-funding ties up capital and asks the household to leave it alone. Either way, the household has to say out loud a scenario it would rather not picture.

Two takeaways for our work.

The first is that the horizon is a plan input, so review it like one. 

Every plan on your desk has an age where it stops looking, and most of us never chose it. In our own fall reviews this year, most plans either had no confirmed horizon on file or had one set years earlier at age 90 or 95. In more than half of the married households in that review set, at least one major income stream was single-life on one spouse with no survivor analysis on file. The peer-professional move worth carrying is writing the horizon, the date it was set, and the survivor case into every annual review — not as a projection flourish, but as a line item.

The second is for the attorneys reading this. Longevity turns incapacity into a central planning case.

The estate plan most clients hold was drafted to move assets at death. A longer life means a longer, likelier stretch where someone is alive and can’t act. POA scope, incapacity-triggered trusteeship, trust funding, and beneficiary synchronization are the documents that run that stretch. They have to be coordinated with how the income is actually owned, or the survivor inherits a plan she can’t operate.

We get to be here, possibly longer than anyone planned, and the work is making sure the person left at the table is still protected.

Hypothetical case study based on patterns from real planning engagements; identifying details have been changed. The plan discussed and the moves described are considerations raised in annual review conversations, not steps reported as implemented. Income figures shown are composite illustrations of the survivor-income inventory and are not performance projections. Pension single-life/joint-life election rules, annuity rider joint/single election timing, powers-of-attorney scope and carrier acceptance, trust-as-beneficiary treatment of qualified money (including SECURE Act distribution rules), and long-term-care underwriting all vary by plan, carrier, and jurisdiction and should be reviewed with the client’s own attorney and CPA. Life insurance and long-term-care coverage are medically underwritten; annuities are not. Past performance is not indicative of future results.