The Family Endowment: Planning for a Longer Life

By Panoramic Capital Partners

The Family Endowment: Planning for a Longer Life

The Full Picture | Panoramic Capital Partners

---

On January 31, 1940, a retired legal secretary from Ludlow, Vermont named Ida May Fuller received the first monthly Social Security check ever issued. Check number 00-000-001, in the amount of $22.54. She was 65 years old, and she had paid a grand total of $24.75 into the system during the three years it had existed.

The actuaries who built the program expected to pay people like her for a little over a decade. That was the honest arithmetic of the era: a man who reached 65 in 1940 could expect roughly thirteen more years.

Ida May Fuller lived to 100. She collected benefits for 35 years, cashing checks well into the Ford administration, and received nearly a thousand times what she had paid in.

Her story usually gets told as charming trivia from nearly 90 years ago; however, it reads better as a warning that many families are all too familiar with today. The very first retirement check in American history went to someone who outlived the system’s assumptions by more than two decades. Nearly everything that has strained retirement planning since is that same surprise, repeated at national scale.

---

What you should take away from this article

• America’s retirement income system was built when the payout window was short and the worker base was deep. Both assumptions have quietly eroded, and the system was never rebuilt to reflect it.
• Retirement ages have barely moved in ninety years while time spent in retirement has roughly doubled. Anchoring to “65” is anchoring to 1935 arithmetic.
• The “4% rule” developed within the financial planning industry to start addressing longer term planning answered a specific question: how much can someone withdraw over 25 to 30 years, based on historical U.S. returns? However, as lifespans continue to stretch, changes to the time horizon or the market regime mean the answer needs to change with it.
• Monte Carlo “probability of success” scores (broadly used in the financial planning industry) can obscure what actually funds your spending each year. Withdrawal rate and income rate are critical inputs to the scores and can provide more clarity.
• A forty-year retirement behaves less like a countdown and more like an university endowment designed to perpetually support the institution. That shift changes how families should set spending, build portfolios, and size the assets they will need.

---

A system built for thirteen years

Retirement, as a system in America, was invented by the railroads. The first formal industrial pension plan in North America appeared in the rail industry in 1874, and American Express, then a railroad freight agency, established the first private-sector pension in the United States a year later. By 1925, more than three-quarters of American railroad workers were covered by a pension plan.

Coverage, it turned out, was not the same as security. The plans were unregulated, frequently terminated, and chronically underfunded, and the Great Depression broke most of them. Older workers who no longer trusted their pensions used their seniority to keep working, which clogged the industry with employees in their seventies and pushed Congress to act. The Railroad Retirement Acts of 1934, 1935, and 1937 created the first federal retirement system, with annuities payable at 65. The Social Security Act of 1935 extended the same basic architecture, and the same age, to most other workers.

A popular myth says the designers chose 65 because most people would die before collecting. That’s not quite right, and the truth is more useful. Life expectancy at birth in the 1930s was about 58 for men and 62 for women, but those figures were dragged down by childhood mortality. A man who actually reached 65 in 1940 could expect about thirteen more years. The system was honest for its time. It simply assumed two things: a modest payout window, and a deep bench of workers behind every retiree. In 1935 there were roughly ten working-age Americans for every person over 65.

Both assumptions have eroded. Today there are about 2.7 workers per beneficiary, projected to fall toward 2.2 by 2045. The 2026 Trustees Report projects that Social Security’s retirement trust fund will be depleted in late 2032, at which point incoming payroll taxes would cover roughly 78% of scheduled benefits unless Congress acts. The program isn’t disappearing, fixes are actually readily available given the right political willpower to start the conversation in earnest, and history suggests some fix will come. The structural point stands: a system designed around a thirteen-year liability now carries a twenty-plus-year one, and the math has never been fully rebuilt to match.

More importantly, the solvency debate misses a key point: even a full repair of Social Security funding wouldn’t solve the family-level problem, because the check already covers a shrinking share of what retirement actually costs. Social Security’s own researchers note that net replacement rates are falling under current law, driven by the rise in the full retirement age from 65 to 67, the growing taxation of benefits, and Medicare premiums that come straight out of the check and grow faster than the check does. The 2026 numbers made the pattern visible: the cost-of-living adjustment raised the average benefit 2.8% while the Medicare Part B premium rose 9.7%, reclaiming roughly a third of the increase before it arrived. For higher earners the arithmetic is starker still, by design. The benefit formula replaces a much smaller fraction of income above average wages and nothing above the taxable maximum, which means many families will find Social Security covering a small and shrinking share of the retirement they actually intend to fund. Solvency determines whether the check arrives in full, but it says nothing about how much of a forty-year retirement the check can carry.

---

The pensions made the same mistake

Corporate America’s version of this story ran in parallel, and ended the same way.

Defined benefit pensions promised lifetime income, and their sponsors made the same actuarial bet the railroads had made: lifetimes would stay short enough, and funding could stay thin enough, for the promise to hold. When Studebaker collapsed in 1963, thousands of workers received a fraction of their promised benefits or nothing at all. Congress responded with ERISA in 1974, which protected workers by making pension promises expensive to keep. Four years later, the Revenue Act of 1978 quietly created section 401(k), and by 1981 the first plan was live.

The migration was swift. In 1980, about 38% of private-sector workers had a defined benefit pension. Today roughly 15% have access to one. Employers weren’t simply changing plan types; they were transferring longevity risk and responsibility to the employees. Under a pension, the institution bears the consequences of retirees living longer than expected. Under a 401(k), the household does.

Here is the part worth sitting with: the institutions that mispriced longevity could freeze their plans and move on. Governments can adjust formulas. Corporations can close plans to new entrants. A family can’t freeze anything. Of the three parties that have taken a turn holding longevity risk over the last 150 years, the one holding it now is the only one with no exit ramp and no younger generation of contributors to spread it across.

More of the retirement problem belongs to individual families today than at any point since retirement was invented.

---

Retirement age is the elephant in the room

Germany created the world’s first state pension system in 1889, with benefits at age 70. The United States settled on 65 in the 1930s. Ninety years later, the full Social Security retirement age has moved to 67. Two years of adjustment, in nearly a century.

Meanwhile, the thing the number points to has transformed. A man reaching 65 in 1940 averaged about thirteen more years. By 1950 the figure was 13 for men and 16 for women; by 2000, 17 and 20; Social Security’s actuaries project roughly 21 and 23 by 2050. For couples the math compounds: for a healthy 65-year-old couple, the odds are roughly a coin flip that at least one spouse is still alive around age 90, and about one in five that one of them reaches 95. Demographers writing in The Lancet (a scientific journal) projected that, if current trajectories hold, more than half of babies born in industrialized nations since 2000 will live to 100. Other researchers think that’s optimistic. Either way, the direction of travel isn’t in dispute.

Instead of adapting as longevity has continued to improve, most people still target a retirement age somewhere between 62 and 65. This is where anchoring does its quiet damage. They picked that number largely because their parents did. Their parents picked it because Franklin Roosevelt’s actuaries did. The number has stayed frozen while the product it describes has changed entirely. Retiring at 65 in 1940 meant funding perhaps thirteen years. Retiring at 62 today can mean funding thirty-five or forty.

---

Separating what you control from what you don’t

1. Spending. Both the level and, just as important, the flexibility. A spending plan that can flex down 10% in a bad year does more for plan survival than most portfolio decisions.
2. Saving. The rate at which capital gets built pre-retirement. For business owners specifically, the discipline of moving wealth out of the business along the way rather than betting everything on a single future transaction.
3. Asset allocation. How the portfolio balances growth against income/stability, and how aggressively it’s positioned relative to the plan’s actual needs.
4. Working years. When and how work ends. This is the most powerful lever in the entire plan and the least discussed. Each additional year of even partial income delays withdrawals, adds compounding time, and shortens the funded horizon on both ends. Given forty-year horizons, the traditional retirement age deserves honest reexamination, and “phased retirement” deserves a seat at the table alongside “retired.”
5. Withdrawal architecture. Which accounts fund spending, in what order, and how much of each year’s check comes from portfolio income versus asset sales.

1. Market returns, and more dangerously, the order in which they arrive.
2. Inflation, which compounds silently across decades.
3. Policy, including tax law and the benefit formulas behind that 2032 date.
4. Longevity itself. How long you and your spouse actually live is the one variable the whole plan exists to absorb, and the one nobody gets to choose.

---

What is the the 4% rule and where does it fall short?

---

What traditional financial planning (via Monte Carlo alone) doesn’t show you

Most modern financial plans get pressure-tested with Monte Carlo simulation: thousands of randomized return sequences, distilled into a single probability of success. The tool is genuinely valuable for what it was built to reveal, which is sequence-of-return risk, the danger of bad markets arriving early in retirement. To that end, we are fans of the tool and use it broadly when working with clients as one piece of the puzzle in understanding long-term financial health.

The problem is what the single score conceals. Consider two households whose plans fund spending very differently. One covers most of its annual spending from portfolio income: dividends, interest, rental cash flow, fund distributions. The other sells assets every quarter to raise cash. A keen observer knows these aren’t equivalent; all else equal, the sale-driven plan carries more risk, since forced selling in down markets is exactly what erodes a retirement. Here’s the catch: the standard simulation prices the portfolio’s total return, not the plumbing of the withdrawal, so the two plans can still post similar headline scores. The risk the keen observer sees never surfaces in the number. When markets fall, the first household keeps cashing roughly the same checks. The second sells more shares at worse prices, which is precisely the mechanism that destroys retirements, and precisely the moment when people abandon otherwise sound plans.

A probability score also says nothing about the texture of the journey: which years get tight, which account gets drawn when, what happens to the income stream if rates or payout policies shift. The simulation optimizes for terminal outcomes. Families live through cash flows.

There’s a second gap the score can’t see: the distance between the plan and its implementation. Every plan is a stack of assumptions about spending, inflation, returns, and where each year’s withdrawal will come from. Assumptions drift. Spending creeps, markets reshuffle the mix, and a withdrawal that was supposed to come from income quietly starts coming from principal. Few advisors ever reconcile actual withdrawals against what the plan assumed, so the document gets built once, admired, and shelved while reality wanders off without it. A plan is only as good as the paper it’s written on. Measurement against that plan, on a schedule, is what turns it into a system.

The remedy isn’t to throw out a helpful tool. It’s to see the tool for what it is: a necessary input to the planning process, but not a sufficient one solely in itself. Probability of success is a statistical measure, and a statistic can’t fully capture the quality good planning actually has to test, which is how resilient or fragile the plan is when reality misbehaves. So how do we add to the toolset in order to triangulate around a better planning solution? Unfashionably old-fashioned ideas: real cash flow planning, tied directly to asset allocation. Every plan should be able to answer two questions in plain numbers. What is the withdrawal rate, meaning total annual spending as a share of the portfolio? What is the income rate, meaning the share of that spending covered by durable portfolio income? The gap between the two is the amount that must come from selling assets, and the size and timing of that gap should drive the allocation, not the other way around. When principal sales are planned rather than forced, sequence risk can be managed instead of just modeled.

---

Shifting the family mindset to planning like an endowment

1. From a withdrawal number to a spending policy. Replace “4% forever” with written rules: how spending behaves in strong years, how it flexes in weak ones, what floor protects the non-negotiables. Flexibility is worth real money; Morningstar’s research finds flexible systems can support meaningfully higher lifetime spending than rigid ones.
2. From total return only to income visibility. Know your income rate, not just your return. Build a floor of durable income (Social Security claimed well, dividends, interest, real estate cash flow, and distributions) that covers essential spending, so markets are negotiating with your wants rather than your needs.
3. From age-based conservatism to growth sized for four decades of inflation. At 3% inflation, prices roughly triple over forty years. A portfolio positioned only for stability quietly fails at its actual job, which is purchasing power. This introduces tension that needs to be managed: the growth assets that defend purchasing power typically produce less current income.
4. From a hoped-for number to an honestly sized corpus. That tension between income today and growth for tomorrow resolves only one way: with more capital. The portfolio needs enough income-productive assets to cover the floor and enough growth assets to defend the next three decades, and lower sustainable withdrawal rates over longer horizons mean the required multiple of spending is simply larger than the old rule implied. Better to size that honestly at 40 than to discover it at 75.
5. From countdown to stewardship. Endowments treat surplus in strong years as future income, not found money. Families running forty-year plans become, in effect, their own pension committee: a written policy, an annual review, and a spouse and next generation who understand how the machine works.

---

The owner’s version of this problem

---

The owner’s version of this problem