A father explains to his daughter that in his day, the company took care of you, and hears himself sounding like a man describing a country that no longer issues visas. He is not wrong about the past. He is wrong about what made it work. The pension he collected was not a reward for a superior work ethic; it was the visible end of a bargain whose invisible half was demographic and institutional: many young workers per retiree, and firms expected to outlive everyone on the payroll.
Both halves are gone. Across the OECD, the old-age dependency ratio, people over 65 relative to the working-age population, rose from 19 percent in 1980 to 31 percent in 2023, and is projected to reach 52 percent by 2060. A promise structured as “the young will fund the old, and the firm will be here in forty years” was rational under the first number and is arithmetic fantasy under the last. The children are not imprudent for declining to believe it. They are reading the same table.
What the pension actually was
Strip the nostalgia and the defined-benefit pension was three things at once: a deferred claim on the firm's future, an instrument that made staying rational, and a signal that the firm expected to exist. It is telling that where genuine long-term commitment devices survive, they cluster together. Blasi, Kruse and Weltmann, tracking privately held ESOP companies against matched non-ESOP firms from 1988 to 1999, found the employee-owned firms were only half as likely to go bankrupt or close, and four times more likely to also offer a defined-benefit pension. Security was never a free-standing benefit; it was the surface of a deeper design in which workers held real claims and the firm planned in decades.
So the honest translation for a younger listener is not “we were loyal and you are not.” It is: “we held instruments that made loyalty pay, and those instruments were discontinued.” Loyalty without a claim is just unpriced risk carried by the employee.
Portability is the same need, re-engineered
What replaces the pension is not one product but a property: portability. Savings that move between employers. Skills and reputations that are legible outside one firm. And, crucially, information. A worker who can see the numbers can price their own deal; a worker who cannot is trusting blind, which is exactly what the old system asked and the new one cannot honestly ask.
The research on pay information is nuanced in ways both generations should hear. Cullen's review in the Journal of Economic Perspectives distinguishes three kinds of pay transparency with opposite effects: horizontal (peer-to-peer) transparency tends to reduce effort and satisfaction among those who discover they are behind; vertical transparency, knowing what the boss makes, can increase effort by revealing the career return; and cross-firm transparency laws tend to compress and lower pay by weakening individual bargaining. Transparency is not virtue in the abstract. It is a design choice, and it can be designed badly.
The Semco version of a portable claim
One company ran this translation decades early. Semco paid a fixed slice of each unit's profit, 23 percent, to the people who produced it, with the books open so anyone could check the math. As Ricardo Semler has told it, the side effect was people turning off lights, because the money had become theirs; his shorthand was that nobody hoards a pizza everyone shares. He liked to test other companies with one interview question: what percentage of last year's profit went to non-executives?
That is the interesting middle ground between the pension era and the gig era. The claim was immediate rather than deferred forty years, so it did not require believing in the firm's immortality. But it was collective and verifiable, so it built exactly the thing the pension used to signal: evidence that the company expected to share its future, not merely rent your present.
Saying it without nostalgia
A senior leader explaining pay to an adult child has one real gift to give, and it is not the story of the gold watch. It is fluency in the underlying deal. Every compensation system answers three questions: who carries the risk, who sees the numbers, and what happens to your claim when you leave. The pension answered them one way for one demographic moment. The portable world answers them differently, not worse.
The failure mode is mourning the old answers. The useful act is teaching the questions, then admitting, without bitterness, that a system in which the worker checks the books beats a system in which the worker trusts the founder's memory of a kinder time.
Sources
- OECD (2025). OECD Employment Outlook 2025. Old-age dependency ratio in the OECD rose from 19% (1980) to 31% (2023), projected 52% by 2060, making old pension arithmetic structurally obsolete. https://www.oecd.org/en/publications/oecd-employment-outlook-2025_194a947b-en.html
- Blasi, J., Kruse, D. & Weltmann, D. (2013). Firm survival and performance in privately held ESOP companies. Matched longitudinal design 1988-1999: ESOP firms half as likely to fail and four times more likely to offer a defined-benefit pension. https://www.researchwithrutgers.com/en/publications/firm-survival-and-performance-in-privately-held-esop-companies
- Cullen, Z., Journal of Economic Perspectives (2024). Is Pay Transparency Good? Horizontal, vertical and cross-firm transparency have opposite effects; cross-firm laws tend to compress and reduce pay. https://doi.org/10.1257/jep.38.1.153