Cohort vs. Conveyor-Belt Financial Planning
Median-by-age data is a snapshot of one era's rules, not a forecast of yours.
When personal-finance content shows 'median savings by age 20, 30, 40, 50, 60', it presents a conveyor belt — implying a 25-year-old can expect to land where today's 60-year-olds are. Nicholas points out this is a category error: those are different cohorts who lived under different rules (DB pensions, mortgage interest tax relief, cheaper houses, 15% interest rates that crushed and rewarded different choices).
A 25-year-old today cannot use the median 60-year-old's balance sheet as a forecast — it's a snapshot of historical conditions, not a trajectory. The framework is to plan from your own cohort's actual ruleset (no DB pension, state pension uncertain, housing detached from wages, longer retirements) rather than extrapolating from the people currently retiring.
The correction is to (1) build your plan on your cohort's enablers, (2) assume safety nets erode rather than persist, (3) shift saving and investing earlier and more aggressively to compensate, and (4) treat any 'expected' inheritance or state benefit as upside rather than baseline.
- Cross-sectional data shows where different cohorts ended up, not where you will end up.
- Each cohort's outcome was shaped by enablers (DB pensions, asset prices, interest rates) specific to its era.
- Treat current retiree balance sheets as the output of a system that no longer exists.
- Plan from your cohort's actual enablers, not from inherited median tables.
- Erosion of safety nets is the base case; their persistence is the upside scenario.
- Inventory your cohort's actual enablersList what is genuinely available to you: workplace DC pension match, ISA/Roth allowances, property markets you can access, remote-work wage arbitrage, low-fee global index funds. These are your levers — not your parents'.Pro tipCompare to your parents' enablers at the same age. The deltas tell you which strategies imported from them are obsolete.
- Strip out assumed-but-uncertain inputsBuild a base plan that does not rely on (a) state pension at current real value, (b) inheritance, (c) house price growth above inflation, (d) decade-long bull market average returns. Add each back as upside scenarios.Warning60% of UK retirees rely on state pension for half their income — building a plan that assumes its persistence at current real value is a single point of failure.
- Set your savings rate against your cohort's hill, not theirsIf housing eats 50% of pay (vs ~8% in the 1970s) and you'll retire later with no DB pension, the savings rate that produced your parents' outcome is too low for yours. Increase contributions earlier; let compounding do the work the property ladder did for them.Pro tipA 15% pension contribution rate from age 25 broadly replaces what a DB pension delivered for the prior generation.
- Diversify across uncorrelated regimesYour parents could win by being long one asset (UK housing) for 40 years. You can't safely bet on any single regime continuing. Spread across global equities, fixed income, and a small allocation to alternatives so a regime change in any one doesn't take you out.WarningDon't confuse new access (crypto, app-based investing) with new diversification — concentration in one volatile asset is concentration regardless of the wrapper.
- Re-plan every 5 years against the actual rulesetThe rules will keep changing — pension allowances, tax wrappers, retirement age, state pension formula. Treat the plan as a rolling document, not a one-shot decision.Pro tipTie the review to a fixed life event (every birthday ending in 0 or 5) so it actually happens.
Anyone who bought their first house in 2007 saw values collapse with a fresh mortgage at peak prices in a recession with their job at risk. A 50-year-old in 2008 saw their house dip but still held a 2-3x gain on purchase price and a job longer-tenured.
Boomers retired with employer-managed Defined Benefit pensions where 'work would sort it out.' Younger cohorts have Defined Contribution: more freedom, far less stability, full longevity and market risk transferred to the individual.
Nicholas describes a video he made looking at median savings by age, and how presenting it as a ladder — '20, 30, 40, 50, 60, you should be here by then' — disguises that each row is a different cohort with different rules. The conversation extends this: today's 60-year-olds had DB pensions and bought houses cheap; today's 25-year-olds will not, and projecting forward from the snapshot is mathematically and structurally wrong.