Most people never become financially independent. That sentence is usually sold as a character study. It is a spreadsheet.

Financial independence (FI) means a portfolio that can fund your spending without wages. Under the 4% first-year withdrawal heuristic, the target is 25 × annual spending. William Bengen (1994) and Cooley, Hubbard, and Walz (1998, the “Trinity” study) derived that heuristic from U.S. market history. It is not a contract with the future.

This piece does three things: (1) show the math of savings rate vs return vs time, including a 10-year-earlier start with the future value written out; (2) separate survey evidence from argument on why people miss; (3) rank what actually moves the date, and give one action for this week.

1. What the official numbers show

Evidence — Federal Reserve, Survey of Consumer Finances 2022 (Aladangady et al., 2023; 2022 dollars):

  • 56% of families spent less than their income in the prior year (59% in 2019). Top usual-income decile: 82% saved. Upper-middle: 66%. Bottom half: 43%.
  • Median family net worth: $192,900 (mean $1,063,700). Under 35: median $39,000. Ages 35–44: $135,600.
  • Median family income (calendar 2021): $70,300. Under 35: $60,500. Ages 35–44: $85,900.
  • Retirement accounts (IRA / defined contribution): held by 54.3% of families; conditional median $86,900, conditional mean $334,000. Just over two-thirds of working-age families participated in some retirement plan (IRA, DC, or defined benefit).
  • Congressional Research Service, same survey: 49.6% of households under 35 and 61.5% of those 35–44 had any retirement-account assets. 4.6% of all households had more than $1 million in those accounts.
  • Direct-plus-indirect stock ownership: 58% of families. Bottom half of usual income: 34%. Upper-middle: 78%. Top decile: 95%.
  • Median home value divided by median family income: more than 4.6. Student debt: 22% of families.

Evidence — Bureau of Economic Analysis. Personal saving as a share of disposable personal income: 5.4% in 2024, 4.6% in 2025, 2.7% in June 2026. This is a national-accounts residual. It is not the FIRE savings rate of a 22-to-45-year-old. Do not paste 2.7% into a retirement calculator as “what Americans save.”

Evidence — BLS Consumer Expenditure Surveys, 2024:

  • Average income before taxes: $104,207. Average expenditures: $78,535.
  • Housing: $26,266 (33.4% of spending). Transportation: $13,318 (17.0%). Combined: half the budget.
  • Personal insurance and pensions: $9,797 (12.5%) — already inside expenditures.
  • Spending by income quintile: $35,046 (lowest) to $150,342 (highest). The highest quintile starts at $155,925 of before-tax income.

You cannot treat $104,207 − $78,535 as a household savings rate. Income is pre-tax; some saving is already booked as an expenditure.

Evidence — planning behavior. Fidelity Retirement Mindset Study (2019; n=1,429 adults ages 23–74): 18% had a “comprehensive written plan.” EBRI/Greenwald 2025 Retirement Confidence Survey: 69% of workers (or a spouse) had saved something for retirement; 64% were currently saving; 54% had tried to calculate a number; 52% had estimated monthly retirement income; 44% had thought about a withdrawal amount. 67% of workers were at least somewhat confident they would have enough; 24% were very confident. High confidence, thin paperwork.

2. The math that sets the date

Two identities:

  1. Target ≈ annual spending / withdrawal rate. At 4%, that is 25×. At about 3.3%, about 30×. The second is a planning choice for a longer-than-30-year withdrawal, not a new historical study.
  2. Years to FI (start at $0, constant savings rate (s), constant real return (r), target multiple (M = 25)):

n = ln(1 + M × r × (1 − s) / s) / ln(1 + r)

Self-computed, (r) = 5% real, (M) = 25:

Savings rate (after tax)Years to FI
5%66
10%51
15%43
20%37
25%32
30%28
40%22
50%17
60%12
70%9

A 50% rate is not a moral command. It is why some people hit FI in their 30s or 40s and most do not. A 10% rate is a 51-year clock — a traditional career, not early independence.

Return still matters. It does not substitute for the rate. Raising the assumed real return from 5% to 7% at a 10% savings rate shortens the clock, but you remain in multi-decade territory. Raising the savings rate from 10% to 25% at 5% real cuts the clock from 51 years to 32. The rate is the first lever.

3. Starting 10 years earlier — with the future value

The previous live version of this article said starting at 25 vs 35 “can mean millions.” Here is the future value, not the slogan.

Assumptions (stated so you can break them):

  • End-of-year contribution: $12,000 (no raise; employer match not modeled separately)
  • Nominal return: 7% per year, compounded annually
  • No fees, no taxes on the growth, no withdrawals
  • Why 7%, not 10%: Damodaran’s S&P 500 series (dividends reinvested) takes $100 at the start of 1928 to $1,157,599 at the end of 2025 — a geometric average of about 10.0% nominal. Seven percent is below that history. It is a planning rate, not a forecast.

Future value of an ordinary annuity: FV = C × [((1 + r)^n − 1) / r]

PathYears contributingFV at age 65
Start at 25, contribute to 6540$2,395,621
Start at 35, contribute to 6530$1,133,529
Gap10 extra years$1,262,092

Same $12,000 annual contribution. The extra decade is more than a million dollars in this table. Change C or r and the gap scales; it does not disappear.

A stricter comparison — same dollars, different calendar:

PathDollars contributedFV at 65
Ages 25–35 only, then stop10 × $12,000 = $120,000$1,262,086
Ages 35–65, never stop30 × $12,000 = $360,000$1,133,529

Ten early years, then nothing, finish ahead of thirty late years. That is compounding, not hustle.

If you are 22–35 and contributing nothing that invests, the 10-year gap is still open. If you are 35–45, you cannot buy those years back. You can still buy rate.

4. Why the savings rate beats return in the years that matter

Year 1, $12,000 saved, 7% of nearly $0 is nearly $0. After 10 years of $12,000 at 7%, the balance is $165,797. One year’s return on that pot is about $11,600 — still one year’s contribution. Only after the balance is several times annual saving do returns dominate the next twelve months.

That is why “I need a better fund” is the wrong first meeting. A 1-point savings-rate increase is a bigger 10-year lever than a 1-point return increase when the pot is small. Allocation becomes the main lever later. Withdrawal policy becomes the main lever when you start spending the pot.

5. Why people miss — label the claims

Lifestyle creep — argument, with a cross-section

CEX 2024: spending rises from $35,046 (lowest income quintile) to $150,342 (highest). That is a snapshot of different households, not proof that you will spend a raise. The argument is still the identity: if spending rises with income, the savings rate does not, and 25× spending gets harder.

Model (not a survey). Take-home $70,000, spend $59,500, save $10,500 → 15% rate. Target = 25 × $59,500 = $1,487,500. At 5% real: 43 years.

A $10,000 raise, all spent: take-home $80,000, spend $69,500, still save $10,500 → 13.1% rate. Target = $1,737,500. Same formula: 46 years. You earned more and bought three extra working years plus a more expensive life.

Save the raise instead: $20,500 / $80,000 = 25.6%. Same formula: about 31 years. That is the needle.

Low savings rate — evidence

SCF: 44% of families did not spend less than income in the prior year (the survey’s “saved” flag). BEA personal saving rate: 2.7% of disposable personal income in June 2026. Those two numbers are not the same concept. Together they say the leftover is small.

Sequence of returns — evidence

Bengen (1994): 4% in year 1, then inflation-adjusted, on a stock / intermediate-government-bond mix, did not exhaust a portfolio in under 33 years in the history he studied.

Trinity (1998), Table 3, inflation-adjusted withdrawals, 1926–1995, 30-year payouts:

AllocationSuccess rate at 4%
100% stocks95%
75% stocks / 25% bonds98%
50 / 5095%
25 / 7571%
100% bonds20%

Success meant money left at year 30. It is overlapping U.S. history, not a simulation of the next 50 years. FIRE withdrawals can last 40–50 years; that is a longer problem than the 30-year tables. Treat 4% as a starting heuristic, not a law. The point of sequence risk: a 30% drop in year 2 while you are withdrawing is not “the average 7%.” It is a hole you then withdraw through.

No written plan — evidence

Fidelity 2019: 18% had a comprehensive written plan. EBRI 2025: only about half of workers have even tried to compute a number; fewer have thought about withdrawals. A plan that exists only as “I’ll have a 401(k)” does not specify spending, the first-year withdrawal rate, or what you will cut after a bad market.

Systemic constraints — evidence, not a pep talk

  • Housing + transportation = 50.4% of average 2024 CEX spending.
  • SCF housing affordability at a modern-survey low (median home >4.6× median income).
  • Retirement-account ownership 49.6% under 35.
  • Stock ownership 34% in the bottom half of income vs 95% in the top decile.
  • Student debt in 22% of families (SCF 2022).

A 50% savings rate is not equally available. The math does not care. The budget does. The useful response is still the rate you can move: the next percentage point, the raise you do not fully spend, the match you do not leave on the table.

6. What actually moves the needle

First: savings rate. It raises the contribution and lowers the target at the same time. That double effect is why the years-to-FI table is so steep.

Second: allocation. Once money is invested, a 50/50 or 75/25 stock-bond mix is what the 4% historical tables were built on. An all-cash or all-bond pile is a different — and, in Trinity’s 30-year inflation-adjusted 4% column, much more fragile — object.

Third: withdrawal plan. Write, on one page: expected annual spending; first-year withdrawal rate; the inflation rule; the rule for a 20%+ drawdown year (for example: freeze the raise, cut flexible spending, delay a large purchase). Do this years before you quit.

Do not invert the order. A clever withdrawal rule will not repair a 5% savings rate. A 9% backtested return will not repair a 5% savings rate. A 25% savings rate in a boring global-equity/bond mix will.

7. One action this week

  1. Take last calendar month. Write take-home pay and spending. Savings rate = 1 − (spending ÷ take-home). Include the 401(k) deferral in saving.
  2. Write four lines: that rate; a 5% real return; a 4% withdrawal; 25 × last year’s spending.
  3. Enter the same inputs in the EarlyFIRE Target FIRE calculator on earlyfire.quest. Run 3.5% and 4.0%. You want a date, not a feeling.
  4. If the date is after 65, change one control this month: raise the workplace deferral by 1 percentage point, or pre-commit the next raise (half to the rate, half to life). Do not start with a new fund menu.

Reopen the calculator when the rate changes.

What this is not

This is not a claim that everyone can FIRE at 40. It is not a claim that 7% or 4% will repeat. It is not a survey of EarlyFIRE members, and it is not an EarlyFIRE Research Team paper — those do not exist here. It is the identities, the official surveys, and two worked future values.

If you remember one sentence: the savings rate sets the clock; allocation and withdrawal policy keep the clock from breaking.

Sources

  1. Aladangady et al. Changes in U.S. Family Finances from 2019 to 2022. Federal Reserve, October 2023.
  2. Congressional Research Service. Distribution of Retirement Account Balances (IF12928).
  3. BEA personal saving / DPI: 5.4% (2024), 4.6% (2025); 2.7% June 2026.
  4. U.S. Bureau of Labor Statistics. Consumer Expenditures — 2024.
  5. Bengen, William P. “Determining Withdrawal Rates Using Historical Data.” Journal of Financial Planning, October 1994.
  6. Cooley, Hubbard, and Walz. “Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable.” AAII Journal, 1998.
  7. Damodaran, Aswath. “Historical Returns on Stocks, Bonds and Bills: 1928–2025.” NYU Stern, January 2026.
  8. Fidelity Investments. Retirement Mindset Study, 2019.
  9. EBRI and Greenwald Research. 2025 Retirement Confidence Survey, Fact Sheet #3.