Retire at 55 With $2 Million or 65 With $4 Million? Pricing the Missing Decade

A financial planner put this question to followers on X in October 2026:

“Would you rather retire at 55 with $2 million or at 65 with $4 million?”

Zach Melloh, CFP, @zachmelloh26¹

Our Engineer Investor account answered with the other half of the trade:

“Everyone sees the extra $2 million. Almost nobody prices the missing decade. $2M at 55 → $4M at 65 implies a 7.2% nominal CAGR (~4% real at ~3% inflation), ignoring contributions and withdrawals. If you’ve already won the financial game, why spend another decade running up the score?”

via @egr_investor²

Both posts compare two portfolio balances on two different dates. A household deciding whether to stop work at 55 needs a different comparison: start from the same $2 million at 55, and ask what each retirement date lets it spend, for how long, and with what chance of running short. We ran that comparison through every U.S. starting year since 1928, then weighed it against what the research says about the decade itself.

At 3% inflation, $4 million at 65 is about $3 million today

Doubling in ten years with no money added or withdrawn takes a compound annual return of

Inflation takes a share of that. At 3% a year, the real return is $1.0718 / 1.03 - 1 = 4.06\%$, and $4 million in ten years buys what $2.98 million buys today. So the later retiree has twice the dollars but about 49% more purchasing power.

A 7.2% nominal return is also no sure thing over a single decade. We tested every rolling ten-year window from 1928 to 2025 using S&P 500 and 10-year Treasury total returns and CPI from Aswath Damodaran’s dataset.³ A 60/40 portfolio, rebalanced yearly, at least doubled in nominal terms in 53 of 89 windows and in real terms in 28. An all-stock portfolio doubled in nominal terms in 64 windows and in real terms in 43. The windows overlap, so they are not independent trials.

The 7.2% figure also overstates what a worker needs from markets, because a worker can add savings. Starting from $2 million, it takes about $59,000 of new savings a year to reach $4 million at a 5% return, or about $114,000 a year at 3%.

What ten more years of work buys

To compare the paths on equal terms, we started both with $2 million at 55 and required the money to last to 95. The early retiree draws spending from 55. The worker lives on pay until retiring, adds nothing to the portfolio and takes nothing out. For each starting year from 1928 through 1986 (the last year with 40 years of data), we found the highest constant inflation-adjusted spending that never ran out. The method is the same one behind the 4% rule, run over a 40-year span.⁴

For the 60/40 portfolio, the worst case was a retirement starting in 1966 for every age from 55 to 64 (1965 for 65), the start of the high-inflation years. Retiring at 55 into that sequence supported $69,200 a year, a 3.46% withdrawal rate. Working to 65 raised it to $107,700, 56% more, for a retirement ten years shorter. With $50,000 a year of new savings on top, the worst case at 65 rose to about $133,600.

The gains get larger with each year. In the worst case, the first extra year of work added about $2,500 a year of spending, the sixth about $3,700, and the tenth about $6,700. Two effects multiply: the untouched portfolio keeps compounding, and each year taken off a retirement raises the sustainable withdrawal rate by more than the year before, because the remaining spending is spread over fewer years. The median path shows the same pattern. Bronshtein, Scott, Shoven and Slavov reach a similar conclusion with a life-cycle model: delaying retirement by 3 to 6 months raises a worker’s sustainable standard of living as much as saving an extra 1% of pay for the last 30 years of a career.⁵

So the dollar return on another year of work rises rather than falls. That is a fair reply to “running up the score”: the score rises faster each year you stay. Whether a household needs it depends on what the portfolio has to pay for.

Can $2 million fund 40 years?

Morningstar’s December 2025 retirement income report gives forward-looking starting withdrawal rates with a 90% modeled chance of lasting, for fixed inflation-adjusted spending. The rate falls as the horizon lengthens: 4.4% for 25 years, 3.9% for 30, 3.5% for 35 and 3.3% for 40. The 30-year maximum comes from portfolios holding 30% to 50% in stocks; the 40-year maximum is the 50% portfolio only.⁶ At 3.3%, $2 million supports $66,000 a year, close to the $69,200 worst U.S. case above.

U.S. history is the most favorable record available. Anarkulova, Cederburg, O’Doherty and Sias rebuilt the safe withdrawal rate test across 38 developed countries. They find that a 65-year-old couple willing to accept a 5% chance of running out of money could withdraw only 2.31% a year, and a 55-year-old faces a longer horizon than that.⁷ A household relying on the table above should expect that a future sequence could be worse than the worst U.S. case.

Spending decides which path fits. With 60/40, no Social Security and no new savings:

• $60,000 a year lasted to 95 in all 59 starting years from 55. The extra decade would buy a bigger budget or a bequest.
• $80,000 a year from 55 ran out in 9 of 59 starting years. It lasted in every year from age 59.
• $100,000 a year from 55 lasted in only 32 of 59. It needed work to 64 to last in every year.

Social Security changes these lines a lot. Add a $40,000 benefit from 67, the same on both paths, and $80,000 a year from 55 lasted in every starting year, while $100,000 a year did from 58.

What the decade is worth

The decade is not guaranteed to whoever postpones it. In the Social Security Administration’s 2023 period life table, 89.4% of men and 93.4% of women alive at 55 are still alive at 65.⁸ The same table cuts the other way: if the two lifetimes are independent, at least one member of a 55-year-old couple reaches 95 about 20% of the time, and the early retiree’s money has to last that long. Period tables ignore future gains in longevity, and higher earners live longer than average, so both figures are likely conservative.

On health and well-being, the best studies disagree, and each measures a different kind of retirement:

• Gorry, Gorry and Slavov use U.S. Health and Retirement Study data and variation from Social Security and pension eligibility rules. They find retirement improves both health and life satisfaction. The satisfaction gain appears within four years; many health gains appear four or more years later.⁹
• Charles finds that the raw correlation between retirement and well-being is negative, because people often retire after bad news. Once he accounts for that, the direct effect of retirement on well-being is positive. His sample is men.¹⁰
• Dave, Rashad and Spasojevic, also with Health and Retirement Study data, find that complete retirement raises difficulties with mobility and daily activities by 5% to 14%, illnesses by 4% to 6%, and lowers mental health by 6% to 9%, relative to the sample average. The effects are smaller for people who are married, physically active, or still working part-time, and larger when retirement is involuntary.¹¹
• Kuhn, Staubli, Wuellrich and Zweimüller study an Austrian rule that let some workers leave the labor force up to three years early on extended unemployment benefits. For men, retiring a year earlier raised the risk of dying before 73 by 5.5% and cut age at death by 2.2 months. For women they find no significant effect.¹²

None of these is a study of a healthy, financially independent 55-year-old choosing to stop. Voluntary early retirement, a layoff and retirement forced by illness are different events, and they show up mixed in most data. The evidence does not support saying early retirement extends life, or that work keeps people healthy.

Work has its own value. Many people get structure, colleagues and purpose from a job they like, and the worker on the 65 path still has weekends, vacations and a family during that decade. For someone who likes the job, the cost of the 65 path is smaller than the ten-year gap suggests; for someone who dreads it, larger.

Plans to work to 65 often end early

The 65 path assumes ten more years of employment at current pay. In the 2026 Retirement Confidence Survey from the Employee Benefit Research Institute and Greenwald Research, workers expected to retire at a median age of 65, while retirees reported a median retirement age of 62, and nearly half of retirees said they retired earlier than planned.¹³ A plan that needs a decade of uninterrupted paychecks carries layoff and health risk that the early retiree has already removed.

Spending, taxes, health insurance and Social Security

The calculator holds spending flat after inflation. Research on actual retirees shows spending is less fixed. Hurst finds the average drop in spending at retirement is limited to food and work-related costs, and Aguiar and Hurst show that food intake holds steady even as food spending falls, because retirees cook and shop with more time.¹⁴¹⁵ Rohwedder, Hurd and Hudomiet find spending falls at older ages partly because interest in some activities declines with health and age, though close to 20% of people over 80 say they are not satisfied with their finances.¹⁶ A household with flexible discretionary spending can retire on less than the table’s worst case implies.

Four costs land on the early retiree that the table omits:

• Health insurance before Medicare. Up to ten years of Marketplace premiums once any COBRA coverage ends. The 400% of poverty-line subsidy cliff returned in 2026, so withdrawal choices set the premium; see The ACA Subsidy Cliff Is Back.
• Account access. The IRS lets you take 401(k) distributions without the 10% additional tax if you separate from service in or after the year you turn 55. IRAs do not qualify.¹⁷ Taxable accounts, 72(t) payments and Roth contributions cover the rest; see Withdrawing From a 401(k) or IRA Before 59½.
• Social Security. Benefits are based on your highest 35 years of indexed earnings, and missing years count as zeros.¹⁸ Someone with 35 high-earning years by 55 gains little from ten more; someone with gaps gains more. Stopping work and claiming are separate choices: for anyone born in 1960 or later, waiting from 67 to 70 raises the benefit to 124%.¹⁹ See When to Claim Social Security.
• Taxes. A $2 million traditional 401(k) and a $2 million taxable account with a high cost basis fund different amounts of spending. In return, the years between the last paycheck and Social Security are often the lowest-income years of a life, which makes them the cheapest time for Roth conversions.

Part-time work between the two dates

The original question offers two choices, but most people have a third. In our model, $40,000 a year of part-time pay from 55 to 60 raised the worst-case spending of the 55 path from $69,200 to $76,300. That is about what retiring fully at 58 bought, with five years of part-time work in place of three more years of full-time work. Dave, Rashad and Spasojevic’s finding that part-time work softens the health effects of retirement points the same way.¹¹

For many people the problem is the current job, its hours or its stress, more than work itself. Reducing hours is also the lever the money and happiness research supports most clearly.

Run your own numbers

Start with your own portfolio, the spending you want, and the Social Security estimate from your account at ssa.gov/myaccount. Then compare three settings: the age you could stop now, the age you planned to stop, and the earlier age with a few years of part-time pay. Pay attention to the line where your target crosses the worst-case curve, which is the earliest retirement age that survived every starting year in the record.

When retiring at 55 wins, and when 65 does

The earlier date looks stronger when:

• Essential spending is well under the worst-case line, or Social Security and a pension cover a large share of it.
• Spending can bend after bad market years.
• Health insurance to 65 is affordable and planned.
• The household places a high value on the years from 55 to 65, for health, family, caregiving or anything else that does not wait.
• Part-time or project work is available if markets go badly.

The later date looks stronger when:

• Spending at 55 would need a withdrawal rate above roughly 3.5% before taxes.
• Most of the portfolio is pre-tax, so taxes take a large share of each withdrawal.
• The household supports dependents, wants to leave a large bequest, or lacks 35 years of covered earnings.
• The work is satisfying and the job looks secure.

Frequently asked questions

Is $2 million enough to retire at 55?

For a 60/40 portfolio, spending of about $69,000 a year, before taxes and without Social Security, lasted to 95 in every U.S. starting year from 1928 to 1986. Social Security, part-time income or flexible spending raise that number. A 38-country study finds U.S. history flatters safe withdrawal rates, so leave a margin.⁷

What return turns $2 million into $4 million in ten years?

7.18% a year compounded, with no contributions or withdrawals. At 3% inflation that is about 4.06% real, and the $4 million buys what about $2.98 million buys today.

Does working one more year matter more than saving more?

Usually, late in a career. Bronshtein and coauthors estimate that delaying retirement 3 to 6 months does as much for sustainable spending as saving an extra 1% of pay for 30 years.⁵

Is it safe to use the 4% rule if I retire at 55?

The 4% rule was built for 30-year retirements. Morningstar’s forward-looking estimate for a 40-year horizon is 3.3%, and in U.S. history the worst 40-year case for a 60/40 portfolio was about 3.5%.⁶

Does retiring early shorten or lengthen your life?

The research does not settle it. U.S. studies using Social Security eligibility rules find retirement improves health and satisfaction, a fixed-effects U.S. study finds the opposite, and an Austrian study of induced early exits finds higher mortality for men only.⁹¹¹¹²

Can I collect Social Security at 55?

No. Retirement benefits start at 62 at the earliest. Retiring at 55 and claiming at 67 or 70 is common, and the portfolio carries the gap.¹⁹

Key takeaways

• Compare from the same starting point. The choice is between spending $2 million from 55 and working while it grows, and both paths should be judged from age 55.
• At 3% inflation, $4 million at 65 is about $3 million in today’s dollars, and a 60/40 portfolio doubled in real terms in 28 of 89 historical ten-year windows.
• In the worst U.S. starting year since 1928, a $2 million 60/40 portfolio sustained about $69,000 a year from 55 and about $108,000 from 65, with no new savings in between.
• The gain from each extra year of work grows. Whether a household needs it depends on its spending.
• The health evidence runs in both directions, and nearly half of retirees in EBRI’s 2026 survey stopped earlier than planned.
• Part-time work is often the better trade. Five years at $40,000 bought about as much security as three more full-time years in our model.

Related guides

• Is $1 Million Enough to Retire? for sizing a portfolio against spending, guaranteed income and flexibility.
• Die With Zero vs. FIRE for how to spend once the timing question is settled.
• Sequence of Returns Risk for why the 1966 retiree fared worst.
• How Much of Your Retirement Should Be a Guaranteed Income Floor? for covering essential spending before retiring early.
• Coast FIRE for the part-time and Barista FIRE versions of the middle path.

Sources and method

1. Zach Melloh, CFP, post on X, October 28, 2026.
2. Engineer Investor, post on X, October 28, 2026. The same author writes Summitward and the Engineer Investor account.
3. Aswath Damodaran, NYU Stern, Historical Returns on Stocks, Bonds and Bills: 1928–2025. S&P 500 and 10-year Treasury total returns; CPI from FRED series CPIAUCNS.
4. Method: for each starting year from 1928 through 1986, a $2 million portfolio at age 55 must last to 95. Returns are calendar-year S&P 500 and 10-year Treasury total returns, rebalanced annually. Withdrawals are taken at the start of each year and indexed to CPI. Before retirement the household adds only the stated savings and withdraws nothing. Bisection finds the highest constant real spending that never depletes the portfolio. No taxes, fees or health insurance premiums. The same code runs the calculator; its outputs were checked against an independent Python implementation on the same data.
5. Gila Bronshtein, Jason Scott, John B. Shoven and Sita N. Slavov, “The Power of Working Longer”, Journal of Pension Economics and Finance 18(4), 2019, 623–644. The model assumes Social Security is claimed at retirement.
6. Morningstar, The State of Retirement Income: 2025, December 3, 2025, Exhibit 6, p. 9. 90% modeled success, fixed real withdrawals, forward-looking return assumptions, Social Security excluded.
7. Aizhan Anarkulova, Scott Cederburg, Michael S. O’Doherty and Richard W. Sias, “The Safe Withdrawal Rate: Evidence from a Broad Sample of Developed Markets”, Journal of Pension Economics and Finance 24(3), 2025, 464–500.
8. Social Security Administration, Actuarial Life Table, period life table for 2023 as used in the 2026 Trustees Report. Survivors per 100,000 at 55, 65 and 95: men 88,436, 79,084 and 6,333; women 93,577, 87,399 and 12,612.
9. Aspen Gorry, Devon Gorry and Sita N. Slavov, “Does Retirement Improve Health and Life Satisfaction?”, Health Economics 27(12), 2018, 2067–2086.
10. Kerwin Kofi Charles, “Is Retirement Depressing? Labor Force Inactivity and Psychological Well-Being in Later Life”, Research in Labor Economics 23, 2004, 269–299.
11. Dhaval Dave, Inas Rashad and Jasmina Spasojevic, “The Effects of Retirement on Physical and Mental Health Outcomes”, Southern Economic Journal 75(2), 2008, 497–523.
12. Andreas Kuhn, Stefan Staubli, Jean-Philippe Wuellrich and Josef Zweimüller, “Fatal Attraction? Extended Unemployment Benefits, Labor Force Exits, and Mortality”, Journal of Public Economics 191, 2020, 104087 (DOI 10.1016/j.jpubeco.2019.104087). Figures are from the published abstract; the earlier working paper reported larger ones.
13. Employee Benefit Research Institute and Greenwald Research, 2026 Retirement Confidence Survey, April 21, 2026.
14. Erik Hurst, “The Retirement of a Consumption Puzzle”, NBER Working Paper 13789, 2008. Published as “Understanding Consumption in Retirement: Recent Developments,” in J. Ameriks and O. Mitchell, eds., Recalibrating Retirement Spending and Saving, Oxford University Press, 2008.
15. Mark Aguiar and Erik Hurst, “Consumption versus Expenditure”, Journal of Political Economy 113(5), 2005, 919–948.
16. Susann Rohwedder, Michael D. Hurd and Péter Hudomiet, “Explanations for the Decline in Spending at Older Ages”, NBER Working Paper 30460, 2022.
17. IRS, Retirement Topics: Exceptions to Tax on Early Distributions. Separation from service in or after the year the employee reaches 55 (50 for certain public safety employees); qualified plans yes, IRAs no.
18. Social Security Administration, Retirement Benefits: Stopping Work: “We base your retirement benefit on your highest 35 years of earnings” and “We use a zero for each year without earnings.”
19. Social Security Administration, Delayed Retirement Credits and the 1960-or-later delayed retirement page: “If you start receiving retirement benefits at age 70, you’ll get 124 percent of the monthly benefit.” Earliest claiming age 62 from Born in 1960 or Later.

Editor’s note

Educational content, not financial, tax or legal advice. Every dollar figure from our model is a historical outcome on U.S. data under the stated assumptions, not a forecast. A research memo behind this guide quoted Morningstar’s 30- and 40-year rates as if they came from the same portfolio and used the Kuhn et al. working-paper mortality figures; both were corrected against the published documents. Trade press reports 46% of retirees stopping early in the 2026 survey; we quote EBRI’s own “nearly half” because we could not open its fact sheet. Sources verified on October 28, 2026.