How Much Do I Need to Retire?

The most widely used retirement framework is the 4% rule, which emerged from the 1998 Trinity Study. Researchers William Bengen (who originally proposed 4% in 1994) and later Philip Cooley, Carl Hubbard, and Daniel Walz tested withdrawal rates against every 30-year period in U.S. market history going back to 1926.

The finding: a retiree who withdraws 4% of their portfolio in year one, adjusts that dollar amount for inflation each year, and maintains a balanced portfolio (50-75% stocks, rest in bonds) had a 95%+ success rate of not running out of money over 30 years. That includes retirees who started during the Great Depression, the 1970s stagflation, and the dot-com bubble.

The math is simple. The inverse of 4% is 25. So your retirement number equals:

Why some experts now say 3.5%: The original study used 30-year periods. If you plan to retire early (say, at 55) or simply want to plan for a 35-40 year retirement, 4% has a higher failure probability over those longer horizons. Updated research from Michael Kitces, Wade Pfau, and the ERN (Early Retirement Now) blog suggests 3.25-3.5% is more appropriate for longer retirements. At 3.5%, you need roughly 28.6x your annual spending instead of 25x.

Bottom line: the 4% rule is a excellent starting point, not a guarantee. The more conservative you are, the more margin of safety you have. The table below shows the exact numbers at different withdrawal rates.

Many financial advisors use a simpler benchmark: plan to replace 80% of your pre-retirement income in retirement. The logic is that some current expenses disappear in retirement — you no longer save for retirement (obviously), payroll taxes drop, commuting costs vanish, and your mortgage may be paid off.

When the 80% rule works: If you currently earn $100K and spend about $80K, the rule aligns well. You need $80K/year, meaning $2M at a 4% withdrawal rate. It is a useful gut check.

When the 80% rule fails: If you are a high saver who only spends 50-60% of income, the 80% rule dramatically overshoots your actual need. If you earn $150K but live on $60K, you do not need $120K/year in retirement — you need $60K. Conversely, if you plan to travel extensively or have expensive hobbies, 80% may be too low.

The better approach: Track your actual spending for 3-6 months. Use your real expenses, not your income, as the foundation. Then add buffers for healthcare (which goes up), subtract items that go away (commuting, retirement savings), and you have a far more accurate target than any rule of thumb.

Your retirement number is built on one foundation: how much you actually spend. Not how much you earn. Not what your neighbor spends. Your real, tracked, documented expenses.

Most people dramatically underestimate their spending. Studies show Americans underreport spending by 20-30% compared to actual credit card and bank statements. This is the single biggest source of retirement planning errors.

Expenses that change in retirement: Commuting costs drop to near zero. Work clothes, lunches out, and dry cleaning go away. But travel spending often doubles. Healthcare costs rise significantly. And hobbies that were limited by work hours now consume real money. A reasonable estimate is that total spending stays roughly the same in the first decade of retirement, then gradually decreases after age 75 as activity levels slow, then rises again in the final years due to medical and long-term care costs. This is called the “retirement spending smile.”

Social Security is the largest source of retirement income for most Americans — and the most commonly misunderstood. The average monthly benefit in 2026 is approximately $1,907/month ($22,884/year). The maximum benefit for someone who earned the taxable maximum for 35 years and claims at age 70 is about $4,873/month.

When you claim matters enormously. You can start benefits as early as age 62 or as late as 70. Each year you delay past 62 increases your monthly benefit by approximately 6-8%. Claiming at 62 gets you 70% of your full benefit. Claiming at 70 gets you 124%. That is a 77% difference in monthly income for the rest of your life.

Healthcare is the wildcard in retirement planning and the expense most people dramatically underestimate. Fidelity's annual Retiree Health Care Cost Estimate pegs the average 65-year-old couple at approximately $315,000 in lifetime healthcare costs in retirement. That figure covers Medicare premiums, copayments, prescription drugs, and out-of-pocket expenses, but excludes long-term care (nursing home, assisted living), dental, and vision.

Long-term care is the elephant in the room. The median annual cost of a private nursing home room is approximately $108,000. Assisted living averages $54,000/year. About 70% of people over 65 will need some form of long-term care. Medicare does not cover it. This is the expense that bankrupts retirees who thought they had enough. Options include long-term care insurance (buy in your 50s before it gets expensive), self-insuring (setting aside $200K-$400K), or hybrid life/LTC policies.

Inflation is the silent tax on your retirement savings. $1 million today will not buy $1 million worth of goods in 20 years. At the historical average of 3% inflation, $1M has the purchasing power of only $554,000 two decades from now. At 4% (which we have recently experienced), it drops to $456,000.

This is why simply stockpiling cash in a savings account is not a retirement plan. Even at 5% interest, a savings account barely keeps pace with 3% inflation after taxes. You need investments that grow faster than inflation — which is why stocks (historically 10% nominal, 7% real) remain essential even in retirement.

Your $1.5 million portfolio is not $1.5 million of spending money. Depending on which accounts hold your savings, you could owe federal and state income taxes on every dollar you withdraw. Understanding the tax treatment of different account types is critical to knowing your real retirement number.

Social Security taxation: Up to 85% of your Social Security benefits are taxable if your “combined income” (AGI + nontaxable interest + half your SS benefit) exceeds $34,000 (single) or $44,000 (married). Most retirees with any significant investment income will pay taxes on their Social Security. This is another reason Roth conversions before claiming SS are valuable — Roth withdrawals do not count toward combined income.

State taxes: Nine states have no income tax (Alaska, Florida, Nevada, New Hampshire, South Dakota, Tennessee, Texas, Washington, Wyoming). Thirteen states tax Social Security benefits. Where you retire can save or cost you thousands per year. See our best states for taxes guide →

The optimal withdrawal order: Generally, draw from taxable accounts first (to take advantage of low capital gains rates), then traditional accounts (to fill low tax brackets), then Roth accounts last (letting them compound tax-free as long as possible). Meanwhile, do strategic Roth conversions each year to “fill up” low tax brackets.

The risk you do not hear about enough: outliving your money. A 65-year-old man has a 50% chance of living to 85 and a 25% chance of reaching 92. For women, the numbers are even higher — 50% chance to 88, 25% to 94. A 65-year-old couple has a 50% chance that at least one partner lives past 92.

That means a “30-year retirement” is not a worst-case scenario — it is the median for many couples. Planning for only 20 years of retirement when you might need 30 or 35 is one of the most consequential mistakes in financial planning.

The paradox of retirement planning: you need to save as if you will live to 95, but you might not make it past 75. This is not a reason to under-save — running out of money at 87 is catastrophically worse than leaving some behind. Err on the side of having too much. Your heirs will not complain.

Imagine two retirees, both starting with $1.5 million and withdrawing $60,000/year. Over 30 years, both experience the same average return of 7%. But the order of returns is different.

This is sequence of returns risk — and it is the reason why the first five years of retirement are the most dangerous to your portfolio. You have zero control over what the market does in those years. What you can control: