The 4% rule is the most cited number in retirement planning. It says: if you withdraw 4% of your portfolio in the first year of retirement and adjust for inflation each year after, your portfolio has historically survived 30 years in nearly every market scenario.
But what if your retirement lasts 40 or 50 years? And does it work for Canadians with CPP and OAS? Here's the full picture.
Where the 4% Rule Came From
In 1994, financial planner William Bengen analyzed historical US market data going back to 1926. He found that a 50% stocks / 50% bonds portfolio withdrawing 4% in year one (then adjusting for inflation each year) survived every 30-year rolling period in the historical record — including the Great Depression and the stagflation of the 1970s.
The "Trinity Study" (1998) extended this work and arrived at the same conclusion: 4% had a 95%+ success rate over 30 years. That rate became the standard benchmark for retirement planning.
The FIRE Problem: 30 Years Isn't Enough
The original research was designed for retirees at age 65 with a 30-year horizon. If you retire at 45, you may need your money to last 50 years. Every additional decade of retirement introduces more risk.
| Retirement Length | Safe Withdrawal Rate | Portfolio Multiple Needed |
|---|---|---|
| 30 years (retire at 65) | 4.0% | 25× annual expenses |
| 35 years (retire at 55–60) | 3.7% | 27× annual expenses |
| 40 years (retire at 50) | 3.5% | 29× annual expenses |
| 50 years (retire at 40–45) | 3.25% | 31× annual expenses |
The order of market returns matters enormously. Retiring in 2000 (dot-com crash) or 2008 (financial crisis) and withdrawing 4% from a falling portfolio depletes your balance faster — leaving less capital to recover when markets rebound. This is why early retirees often use a cash buffer of 1–2 years of expenses to avoid selling equities in a downturn.
Does the 4% Rule Still Work in 2026?
Some researchers argue current market conditions — lower bond yields compared to historical averages and elevated stock valuations — make 4% too aggressive going forward. Vanguard and others have suggested 3.3–3.5% as a more conservative starting point for new retirees today.
The counterarguments: the rule has survived every 30-year period including worse starting conditions; and retirees who spend flexibly (reducing withdrawals in bad market years by 10–15%) dramatically improve long-term survival rates without meaningfully impacting lifestyle.
The Canadian Advantage: CPP and OAS Reduce Your Required Portfolio
For Canadians, CPP and OAS function as a guaranteed income floor that reduces how much your portfolio needs to generate. If your retirement spending is $60,000/year and CPP + OAS covers $20,000 of that, your portfolio only needs to support $40,000/year.
At 4%, that means you need $1,000,000 — not $1,500,000. CPP and OAS effectively reduce your required FIRE number by the capitalized value of those benefits.
| Scenario | Annual Spend | CPP/OAS | Portfolio Draw | FIRE Number (4%) |
|---|---|---|---|---|
| No government benefits | $60,000 | $0 | $60,000 | $1,500,000 |
| Average CPP + OAS | $60,000 | $18,456 | $41,544 | $1,038,600 |
| Maximum CPP + OAS | $60,000 | $32,000+ | $28,000 | $700,000 |
Practical Adjustments That Improve Safety
- Guardrails method: If your portfolio drops 20%, reduce spending by 10%. If it rises significantly, you can increase spending. Dynamic withdrawals dramatically improve long-term outcomes.
- Cash buffer: Keep 1–2 years of expenses in cash or short-term bonds. Draw from this in market downturns instead of selling equities at depressed prices.
- Delay CPP/OAS: Every year you delay CPP past 65 increases it by 8.4%. Delaying to 70 increases your benefit by 42% — a powerful inflation-indexed income stream that reduces portfolio dependence.
- Part-time work (Barista FIRE): Even modest income ($10,000–$20,000/year) in early retirement dramatically extends portfolio life by allowing it to compound untouched during your most active years.
Frequently Asked Questions
Free Calculators
Model your retirement withdrawal rate with Monte Carlo simulation