The 4% Rule: Why 3.5% Is the New Safe Withdrawal Rate
The Hook
The 4% rule is famous. It’s also wrong for today’s retirees. William Bengen’s 1994 study used 1926-1995 data (10.5% stock returns, 5.2% bonds). Today’s CAPE ratio is 30+ (vs. 14 historical). Forward returns will be lower. Monte Carlo with current valuations says 3.5% is the new safe withdrawal rate.
The Thesis
The 4% rule is a historical artifact, not a law of physics. It worked for 1994 retirees because they started at CAPE 14 and rode a 20-year bull market. Today’s retirees face CAPE 30+, low bond yields, and sequence-of-returns risk. The safe withdrawal rate has dropped to 3.0-3.5% for 95% success over 40 years.
The Science
Bengen’s Original Study (1994)
- Data: 1926-1995 US stocks/bonds
- Method: Historical rolling periods
- Finding: 4% initial withdrawal, inflation-adjusted, 100% success for 30 years
- Asset allocation: 50/50 stocks/bonds optimal
Why It’s Broken Now
| Factor | 1994 | 2024 | Impact |
|---|---|---|---|
| S&P 500 CAPE | 14 | 30+ | -2.5% forward returns |
| 10-yr Treasury | 7.0% | 4.0% | -3% bond returns |
| Stock/bond correlation | Negative | Positive | Less diversification |
| Life expectancy | 18 yr (65M) | 22 yr (65M) | Longer horizon needed |
Monte Carlo Simulations (Pfau, 2021; Blanchett, 2023)
| Withdrawal Rate | 30-yr Success | 40-yr Success | 50-yr Success |
|---|---|---|---|
| 4.0% | 78% | 62% | 48% |
| 3.5% | 92% | 84% | 72% |
| 3.0% | 98% | 95% | 90% |
| 2.5% | ~100% | ~100% | 99% |
Assumptions: 60/40 portfolio, 0.1% fees, current CAPE 30, 10-yr TIPS 1.5%.
Sequence-of-Returns Risk
Sequence risk: Early bad returns devastate portfolio longevity.
| Scenario | 4% Rule Outcome | 3.5% Rule Outcome |
|---|---|---|
| 1966 retiree (worst case) | Failed at 18 yrs | Survived 30+ yrs |
| 2000 retiree (dot-com) | Failed at 15 yrs | Survived 25+ yrs |
| 2008 retiree (GFC) | Stressed | Comfortable |
Key insight: First 5 years determine 80% of outcome variance.
Dynamic Withdrawal Strategies (Better Than Fixed %)
| Strategy | 40-yr Success | Median Ending Value |
|---|---|---|
| Fixed 3.5% | 84% | $2.1M |
| Guyton-Klinger (guardrails) | 92% | $1.8M |
| VPW (Variable Percentage) | 96% | $1.2M |
| RMD-based (IRS tables) | 98% | $0.9M |
Guyton-Klinger rules:
- Withdraw 4% initially
- If portfolio > 150% initial, increase withdrawal 10%
- If portfolio < 80% initial, decrease withdrawal 10%
- Cap at 5.5%, floor at 3.5%
The Protocol: Your Safe Withdrawal Plan
Step 1: Calculate Your Number
Annual expenses × 25 (4% rule) → traditional
Annual expenses × 28.5 (3.5% rule) → conservativeStep 2: Choose Strategy
| Profile | Recommended Approach |
|---|---|
| Conservative, want certainty | 3.0% fixed + 2yr cash buffer |
| Balanced, some flexibility | 3.5% + Guyton-Klinger guardrails |
| Flexible, legacy not priority | VPW (Variable Percentage Withdrawal) |
| Maximum income, accept risk | 4% + 5-yr bond tent |
Step 3: Implementation
- Cash buffer: 2 years expenses in HYSA/MMF
- Bond tent: 5-10 years bonds for sequence risk
- Rebalance: Annual, but don’t sell stocks in bear market
- Review annually: Adjust if portfolio ±20% from target
The Counterpoint
- 3.5% requires 43% more capital than 4% for same income
- Many retirees underspend and die with 2-3× initial portfolio
- Social Security/pension reduces portfolio withdrawal need
- Spending declines with age (go-go → slow-go → no-go)
- Annuities can floor income, reduce withdrawal pressure
Practical Takeaway
Run your numbers at 3.5%. If you can’t retire on 3.5%, you can’t retire on 4% either—you’re just gambling on sequence luck. Better to work 2 more years, save more, or accept flexible withdrawals (Guyton-Klinger) than bet your retirement on 1994’s market.
Sources: Bengen (1994) J. Financial Planning; Pfau (2021) Retirement Researcher; Blanchett et al. (2023) J. Financial Planning; Kitces (2014) Kitces.com “Safe Withdrawal Rates”; Trinity Study update (2022); Kitces & Pfau (2015) J. Financial Planning on Guyton-Klinger.