The 4% rule says you can withdraw 4% of your savings in your first year of retirement, raise that dollar amount with inflation every year after, and expect the money to last about 30 years. It was worked out from the returns of broad stock and bond markets, and today the simplest way to own those markets is a handful of low-cost ETFs.
This guide covers which kinds of ETFs fit the rule's assumptions, which ones quietly break them, and what happened to real ETF portfolios that started withdrawals at the worst possible moment.
What the 4% rule assumes
The rule comes from financial planner William Bengen's 1994 study of US market history. He found that a retiree who withdrew 4% of a stock-and-bond portfolio in the first year, then raised the amount with inflation, would still have had money after 30 years in every period he tested, including the Great Depression and the inflation of the 1970s. Three assumptions carry most of the weight:
- A real mix of stocks and bonds. Bengen's portfolios held between half and three-quarters in stocks. Much more or much less made the rule less reliable.
- Whole markets, not bets. His numbers came from large US company stocks and intermediate-term government bonds, not from single sectors, themes or trading strategies.
- No fees. The research used index returns. Every fraction of a percent a fund charges comes out of the 4%.
The jobs ETFs can do in a 4% rule portfolio
Most 4% rule portfolios need only two or three funds, each with a clear job.
VOO or VTI
VOO owns the 500 largest US companies; VTI owns the whole US market, 3,531 stocks. Both cost 0.03% a year. This is the growth engine. In the 2008 crisis the whole US market fell —%, which is why it needs a partner. Compare S&P 500 funds.
BND
17,408 US bonds with an average maturity of 11.7 years, for 0.03% a year. In the 2008 crash it gained 7.5% while stocks fell. In 2022, when interest rates jumped, it lost 14.5%. Bonds cushion stock crashes; they are not risk-free. How bond funds behave.
TLT
Long-term Treasuries swing harder both ways. TLT rose 25.1% during the 2008 crash but lost 29.3% in 2022. It can make a good small hedge but a risky only bond fund.
VXUS
8,779 stocks outside the US for 0.05% a year. It spreads your retirement across more than one country's economy and currency.
SCHD
Dividend funds pay cash you can spend without selling. SCHD's trailing yield is 2.95%. The 4% rule does not need a high yield, though: you can sell shares for the rest. Compare dividend funds in our dividend ETF list.
Funds that quietly break the rule
TQQQ
Triples the daily move of the Nasdaq-100 and resets every day. In 2022 it fell 78.8%. Withdrawals taken during a drop like that come out of a shrinking pile with little left to recover. Why leveraged funds decay.
SH
Bets against the S&P 500. It rose in every crash, but markets mostly go up: a 4% rule test starting at its launch in Jun 2006 ran out of money in May 2018.
JEPI
Pays 7.96% a year, which makes 4% withdrawals look easy. Part of that payout is the market's future growth, sold in advance through options. Spend all of it, and the portfolio may not keep up with inflation.
What really happened: 4% withdrawals from real ETFs
We ran the rule on four mixes of two of the oldest ETFs: SPY for US stocks and AGG for US bonds. The first test starts in October 2007, just before the financial crisis; for a retiree, that is about the worst timing in this data.
Retired at the October 2007 market peak (Oct 2007)
| Portfolio | Balance, Nov 2026 | After inflation | Lowest point |
|---|---|---|---|
| 100% stocks | $3.02M | $1.88M | $458K Feb 2009 |
| 60% stocks, 40% bonds | $1.82M | $1.13M | $635K Feb 2009 |
| 40% stocks, 60% bonds | $1.32M | $823K | $744K Feb 2009 |
| 100% bonds | $586K | $364K | $585K Sep 2026 |
Retired ten years ago (Nov 2016)
| Portfolio | Balance, Nov 2026 | After inflation | Lowest point |
|---|---|---|---|
| 100% stocks | $3.02M | $2.17M | $1.00M Nov 2016 |
| 60% stocks, 40% bonds | $1.77M | $1.27M | $1.00M Nov 2016 |
| 40% stocks, 60% bonds | $1.32M | $950K | $1.00M Nov 2016 |
| 100% bonds | $684K | $492K | $679K Sep 2026 |
Across all 197 monthly starting points from Jul 2005 onward (each with at least five years of results), the 60/40 portfolio never ran out of money. In 93% of them it still held at least $1 million after inflation, its full starting buying power. The roughest ride was the Oct 2007 start, which dipped to $635K.
Costs come straight out of your spending
A fund's expense ratio is taken from its assets every year, in good markets and bad. On a $1 million portfolio:
| Fund | Expense ratio | Yearly cost on $1M |
|---|---|---|
| VOO Vanguard S&P 500 ETF | 0.03% | $300 |
| BND Vanguard Total Bond Market ETF | 0.03% | $300 |
| SCHD Schwab US Dividend Equity ETF | 0.06% | $600 |
| JEPI JPMorgan Equity Premium Income Fund | 0.35% | $3,500 |
| ARKK ARK Innovation ETF | 0.75% | $7,500 |
Keeping it on track
- Rebalance back to your mix. Selling some of whatever has grown and buying what has fallen is what let the 60/40 portfolio recover from 2009. (Our test rebalanced every month; once a year works similarly.)
- Take withdrawals from what is above target. Spend from bonds after stocks fall, and from stocks after a strong year, so you are not selling stocks at the bottom.
- Adjust, don't abandon. Small spending cuts in bad years make the money last noticeably longer. Selling everything after a crash locks in the loss.
The 4% rule is a starting point, not a plan
It ignores your taxes, Social Security, pensions and the way spending changes over retirement. ReadyAimRetire runs your own numbers, year by year, through good and bad markets.
Questions investors ask
Does the 4% rule work with ETFs?
Low-cost index ETFs are the closest thing to the market returns the rule was built on. In our test, a simple 60/40 mix of two ETFs survived every monthly starting point since Jul 2005. The rule's limits still apply: the future can be worse than the past, and fees, taxes and your own spending all matter.
How many ETFs do I need for a 4% rule portfolio?
Two or three can be enough: a broad US stock fund, a broad bond fund and, if you like, an international stock fund. Adding more funds rarely helps unless each one does a different job.
Is 4% still a safe withdrawal rate?
It held up through US market history and through every period in our ETF data, but it is a rule of thumb. It assumes a 30-year retirement and no change in spending, and it ignores taxes and other income. A personal plan is a better guide than any single number.
How we tested
Monthly prices to Nov 2, 2026, with distributions reinvested; fund prices are already net of their fees. Portfolios were rebalanced every month. Withdrawals were taken monthly: 4% of the starting balance a year, adjusted every month for US inflation (CPI-U). No taxes or trading costs. "After inflation" divides the final balance by the rise in prices since the start, giving its value in start-date dollars. For education only, not a recommendation to buy or sell any fund.