SPY plus enough T-bills for 20 starting monthly withdrawals.
90/10 SPY and T-bills portfolio: returns and retirement test
Strategy in short: Hold 90% in SPY and 10% in three-month T-bills. Take withdrawals from T-bills first. Once every 12 months, rebalance back to 90/10.
Retirement answer: The T-bill cushion supported only slightly more starting spending: 3.5% versus 3.4% a year funded 30 years in at least 95% of both sets of histories. These are historical thresholds, not recommended spending.
Return comparison: Holding only SPY made more money. $10,000 grew to $251,080 with 10% in T-bills and $309,542 in SPY.
Growth of $10,000 — logarithmic scale
The cash cushion did reduce the deepest fall—from about 51% to 46%. The price was lower growth: 10.1% a year instead of 10.8%.
See what the 10% in T-bills actually changed
T-bills absorbed part of every fall, but they also missed part of every rise. Once a year, rebalancing sold SPY after strong periods and bought SPY after weak periods to restore 90/10.
Between those annual resets, the cash share drifted between 7.5% and 15.4%. The strategy therefore kept a cash cushion, but not a permanently fixed one.
| Result | 90% SPY / 10% T-bills | Always SPY |
|---|---|---|
| Ending value of $10,000 | $251,080 | $309,542 |
| Annual return | 10.1% | 10.8% |
| Deepest fall | -46.4% | -50.8% |
Would holding more T-bills work better?
More cash reduced the fall further, but it still did not create a large retirement advantage.
A 67% SPY / 33% T-bill version fell 35.5% and grew $10,000 to $148,560. Its 30-year withdrawal level that funded at least 95% of both history tests was 3.6%, only 0.1 percentage point above the 90/10 result.
This sensitivity points to a trade-off, not one magic cash allocation: more T-bills softened losses and also gave up more of SPY's long-run growth.
90% SPY / 10% T-bills10.1% a year-46.4% deepest fall
Always SPY10.8% a year-50.8% deepest fall
February 1993–June 2026. SPY distributions are reinvested. The strategy restores 90/10 once every 12 months and pays 0.10% of SPY bought or sold. There are no withdrawals in this graph. Equal percentage moves take equal vertical space. Past results do not predict future results.
Did the smaller falls support more monthly spending?
A little. For a 30-year retirement, the highest starting withdrawal rate that funded at least 95% of both history tests was 3.5% with the cash cushion and 3.4% with SPY.
Cash helped most when shares were depressed. Withdrawals came from T-bills first, so fewer SPY shares had to be sold during a fall. The lower stock allocation also meant less growth was available when markets rose.
Example: You retire before the financial crisis with $500,000 and need $2,500 a month—6% a year.
The inflation-adjusted spending now equaled 13.0% a year.
The same spending now equaled 13.8% a year.
This follows the real monthly returns from November 2007 through March 2012. Spending rises with inflation and occurs before each month's return.
$500,000 with $2,500 monthly withdrawals through the financial crisis
A 6% starting withdrawal was still demanding. At that rate, 90/10 funded 342 of 401 ordered starts and 3,217 of 5,000 mixed histories. Always SPY funded 344 and 3,270, respectively. The lower drawdown did not overcome the lower growth at this higher spending level.
What funded at least 95% of both history tests?
The 90/10 strategy supported a 3.5% starting annual withdrawal, compared with 3.4% for SPY. On $500,000, that difference is about $42 a month.
These are historical thresholds, not recommended spending or a promise of future success.
Test whether 90% SPY and 10% T-bills can fund your early retirement
Enter your portfolio, monthly spending, and how long the money needs to last.
$2,500 a month for 30 years starts at 6.0% of the portfolio each year.
In these backtests, sometimes. Neither strategy funded this early retirement in at least 8 out of 10 histories tested.
401 starting months; the fund's record repeats in order when needed
- 90% SPY / 10% T-bills
- 342 funded
- Always SPY
- 344 funded
5,000 histories built from 8–12-year pieces
- 90% SPY / 10% T-bills
- 3,217 funded
- Always SPY
- 3,270 funded
A path funded the goal only if it paid every inflation-adjusted monthly withdrawal for 30 years. Results use the nearest 0.1 percentage-point starting withdrawal rate. They reuse one 33-year market record and are not a probability about your future.
How the FIRE calculator works
The calculator applies the portfolio value, monthly withdrawal, and number of years you enter. The first withdrawal occurs before the first month's return, and later withdrawals rise with U.S. inflation. A history counts as funded only if it pays every monthly withdrawal for the full period and finishes above zero.
One result begins at each of the 401 months in the fund record and follows that record forward in order, returning to its first month when more months are needed. The other builds 5,000 longer histories from linked 8-, 10-, and 12-year pieces of the same record. The calculation charges 0.10% whenever SPY is bought or sold and excludes personal tax.
Strategy in detail
The portfolio starts with 90% in SPY and 10% in three-month Treasury bills. Each month's withdrawal comes from T-bills first. If they run out, only the SPY needed for spending is sold. At the start and once every 12 months, the portfolio returns to 90/10.
The theory is that a small cash reserve gives retirement spending somewhere to come from while stocks are down. Annual rebalancing also buys some SPY after weak years and sells some after strong years. Investor.gov explains that rebalancing restores a portfolio to its chosen asset mix and that investors often do it on an infrequent calendar schedule.
The test supports that mechanism, but shows its size. Ten percent in T-bills reduced the worst fall by about four percentage points. It improved the conservative 30-year withdrawal threshold by only 0.1 percentage point and reduced the ending value by about $58,462.
Can I trust these results?
Trust the direction more than the exact 0.1 percentage-point advantage.
The direction is credible because the rule is simple and was not fitted to a market signal: holding less stock should normally soften stock losses and give up some stock gains. The complete direct SPY history showed both effects.
The exact retirement advantage is fragile. It is one step on a 0.1-point grid, and both history tests reuse the same 33-year U.S. record. At a 6% starting withdrawal, always holding SPY actually funded slightly more histories.
A 10% T-bill cushion is a modest risk trade-off. This test does not establish it as clearly better than simply holding SPY.
Methodology details for the nerds
How I ran the return comparison
Both portfolios start with $10,000 in February 1993 and follow all 401 months through June 2026 in order. SPY distributions are reinvested. The 90/10 portfolio restores its target at the opening and every 12 months. The comparison holds SPY throughout.
I subtract 0.10% of every SPY purchase or sale, including the opening allocation and annual rebalances. T-bills use the published three-month Treasury rate. The growth graph has no withdrawals, inflation adjustment or personal tax.
How the retirement histories were built
The first panel begins in each of the 401 possible months and keeps the market record in order, repeating it when a longer retirement needs more months. The second panel links 5,000 sequences of 8-, 10- and 12-year pieces from the same record.
Portfolio holdings, withdrawals, inflation and the annual rebalance clock continue across every join. Nothing resets when one historical piece ends and the next begins.
Data and important limits
SPY returns use adjusted prices. T-bill returns use Federal Reserve interest-rate data, and withdrawals rise with U.S. Bureau of Labor Statistics CPI data.
The model excludes personal tax, currency changes and country-specific account rules. Rearranged histories are stress tests, not independent forecasts. This is educational research, not personal investment advice.
What evidence would change the conclusion?
A later independent period in which the 90/10 portfolio repeatedly supported meaningfully more spending would make the case stronger. If the advantage remains one small grid step—or disappears after tax, different rebalancing dates or longer history—the practical conclusion should stay that this is a trade-off, not a superior strategy.
Keep comparing
See where this strategy ranks.
Use this result
See how much you need to retire—and compare what to invest in.
Enter what you have and what you want to spend each month. Compare three portfolios against the same market history. Your financial figures stay in your browser.
Build my free FIRE plan