Lower Expected Returns Early Retirement Math at 3% Real

A McKinsey-backed forecast that ran on Bloomberg tells today's thirty-somethings to expect seven extra working years, or to save almost twice as much, because investment returns are about to fall. The standard rebuttal points out that long-range market forecasts have an almost unbroken record of being wrong. The two camps will never settle their argument, and the arithmetic that actually sets your retirement date never needed them to. The lower expected returns early retirement math is a sensitivity question, and you can compute it tonight: how many years is each point of real return worth at your savings rate?

Standard FIRE planning is a return forecast in disguise. The canonical savings rate to financial independence table, and the 25x rule that sets your FIRE number, both quietly assume roughly 5% real returns. Reprice the plan at 3% real and the stretch is brutally asymmetric: a 10% saver's timeline grows by about 18 years, a 25% saver's by about 8, and a 50% saver's by a bit over 2, while 65 to 70% savers lose closer to one year. Mark the target up from 25x toward 30x and a 50% saver loses roughly 5 more. Since you cannot control the return input, the robust response is to price the assumption and buy the gap back with levers you do control, where permanent spending cuts dominate for a reason most people miss.

The Forecast Debate Skips the Plan Arithmetic

One camp, led by institutional research desks, warns that a coming collapse in investment returns means young savers must work seven years longer or nearly double their saving to match the previous generation's outcome. The other camp counters that forecasters have predicted essentially every market disaster that never happened, and missed most that did.

What neither side does is open a FIRE spreadsheet. Whether forward returns fall is an argument. What a lower return assumption does to your specific timeline is arithmetic, and it has an exact answer. Three questions decide the plan:

1. What return does the famous savings rate table actually assume?
2. How much longer is the road if real returns come in two points lower?
3. Which lever buys those years back cheapest?

Each has a defensible answer, and none requires you to pick a side in the forecasting war.

The Famous Savings Rate Table Assumes 5% Real Returns

The savings rate to financial independence table that circulates every FIRE forum comes from Mr. Money Mustache's 2012 post on the shockingly simple math. Savings rate on one axis, years to financial independence on the other: about 51 years at a 10% savings rate, 32 at 25%, 17 at 50%, and 10.5 at 65%.

What the Table Quietly Assumes

The post states its inputs plainly. Returns of 5% per year after inflation, and a target of 25 times annual spending, which comes from the 4% rule. The closed-form version of the table is:

t = ln(1 + M · r · (1 − s) / s) / ln(1 + r)

where s is your savings rate, r is the real annual return, and M is the spending multiple (25 under the 4% rule). Plug in s = 0.50, r = 0.05, M = 25 and you get 16.6 years, which rounds to the published 17. Every published row reproduces to within a few months under the stated 5% input.

That is the whole point. The timelines are not facts about savings rates. They are facts about savings rates at 5% real returns. Change r and you have a different table. The return is not a footnote to the math; it is the load-bearing input. (The model also assumes constant real income and saving, with taxes and career breaks ignored, so treat every output as the center of a range.)

How Lower Expected Returns Stretch Early Retirement Timelines

How much longer to retire if stocks return 3% real instead of 5%? Hold the 25x target fixed and run the formula both ways:

Savings rate	Years to FI at 5% real	Years to FI at 3% real	Years added
10%	51.4	69.3	+17.9
25%	31.9	39.9	+7.9
50%	16.6	18.9	+2.3
65%	10.6	11.5	+0.9
70%	8.8	9.4	+0.6

The stretch is asymmetric because low savers depend on compounding to do most of the work. A 10% saver must eventually hold 225 years' worth of their annual saving; a 50% saver needs 25 years' worth; a 65% saver needs about 13.5. At 5% real returns, roughly three-quarters of the 10% saver's final portfolio is market growth, versus about a fifth for the 65% saver. Slow the market and the low saver loses the engine of the plan, while the high saver mostly loses a discount.

In dollars: earn $100k, save $50k, spend $50k, target $1.25M. At 5% real you arrive in 16.6 years, having deposited about $831k and collected about $419k of growth. At 3% real you need 18.9 years and deposit about $946k. Same target, $116k more of your own money. So if you have ever wondered how long to FI at a 50% savings rate, the honest answer is 17 years or 19 years, depending on which return the table was quietly using.

Rule of thumb to keep: divide the added-years column by two. Each point of real return is worth roughly 9 years at a 10% savings rate, about 4 years at 25%, about 1.2 years at 50%, and about half a year at 65% and above.

The 25x Rule Is Also a Return Forecast

The 4% rule is not a law of nature. William Bengen's 1994 analysis in the Journal of Financial Planning found that 4% initial withdrawals, adjusted for inflation, survived every 30-year window in modern US history for portfolios holding roughly 50 to 75% stocks. The Trinity study by Cooley, Hubbard, and Walz later confirmed high success rates for similar withdrawals on historical data.

An early retiree breaks two of those conditions. The horizon is not 30 years, it is 50 or more. And the historical sample embeds the very returns now in question. Longer horizons hurt because of sequence of return risk: losses in the first decade of withdrawals do permanent damage, since you are selling assets at depressed prices. Forward-looking withdrawal research, including the Safe Withdrawal Rate series at Early Retirement Now, commonly lands nearer 3% to 3.5% sustainable for long retirements on stock-heavy portfolios, which is a FIRE number of roughly 29x to 33x.

So does the 4% rule work if returns are lower? It was always an empirical result about a specific sample, not a guarantee. Prudent repricing means allowing that 25x may become 30x. Stack both effects, slower accumulation and a bigger target:

Savings rate	5% real, 25x target	3% real, 30x target	Total stretch
10%	51.4	74.7	+23.4
25%	31.9	44.3	+12.4
50%	16.6	21.7	+5.1
65%	10.6	13.4	+2.8

At a 50% savings rate, the two concerns together stretch 17 years to about 22. At 65%, under 11 becomes about 13.5. A 25% saver passes 44 years. You do not have to accept 30x; it is a margin decision. But the table above is the price of that margin, denominated in years.

Pricing the Real Return Assumption Instead of Predicting It

Plan to the range, not to a point. Respectable expert projections span from the mid-single-digit nominal returns for US equities that recent vintages of Vanguard's capital markets model have generally projected over the coming decade, down to the near-zero or negative real projections for US large caps that GMO's 7-year asset class forecast has printed in several recent editions. Two credible desks, two points of real return apart, and that width is the finding. A point estimate is unusable whichever camp issues it, because your plan inherits the full width of their disagreement. Plan across a 3% to 5% real band instead, and the argument stops being a bet you are making and becomes a sensitivity you have already priced.

Bogle's return decomposition explains why the spread exists at all. Expected stock returns are dividend yield plus earnings growth plus speculative return, the change in valuation multiples. The first two terms are observable and modest; you can look both up tonight. Essentially all of the expert disagreement lives in the speculative term, which is precisely the term nobody can forecast, and at elevated starting valuations that term has historically subtracted more often than it added. The forecast war is a dispute not about data but about the one input that is unknowable in advance.

The track record of market forecasts is poor enough that waiting for a winner is not a strategy. The lesson is not to ignore forecasts; it is to price the assumption instead of picking a number. Run the four corners, 5% and 3% real crossed with 25x and 30x, and treat the spread as the plan. A plan that only works at 5% is not a plan; it is a hope with a spreadsheet attached.

The Levers That Close the Gap, Ranked

Back to the $100k earner saving and spending $50k each. Repriced at 3% real with a 30x target, the timeline is 21.7 years against the original 16.6. The gap is 5.1 years. Rank the levers by gap closed per unit of sacrifice.

1. Cut spending permanently, because it works twice

Cut spending from $50k to $41k. Saving rises to $59k, and the target falls to 30 × $41k = $1.23M, below the original $1.25M. Run the formula: 16.4 years. An 18% spending cut restores the entire original timeline. Each dollar cut adds a dollar of annual deposits and removes $30 from the required target. It is the only lever that improves both sides of the FIRE identity, which is why it beats every alternative per dollar.

2. Raise income, which works once

Keep the $50k lifestyle, and the target stays $1.5M. To finish in the original 16.6 years at 3% real you need to save about $71k a year, meaning roughly a 20% raise with every new dollar saved. The same restoration that cost $9k of spending cuts costs about $21k of gross income. Raises absolutely work, and high earners often cannot cut 18%, but per dollar the cut is more than twice as potent.

3. Make withdrawals flexible, which trims the target

Research on retirement spending guardrails, including Guyton-Klinger-style rules popularized by Michael Kitces, suggests initial withdrawal rates meaningfully above 4% can be sustainable when spending flexes with portfolio performance. That pulls a 30x target back toward 25x, clawing back about 2.8 of the 50% saver's 5.1 years. The cost is real: you must actually cut spending in bad markets. It is the spending-cut lever again, just scheduled for later.

4. Recover cost drag, the controllable point of return

Fees are the one slice of real return you control outright. Broad-market index funds commonly charge 0.03% to 0.10% per year, while many actively managed options charge around half a percent to 1%, so the recoverable gap can approach a full percentage point. Turning 3% into 4% takes the 50% saver from 21.7 to about 20.1 years at a 30x target, or 18.9 to 17.7 at 25x. The lever is capped at whatever drag you were paying, but it is effortless and risk-free.

5. Work longer, which settles one-for-one

The damage is denominated in years, and extra work years pay in exactly that currency. Holding the 4% rule fixed, the 50% saver needs about 2 to 3 more years; accepting the 30x target makes it about 5. Delaying also shrinks exposure to a bad first decade of withdrawals. It is the least glamorous lever and the only one guaranteed to work.

6. Shift allocation, the circular lever

Adding equity risk to offset a lower equity premium is a strange bet: you are increasing exposure to the exact variable in doubt. Glidepath tweaks can be defensible, but allocation cannot be the whole answer to a return downgrade.

Rank	Lever	Buys back (50% saver)	Main caveat
1	Cut spending to $41k	5.3 years	The cut must be permanent
2	~20% raise, fully saved	5.1 years	Works once; taxes take a share
3	Guardrail withdrawals	~2.8 years	Requires real flexibility
4	Recover ~1 point of fees	~1.6 years	Capped at prior drag
5	Extra work years	1:1 with years accepted	Non-negotiable, unglamorous
6	Allocation shift	Uncertain	Bets on the premium in question

Stress-Test Your Own FIRE Plan

One sitting with a spreadsheet buys two outputs: your four-corner range, years to FI at 5% and 3% real crossed with 25x and 30x, and your personal sensitivity, the years each point of real return costs at your savings rate.

1. Compute four corners. Using the formula above, find years to FI at 5% and 3% real, each at 25x and 30x. Your plan's honest range is the spread, not any single cell.
2. Extract your sensitivity. Years added per point of real return: about 0.5 at a 65% savings rate, 1.2 at 50%, 4 at 25%, 9 at 10%. Write yours down; it is the number this whole debate reduces to.
3. Price your lever. For a spending cut, recompute the savings rate and the target together. For a raise, only the savings rate moves. For fees, add recovered drag to r.
4. Choose the withdrawal rule now. Fixed 4% means pricing a 30x target. Guardrails mean you can justify 25x, but only if you will genuinely cut spending in drawdowns.
5. Recheck annually against realized returns. If your realized real return keeps trailing the band, that is data. Adjust the savings side rather than the forecast.

Three loops, closed. The famous table assumes 5% real returns, and its timelines reproduce only under that input. The sensitivity ladder prices the rest of the question, because low savers lean on compounding and high savers lean on deposits. The canonical case is the 50% saver, whose 17 years stretch to about 22 under 3% real and a 30x target. And the lever that dominates is the permanent spending cut, the only move that raises saving and lowers the target at once.

You still cannot control the return. You now know exactly what it costs, and exactly which levers buy it back.