The biggest hidden assumption in every rent-vs-buy calculator is the return you assume on the money you don't put into a house. It's called the opportunity cost of the down payment, and it swings the answer more than any other single input — including the mortgage rate. Assume 2% return on the alternative and buying almost always wins. Assume 8% return and renting almost always wins. Neither of those numbers is defensible on its own.
Here is the honest historical evidence, how to pick a defensible assumption for your own risk tolerance, and a sensitivity table showing exactly how much the break-even year moves at 3%, 5%, 7%, and 9% real return.
Why this input dominates the calculation
Suppose you have $80,000 you could put toward a house. If you rent instead, that $80,000 goes into a portfolio. Over ten years:
- At 3% real return per year, $80,000 becomes $107,500 → an $27,500 gain
- At 5% real return per year, $80,000 becomes $130,300 → a $50,300 gain
- At 7% real return per year, $80,000 becomes $157,400 → a $77,400 gain
- At 9% real return per year, $80,000 becomes $189,400 → a $109,400 gain
The gap between the top and bottom of that range is $80,000 over ten years. That's the same order of magnitude as your entire down payment. Whether renting or buying wins the ten-year comparison depends almost entirely on which of those numbers you plug in as the alternative.
And that's before you compound it out to 20 or 30 years, where the gap swells to hundreds of thousands.
The takeaway: the opportunity-cost assumption is the single largest lever in the calculation, and calculators that let you skip it or hide a default at "1%" are cheating.
The historical evidence — nominal vs real, and why it matters
Every serious return number should be quoted in real (inflation-adjusted) terms, because a rent-vs-buy comparison already treats rent, prices, and taxes on an inflation-consistent basis. Mixing a nominal return on the alternative with a real growth rate on the house is a common mistake and it makes the alternative look artificially good.
Sources for the numbers that follow: NYU Stern (Damodaran) 1928–2025 historical return data set for US equities and Treasuries; Federal Reserve H.15 for the 3-month T-bill; BLS CPI-U for the inflation adjustment.
US large-cap stocks (S&P 500)
| Period | Nominal return | Inflation | Real return |
|---|---|---|---|
| 1928–2025 (98 years) | 10.1% | 3.0% | 6.9% |
| 1950–2025 (76 years) | 11.7% | 3.3% | 8.1% |
| 1970–2025 (56 years) | 10.5% | 3.8% | 6.5% |
| 1985–2025 (41 years) | 11.4% | 2.7% | 8.5% |
| 2000–2025 (26 years) | 8.2% | 2.6% | 5.5% |
| 2010–2025 (16 years) | 12.9% | 2.6% | 10.0% |
The centered range for US large-cap stocks over multi-decade windows is 6% to 8% real. Anyone who tells you they can predict which of those will occur over the next ten years is lying. The right assumption for planning purposes is somewhere in the middle: 6.5% real is a defensible baseline for a diversified US large-cap portfolio, but any calculator input from 5% to 8% real is inside the historical envelope.
Global stocks (developed markets)
Slightly lower than US-only. MSCI World returns 1970–2025 have been about 5.5%–6.5% real depending on your currency hedging assumptions. If you're saving toward a US-dollar house, unhedged global stocks add currency noise; hedged global stocks land closer to US-only.
US Treasury bonds (10-year)
Very different profile. Over 1928–2025 the 10-year Treasury returned roughly 4.9% nominal / 1.9% real. Over the 1970–2025 period it was 6.7% nominal / 2.9% real (the bond bull market since 1982 raised realized returns above expected returns). Since 2010 the real return has been essentially zero or slightly negative in some years.
The right assumption for a bond-heavy alternative portfolio is 1% to 3% real. If the alternative is entirely long-dated Treasuries, use 2%.
Short-term Treasury bills (cash equivalents)
3-month T-bills 1928–2025: 3.4% nominal / 0.4% real. This is close to zero, over the very long run. In 2020 T-bills yielded 0.1% nominal; in July 2026 they yield 4.35% nominal (T-Direct 4-week bill). The real yield right now is roughly 1.5%–1.8%, which is unusually high by historical standards. That won't last.
The right assumption for a cash-equivalent alternative portfolio is 0% to 1.5% real. Higher than that assumes today's real yields persist, which is not the base rate.
60/40 balanced portfolio
The most common "moderate risk" alternative: 60% US large-cap, 40% US Treasuries. Historically 4% to 5.5% real over multi-decade windows.
Pick a defensible return for the calculator
The right assumption for you depends on what you would actually do with the money. Not what a person with better discipline would do. Not what a robo-advisor's marketing suggests. Not what your uncle claims his portfolio returned.
Three anchors:
| Realistic alternative allocation | Defensible real return assumption |
|---|---|
| 100% low-cost global stock index, held through the drops | 6.5% |
| 60/40 balanced, held through the drops | 5.0% |
| High-yield savings / CDs / T-bills | 1.5% |
| "Some in stocks but I'd panic-sell in a bear market" | 3.0% |
| "I'd probably just spend some of it" | 0% |
The last row is a joke, but it's also true. The alternative-portfolio return you can defend is capped by the alternative-portfolio you'd actually hold and actually leave alone through a 30% drawdown. For most Americans, that number is honestly closer to 4% real than to 7% real, because the honest allocation includes some cash cushion and the honest behavior includes some suboptimal trading.
We default the calculator to 5.0% real because that's what a 60/40 portfolio has historically returned and it's roughly what a US-average household with a house-purchase timeline actually holds. Adjust for your situation.
Sensitivity: break-even year at 3%, 5%, 7%, 9% real
Now the honest table. Median case: US median rent $2,320/month, US median price $432,500, 30-year fixed at 6.72%, 15% down, 1% property tax, 3% rent growth, 3% home price appreciation nominal, 8-year length of stay, 6.5% selling costs. From our 2026 rent-vs-buy piece.
| Assumed real return on alternative | Break-even year |
|---|---|
| 9% real (100% equities, held perfectly) | 11.2 years |
| 7% real | 8.6 years |
| 5% real (60/40 baseline) | 6.7 years |
| 3% real | 3.9 years |
| 1% real (near-cash) | 2.4 years |
Look at the top and bottom rows. At 9% real assumed return, the break-even is at 11.2 years — meaning if the alternative return you can defend is 9% real, you should rent unless you're planning to stay 11+ years in the same house. At 1% real assumed return, the break-even is at 2.4 years — meaning if you'd honestly leave the money in savings, buying wins fast.
Same house. Same rent. Same mortgage rate. The break-even year quadruples based on which alternative return you plug in.
Two common mistakes
Mistake 1: using the historical average of the highest-return asset available in isolation. People plug in 8% or 10% because "that's what stocks return," and don't reflect the fact that they wouldn't actually be 100% in stocks with an amount of money they might need in a few years. The alternative portfolio for a housing decision should be more conservative than a 30-year retirement portfolio, because the housing decision is being made on a 5–15 year horizon.
Mistake 2: using nominal return on the alternative while treating everything else on a real basis. If you plug in "8% return on invested down payment" as a nominal figure, but you've set rent growth to 3% (which is real-basis, absent inflation) and home price growth to 3% (again, real-basis), you've quietly given the alternative a 3-percentage-point real edge. Match units — either everything is nominal or everything is real. We use real throughout, and default the return input to 5.0% real.
A concrete example: two identical households, different assumptions
Household A and Household B both make $130k, both have $80k saved, both are considering the same $500,000 house with a $2,500/month rental alternative in the same neighborhood. Both would put 15% down and take a 6.72% 30-year fixed.
- Household A plugs in 3% real return on the alternative — they'd hold cash and short-term bonds because they'd panic-sell equities in a downturn. Their honest number.
- Household B plugs in 7% real return on the alternative — they hold a 90/10 stock/bond portfolio, they've weathered 2020 and 2022 without touching it, and 7% is inside the range of what a global stock portfolio has actually returned over multi-decade periods.
The calculator gives:
- Household A: break-even year 3.2. Verdict: buy. In 10 years buying is ahead by $58,000.
- Household B: break-even year 8.9. Verdict: rent (if they'll be in this house less than 9 years). In 10 years buying is ahead by just $7,000; in 20 years buying is ahead by $95,000.
Same house. Same paycheck. Same mortgage rate. Totally different recommendation. That's the power of the assumption you make about your own discipline and own allocation.
Neither household is wrong. What matters is that each of them is using an alternative-return input that reflects what they would actually do, not a made-up default someone else chose for them.
The flagship rule
If you take one number away from this article, take this: the return you assume on the invested down payment is the input you most owe yourself an honest answer on. Not the mortgage rate — that one you can look up. Not the appreciation rate — that one is a genuine unknown for everyone. The alternative return is a knob that reflects your own behavior in a downturn, and you know that number better than any calculator does.
Set it too high and you'll rent when you should've bought. Set it too low and you'll buy when you should've rented. Set it to what you'd actually achieve — which for most people is 3% to 6% real, not 8%+ — and the rest of the math follows honestly.
Ready to see what the break-even looks like on your return assumption? Try the rent-vs-buy calculator → — the alternative-return input is a top-level knob defaulted to 5.0% real, with a citation to the historical evidence above and full sensitivity plotting.