Key takeaways
- On the median UK local authority rent-to-price ratio of 4.44%, with a 25% deposit and a 4.92% mortgage, the buyer is £79,800 ahead of the renter after ten years.
- House price growth dominates. Across the 1.27% to 12.23% range of rolling ten-year UK windows since 1991, that ten-year gap swings by £555,000.
- The rent-to-price ratio comes second, worth £208,500 across the 2.77% to 7.02% spread measured across 316 local authorities in August 2026.
- The mortgage rate is worth £196,900 across its 1995 to 2026 range and the renter's investment return £171,800, so neither settles the question alone.
- At 1.27% house price growth the buyer is £35,100 behind at ten years. At 12.23% they're £519,900 ahead. Same house, same mortgage.
The rent vs buy break-even lands in month seven, and that isn't the finding
You want to know how long you'd have to stay in a house before buying it beats renting it. For a UK buyer in September 2026, on the middle set of assumptions, the answer is about seven months. After that the buyer stays ahead. Ten years in, the gap is £79,800.
That number is close to useless on its own, and the reason is the whole point of this piece. Move one input, house price growth, across the range the United Kingdom has actually delivered over rolling ten-year periods since 1991. The same ten-year gap now runs from minus £35,100 to £519,900. Nothing else changed: same house, same mortgage, same rent.
So the rent vs buy break-even is less a calculation than a growth forecast wearing a calculation's clothes. The useful question isn't which side wins. It's which assumption you're really arguing about when you argue about renting versus buying. What follows builds the model, then moves each input across its observed historical range, one at a time, and measures what happens to the ten-year wealth gap.
Every input in the rent vs buy break-even model, and where it came from
The base case is a UK home at £272,611. That's the UK House Price Index average for July 2026, published by the Office for National Statistics. The buyer puts down 25% and borrows the rest over 25 years on a repayment mortgage at 4.92%, which was the Bank of England's quoted rate for a two-year fix at 75% loan to value in August 2026.
Stamp duty on that price is £3,631: nothing on the first £125,000, 2% on the slice up to £250,000, and 5% above it. Add the £150 Land Registry portal fee that has applied to the £200,001 to £500,000 band since December 2024. Running costs are £780 a year, or £65 a month. That's the £11.90 a week of maintenance and repair plus £3.10 a week of structure insurance the Office for National Statistics measured across all UK households, renters included, in its Living Costs and Food Survey for the financial year ending 2025. Loading the same spending onto owners alone would raise it, which is one of several ways the model leans towards the buyer.
The renter's side is simpler. They pay rent, and they invest what the buyer handed over at the counter: the deposit, the stamp duty and the fee, £71,933 in all. Each month they invest the difference between the buyer's outgoings and their own rent. When the rent is the larger number, the difference comes out of the portfolio instead.
Rent is set as a ratio to the price of the house, because that ratio is the thing that actually varies. The base is 4.44% a year gross, the median across 316 local authorities where the August 2026 rent series and the July 2026 price series can be matched by area code. On a £272,611 house that's £1,009 a month, against a mortgage payment of £1,186 and £65 of upkeep, so the renter banks about £242 a month. House prices grow at 4.52%, the median rolling ten-year rate since 1991. Rents grow at 2.92%, the compound rate for England between January 2005 and February 2025. The renter's portfolio compounds at 9.08%, the MSCI World index's annualised gross return in dollars since the end of 1987.
That's the whole model. There are no selling costs, no conveyancing and no survey in it. Those exclusions favour the buyer, and they get their own section below.
House price growth swings the ten-year gap by £555,000
Nationwide's monthly index starts in January 1991. Every overlapping ten-year window in it up to August 2026 gives 308 readings of what a decade of UK house price growth has actually looked like. The worst was 1.27% a year, for the decade from October 2007. The best was 12.23%, for the decade from January 1997. The median was 4.52%.
Run the model at the bottom of that range and the buyer is £35,100 behind after ten years. Run it at the top and they're £519,900 ahead. That's a swing of £555,000, and it's larger than the next two levers in the model put together. The chart plots all five bars on the same scale.
Two things follow. Anyone quoting you a break-even period has embedded a house price forecast in it, whether they said so or not. And the range isn't exotic. Both endpoints are decades that happened, to the same country, inside the working life of anyone buying now.
It also explains why the argument never resolves. Someone who bought in 1997 and someone who bought in 2007 are both describing what really happened to them, and they're describing opposite outcomes. Nationwide's inflation-adjusted series makes the second case plain. In real terms, UK house prices peaked in 2007 Q3 at £369,595 in today's money, and the 2026 Q2 level of £278,784 is still 24.6% below that. Over the full series, from 1975 Q1, real growth has run at 1.36% a year. Set that real record against shares and you have the argument in our piece on your house as an investment.
The rent-to-price ratio is the second lever, and it runs from 2.77% to 7.02%
The next largest input is the one most calculators ask you to guess. Matching the August 2026 rent series to the July 2026 price series across 316 local authorities gives a gross rent-to-price ratio with a median of 4.44%. The tenth percentile is 3.59% and the ninetieth is 5.52%. The extremes are 2.77% in Derbyshire Dales and 7.02% in Newcastle upon Tyne.
Move the base case across that spread and the ten-year gap runs from minus £2,100 to £206,400, a swing of £208,500. At the lower quartile of 3.99% the buyer is £57,700 ahead at ten years. At the upper quartile of 5.02% they're £108,300 ahead.
The mechanism is plain enough. The rent-to-price ratio decides whether the renter is banking money each month or burning it. At 4.44% the rent is £1,009 against the buyer's £1,251 of mortgage and upkeep, so the renter saves £242 a month. At 2.77% the rent is £629 and the renter saves £621. That's a different investment plan, not a rounding difference.
It also means a national answer to renting versus buying isn't portable to your street. The same house at the same price is a different trade in Newcastle and in the Derbyshire Dales, and nothing about the buyer changed. This is the arithmetic sitting underneath the claim that rent is dead money, which turns out to be true of a large slice of a mortgage payment too.
The mortgage rate and the return on the deposit finish third and fourth
The Bank of England's quoted two-year fixed rate at 75% loan to value has 380 monthly readings, from January 1995 to August 2026. It peaked at 8.38% in February 1995 and bottomed at 1.2% in September 2021, averaging 4.276%. Move the model across that range and the ten-year gap goes from £176,100 at the low end to minus £20,800 at the high end, a swing of £196,900.
The renter's investment return does slightly less work. Take the range from 3.4%, the nominal return on United States Treasury bills from 1900 to 2025 in the Dimson, Marsh and Staunton database, up to 13.56%, the MSCI World index's annualised return over the ten years to August 2026. The gap runs from £158,400 down to minus £13,400, a swing of £171,800.
That ordering is the surprise in the model. The return the renter earns on the deposit is the assumption people argue about hardest, and it moves the answer less than the growth rate on the house does. Leverage is why. The buyer is exposed to house price growth on the whole £272,611, while the renter's portfolio starts at £71,933. A point of house price growth and a point of investment return simply aren't the same size of thing. The same effect drives our piece on mortgage leverage, where a modest growth rate turns into a much larger return on a deposit.
Rent inflation comes fifth. Across England's observed range, from minus 1.34% in April 2010 to 9.24% in November 2024, the ten-year gap moves from £45,900 to £146,000. A £100,100 swing is real money, but it's the smallest of the five.
The counter-case: this model is built to flatter the buyer
The strongest objection is that the base case leaves out costs only the buyer pays. There's no conveyancing, no survey and no estate agent fee, because none of those has a published tariff the way stamp duty and the Land Registry fee do. The direction of that bias is knowable even when the size isn't. Charging a 3% selling cost cuts the ten-year gap from £79,800 to £67,100, which is a dent rather than a reversal. Itemised on the July 2026 average home, the cost of moving house comes to £9,329, of which only £3,800 carries a published tariff.
The second objection goes to the rent-to-price ratio itself. The rent series and the price series measure different housing. The rented stock skews towards flats and smaller homes, and the transacted stock doesn't. Matching by local authority controls for geography but not for property type, and the national figures show how much that matters. Matched by type across the UK, the implied gross ratio runs from 4.29% on detached homes to 8.51% on flats. That isn't a credible like-for-like spread. It's a mix artefact, and it's exactly why the model treats the ratio as a range rather than a point estimate.
Third, the renter's 9.08% is a gross dollar return on a developed-market equity index since the end of 1987. A sterling investor pays fund charges and some withholding tax, and carries currency risk on top. That flatters the renter, which partly offsets the missing costs on the buyer's side. Two errors pointing in opposite directions don't cancel to zero, though, and it would be dishonest to pretend they do.
Fourth, the model holds the mortgage at 4.92% for the full 25 years. A borrower on the two-year fix that rate is quoted for doesn't. They refix into whatever rate exists at the time, which is precisely why the 1.2% to 8.38% range matters more than the starting rate does.
What the data cannot tell you
Every range here is a sample of the past. The 308 ten-year windows since 1991 overlap heavily, so they're nowhere near 308 independent observations of a decade. A backtest is not a forecast, and a country that ran one housing cycle inside that sample could run a different one next.
The deflator matters as well. Nationwide's real series converts nominal prices using the Retail Prices Index. A different price index would move the 1.36% a year that UK house prices have returned in real terms since 1975 Q1, and with it the case for treating the housing inflation hedge as dependable.
The equity figures are United States data doing global work. US equities returned 6.6% a year in real terms from 1900 to 2025 in the Dimson, Marsh and Staunton database, and that market today accounts for around 62% of total world equity market value. Survivorship isn't a small problem here.
The housing growth range is one country's, too. The S&P Cotality Case-Shiller national index for the United States stood at 336.663 in June 2026 against a January 2000 base of 100, which works out at 4.70% a year in nominal terms. That's close to the UK median. The path was not, and a path with a crash in the middle of it is not the same asset as one without.
One more limit is worth naming. The model compares wealth, and wealth is not housing. It cannot tell you what a secure tenancy is worth, or what it costs to move every twelve months, or what a leaking roof does to a household with no savings left after the deposit. None of that shows up in any of the five bars.
What would change the conclusion
If house price growth over your particular decade lands near the bottom of its historical range, the arithmetic reverses. At 1.27% growth combined with a 2.77% rent-to-price ratio, the buyer is £117,000 behind after ten years. That pairs a national growth rate from one decade with a local rent ratio measured today, so it's a construction rather than an observation. Neither number is invented.
Two thresholds in the model are worth knowing. At a 5.22% rent-to-price ratio the rent exactly equals the mortgage payment on this house, and at 5.51% it covers the upkeep too. Above that the renter is spending more each month than the buyer, and buying wins on cash flow before it wins on anything else. Below about 4.25% the buyer's ten-year lead survives, but the renter's portfolio overtakes at some point inside the model's 40-year horizon. At a 4% ratio the ten-year gap is still £58,200. The long-run one isn't.
The financing details move it less than people expect. First-time buyers relief removes stamp duty below £300,000 and lifts the ten-year gap from £79,800 to £88,500. A 10% deposit at the Bank of England's quoted 90% rate of 5.16% in August 2026, which is 0.24 percentage points above the 75% rate, lifts the gap to £94,000 rather than cutting it, because leverage runs in the buyer's favour at these growth rates. That's not a free lunch. It's the same leverage that left real prices 24.6% below their 2007 peak and has held them there ever since.
So the thing to watch isn't the break-even year at all. It's the rent-to-price ratio on the specific street, which you can measure this afternoon, and the growth rate you're silently assuming when you quote anyone a break-even at all. Two decades of UK data say the first is knowable and the second isn't.