Renting or Buying: the Break-Even in Years, and What Actually Moves It

The verdict first, and the model it comes out of

Buying wins from year 6 to year 8, depending on where you buy. That is the answer, and everything after it is the reason. The model behind it is a wealth comparison, not a monthly-payment comparison: at the end of each year, the buyer's position is the house at market value, minus the cost of selling it, minus the loan still outstanding; the renter's position is the deposit and the purchase costs invested from day one, plus every year's difference between the owner's outgoings and the rent, compounded. The break-even is the first year the buyer's position is the larger of the two.

The inputs are stated so you can disagree with them precisely rather than vaguely. Price 300,000 euros; deposit 20 %; a 25-year fixed loan at 3.2 % on the rest; running costs — maintenance, insurance, property tax, building charges — at 1.2 % of the price a year; prices and rents both growing 2 % a year; a gross rental yield of 4.2 %, so 12,600 euros of rent a year at the start; the renter earning 3 % a year after tax on everything invested; and 4 % to sell. Only the purchase costs are sourced facts here — the rest are assumptions, and they are the same assumptions in every country so that the country column is the only thing changing.

What a purchase costs on the way in: France, Spain, Portugal in 2026

In France the charge is called frais de notaire and the name is a lie of scale: the notary's own regulated fee is a small part of it, and the bulk is the transfer duty, which since 1 January 2026 runs from 5.81 % to 6.32 % depending on the département — first-time buyers of a main home keep the earlier rate. Add registration and disbursements and an existing home costs roughly 7 to 8 % of the price to buy. A new build, whose first sale falls under VAT rather than the full transfer duty, costs about 2 to 3 %. That gap inside a single country is worth three years of break-even on its own, which is more than the whole distance between Madrid and Catalonia.

Spain has no national answer at all, because the transfer tax on a resale is set by each autonomous community: 3 % in Ceuta and Melilla, 4 % in the three Basque provinces, 6 % in Madrid, 7 % in Andalusia, 9 % in Cantabria and Castile-La Mancha, and 10 to 13 % at the top of the Catalan and Balearic bands. Alongside it, Spain's notary and registry fees are startlingly small — 0.2 to 0.5 % — because the two statutory scales that set them are 1989 pesetas converted to euros in 2002 and never revalued since, so the same nominal fee shrinks as a share of every rising price. Portugal charges a progressive IMT, gentler on a permanent main home than on a second home and fully exempt below a threshold, plus a flat 0.8 % stamp duty, for a total of roughly 6 to 10 %.

The ranking of levers, computed rather than asserted

Run the same model three times, changing one input at a time, and the hierarchy is unambiguous. Purchase costs from 4 % to 11 % of the price — the entire European spread, from a Basque province to a top-band Catalan flat — move the break-even from year 5 to year 9. The gross rental yield from 3.0 % to 5.5 % — roughly the distance between central Paris and Lisbon — moves it from year 18 to year 4. The loan rate from 2.0 % to 4.5 %, with the renter's return moving in step so the comparison stays fair, moves it from year 4 to year 22. On this model the rate is the biggest single lever, the rent-to-price ratio is a close second, and the purchase cost is a distant third.

This matters because the popular version of this argument has the order backwards, and the practical advice that follows from it is wrong. If you believe purchase costs dominate, you shop for a cheap-transfer-tax region and stop thinking. What the arithmetic says instead is that a Madrid flat at a 3.0 % gross yield has a worse break-even than a Catalan flat at 5.0 %, even though Catalonia charges nearly twice the transfer tax. The transfer tax is a fact about the region; the rent-to-price ratio is a fact about the specific building, and it is the one you can actually shop for.

Why purchase costs and the interest rate multiply instead of adding

Here is the result that rescues the intuition the ranking above appears to demolish. Purchase costs are worth almost nothing when money is cheap and a great deal when it is expensive. At a loan rate of 2.0 %, moving purchase costs from 4 % to 11 % moves the break-even from year 3 to year 5 — two years across the whole spread. At 3.2 % the same move is worth four years, from 5 to 9. At 4.5 % it is worth fourteen years, from 15 to 29. The one-off bill never changes; what changes is the size of the annual advantage it has to be recovered out of, and a high rate shrinks that advantage.

The practical reading is that the two questions are not separable. A buyer in a 2 % world can be casual about a Catalan transfer tax; the same buyer in a 4.5 % world is signing up to fourteen extra years of ownership for the same choice. This is also why the French new-build exemption changes behaviour so sharply in high-rate years and barely at all in low-rate ones: 2 to 3 % against 7 to 8 % is a small nudge when the annual advantage is fat, and a decisive one when it is thin.

The exit, which most comparisons quietly leave out

A break-even year only means something if you can reach it, and reaching it means not selling before it. The model above charges 4 % to sell, which is an assumption about estate-agent commission and nothing else. It does not charge capital gains tax, because in all three countries the sale of a permanent main home is treated favourably: France exempts it outright, Portugal exempts the gain to the extent it is reinvested in another permanent main home within the statutory window, and Spain exempts reinvestment in a habitual residence and exempts sellers over 65 entirely. Sell a property that is not your main home and none of that applies, and the break-even you computed becomes fiction.

The other exit cost is the one nobody prices: mobility. A renter who moves for a job loses a deposit and a month's notice. An owner who moves in year 3 has paid the whole entry bill and recovered none of it, which in France is roughly 22,500 euros on a 300,000-euro flat, plus the selling commission on the way out. That is the real reason the break-even year is the right question — not because renting or owning is better, but because the answer is a statement about how long you intend to stay, and everything else is a modifier on it.

What would change this verdict

Four things, in order of force. Flat house prices instead of 2 % growth push the break-even from year 7 to year 20 in the French case — house price growth is doing more work in this model than any tax. A gross rental yield below 3.5 % pushes it past a decade regardless of where you buy. A loan rate above 4.5 % pushes it past twenty years unless prices are rising faster than the assumption. And a renter who does not actually invest the deposit — who spends it — hands the verdict to the buyer at almost any purchase cost, which is the least mathematical and most common reason buying wins in practice.

Frequently asked questions

Does a higher transfer tax always mean a later break-even?

All else equal, yes, but all else is almost never equal. In this model each extra 1.75 points of purchase cost is worth about one additional year, so the whole spread from a Basque province at roughly 4.4 % to a top-band Catalan flat at roughly 10.4 % is worth about four years. A rent-to-price ratio two points lower is worth more than that on its own. So a high-tax region with generous rents can beat a low-tax region with thin ones, and the only way to know is to run both with the same assumptions rather than compare the tax rates.

Is the French new-build exemption really worth three years?

On these assumptions, yes: the same house at 2.5 % entry cost breaks even in year 4 against year 7 at 7.5 %. The reason is that a first sale within the VAT regime escapes the full transfer duty and pays a much smaller registration charge instead. Two cautions before you treat that as free money. A new build is typically priced above a comparable resale in the same street, and any price premium eats the saving directly; and the comparison assumes both properties command the same rent, which is exactly what a buyer should check rather than assume.

Why does the model make the renter invest the deposit?

Because otherwise the comparison is rigged. The buyer converts cash into an asset; if the renter's identical cash simply disappears from the model, buying wins by construction and the exercise proves nothing. Giving the renter a 3 % net return is a deliberate, conservative choice: it is above a regulated savings account and well below a long-run equity return, and it is stated as an assumption rather than a fact. Raise it to 5 % and the French break-even moves out by several years; drop it to 1 % and buying wins almost immediately. Anyone quoting a rent-versus-buy break-even without saying what the renter does with the deposit is quoting half a calculation.

What happens to the break-even if house prices stop rising?

It roughly triples. On the French case, 2 % annual price growth gives a break-even in year 7; 1 % gives year 12; flat prices give year 20. That single input outweighs every tax discussed in this article put together, and it is the one input nobody can source, because it is a forecast rather than a rule. The honest way to use the model is therefore to run it at zero growth first and treat that as the floor: if buying still works when the house never gains a cent, the decision is robust. If it only works at 3 % growth, what you have is a bet on the market, not a housing decision.

Do these purchase costs include the estate agent?

No, and that omission is deliberate and consistent across all three countries. The figures cover taxes, notarial and legal fees and land-registry charges — the part that is set by law and that no negotiation touches. Estate-agent commission is negotiated, varies by market and is often already inside the advertised price, so folding a single national average into the same number would make the tax comparison meaningless. Where a country's market quotes a higher headline than the one here, the gap is usually the commission. The model charges 4 % on the way out for exactly that reason, and that 4 % is an assumption, not a sourced figure.