The Two Numbers That Set Your Borrowing Capacity

The ratio gives you a payment, not a house

Every borrowing-capacity rule in every country starts the same way: a share of income, minus what you already owe each month. That is the whole of step one, and it produces a monthly payment. It knows nothing about house prices, nothing about the neighbourhood you want and nothing about how much you actually need. It is a solvency test dressed up as a shopping budget, and the reason it feels arbitrary is that it is arbitrary — the percentage is a supervisory or commercial choice, not a fact about your life.

The American version applies the share to gross income, which is the single most consequential detail in the whole exercise. On $7,500 a month gross and $900 of existing debt, a 43 percent back-end ratio leaves $2,325 for housing — but $2,325 out of a take-home figure that might be $5,600 is 41.5 percent of the money that actually arrives. The continental European rules apply the share to net income precisely to avoid that gap, which is why a French 35 percent and an American 43 percent are much closer than they look. Whenever someone quotes you a ratio, the first question is which income it is a share of; the second is whether the escrowed property tax and insurance sit inside it.

The rate turns the payment into a principal

Step two is the annuity formula run backwards. A payment of P for n months at monthly rate r supports a principal of P × (1 − (1 + r)^−n) ÷ r. That fraction is a pure multiplier, and it is worth memorising the size of it: at 6.5 percent over 30 years it is 158.2, so every $100 of monthly payment buys $15,820 of loan. At 5.5 percent it is 176.1, at 7.5 percent 143.0. The whole of the market's effect on you sits in the difference between those three numbers, and none of it has anything to do with your job, your savings or your credit file.

That multiplier is also the reason lengthening the term feels like free money and is not. Going from 15 to 30 years at 6.5 percent raises the multiplier from 114.8 to 158.2 — an extra 38 percent of principal for the same payment. But the interest paid rises far faster than the principal borrowed, because you are renting the money for twice as long. The honest way to frame the choice is not more house versus less house; it is a stated amount of extra principal against a stated amount of extra interest, and the calculator will print both if you ask it the same question twice.

Existing debt is the most expensive thing in the file

Because the ratio subtracts your existing instalments before anything else happens, every monthly obligation you carry is multiplied by the same annuity factor on its way out. At 6.5 percent over 30 years that factor is 158.2, so the $280 student-loan payment in the worked example is not costing you $280; it is costing you $44,299 of borrowing power. A $520 car payment costs $82,270. This is the single most actionable fact in mortgage arithmetic, and it is why paying off a small balance in the months before an application can be worth many times its face value.

There is a trap in the other direction, though. Emptying your savings to clear a debt raises the ratio result and lowers the deposit, and the deposit is a separate constraint that the ratio never sees. If your file is short of cash rather than short of income, clearing debt makes it worse. The correct order is: work out which of the three constraints — the ratio, the rate-driven multiplier, and the cash you must put down — is the one actually binding, and only then decide what to do with a spare thousand.

Insurance, tax and escrow: what sits inside the ratio

The ratio is applied to a payment, but which payment? In the United States the underwritten figure is PITI — principal, interest, taxes and insurance — plus homeowners' association dues where they exist, because taxes and insurance are escrowed and collected with the loan payment. That is why the worked example took $520 out before converting to principal: property tax at roughly 1.1 percent of a $360,000 house is $330 a month, and hazard insurance is another $190. Leave them out and the capacity number is overstated by the multiplier times $520, which is $82,264.

The general lesson is that a capacity figure is only comparable to another capacity figure if both were built from the same definition of the payment. Two lenders quoting you very different maximum loans are usually not disagreeing about you; they are disagreeing about what belongs in the numerator. Ask each of them, in writing, which items they included, and re-run the calculation yourself with the fuller of the two definitions. The answer that survives that test is the one your budget will actually meet.

Why the same borrower gets very different answers in six countries

France is the only one of the six that has turned the ratio into law. The Haut Conseil de stabilité financière capped the taux d'effort at 35 percent and the maturity at 25 years, and made the decision binding on lenders from 1 January 2022 — with a flexibility margin of up to 20 percent of quarterly production so that judgement survives at the edges. Germany has the opposite architecture: no ratio in the statute at all, only a duty under § 505a BGB to assess creditworthiness, with the real discipline supplied by the loan-to-value and by purchase costs that cannot be borrowed.

The United States is the country where the received wisdom is most out of date. For years the General Qualified Mortgage carried a hard 43 percent debt-to-income ceiling; the 2020 revision replaced it with a price-based test, so what the regulation now caps is the annual percentage rate relative to the average prime offer rate — 2.25 points for a large first lien — and not your ratio at all. Debt-to-income limits still govern your file, but they arrive through agency and investor underwriting standards, which means they can differ between two lenders and can change without a rulemaking. If someone tells you 43 percent is the law, they are quoting a rule that no longer exists.

What to do with the number once you have it

Treat the capacity figure as a ceiling produced by someone else's risk appetite, not as a target. The ratio was calibrated so that a portfolio of borrowers does not default too often; it says nothing about whether you can also fund a pension, replace a boiler and survive a year of one income. Run the calculation a second time at a payment you would still be comfortable with if your household income fell by a quarter, and treat the gap between the two answers as the size of the buffer you are choosing not to have.

Then make the three constraints explicit on one line each, because only one of them is binding and it is usually not the one people worry about. Constraint one: the payment your income supports after existing debt. Constraint two: the principal that payment supports at today's rate over the term you will actually accept. Constraint three: the cash you must produce at completion — deposit plus transfer taxes, notary or title, and fees — which no ratio ever tests. Whichever of the three gives the smallest house is your answer, and the other two are noise.

Frequently asked questions

Is the debt ratio calculated on gross or net income?

It depends on the country, and the difference is roughly the whole of income tax and payroll contributions. American underwriting uses gross monthly income, which is why its ratios look high: a 43 percent back-end ratio on gross can be over 55 percent of take-home for a high earner in a high-tax state. French, Spanish, Italian and Portuguese practice uses net income, and the French 35 percent taux d'effort is defined on the income actually received. German banks do not use a ratio at all; they build a household budget from net income and flat-rate living costs. So a 35 and a 43 are not two positions on the same scale, and comparing them without converting is the commonest mistake in this subject.

Does a bigger deposit increase my borrowing capacity?

Not directly, and that catches people out. Capacity is a function of income, existing debt, rate and term; the deposit does not appear in it. What a bigger deposit buys is a bigger purchase price for the same loan, a lower loan-to-value that may unlock a better rate, and in some markets the removal of a mortgage-insurance premium that was eating into the payment. Only the last two feed back into capacity, and they do it indirectly through the rate and through the payment. If you are told that saving more will let you borrow more, ask which of those three channels is meant.

Why does my capacity fall so much when rates rise by one point?

Because the multiplier that converts a payment into a principal is a convex function of the rate, and the effect compounds over a long term. At 6.5 percent over 30 years each dollar of monthly payment supports $158.20 of loan; at 7.5 percent, $143.02. That is a 9.6 percent fall in capacity from a one-point move, so a $1,805 payment that carried $285,571 now carries $258,147. The longer the term, the more brutal it is, because more of the payment is interest and interest is what the rate is charging for. This is also why capacity recovers faster than prices when rates come down, and why the calculator is worth re-running the week you actually apply rather than the month you start looking.

Can a lender lend me more than the ratio allows?

In France, yes, but only within a quota. The Haut Conseil de stabilité financière allows lenders to breach the 35 percent and 25-year criteria for up to 20 percent of their quarterly production, of which at least 80 percent must go to people buying their main home and at least 30 percent of the margin to first-time buyers, leaving a small freely usable remainder. That is the mechanism behind an exception: it is rationed, it is audited, and the strongest files win it. In the United States a lender can exceed any ratio it likes provided the loan meets the ability-to-repay rule, but the loan may then fall outside the standards a secondary-market buyer will accept, which is why the answer often arrives as a higher rate rather than a refusal.

Should I borrow the maximum I am offered?

The maximum is calibrated on the probability that a portfolio of borrowers defaults, not on the probability that your particular household has a bad year. Three questions decide it, and none of them appears in the ratio. Would the payment still be payable on one income for twelve months? Does it leave room for the maintenance a property actually needs, which is commonly budgeted at around one percent of the value a year? And does it leave a retirement contribution intact, because twenty-five years of not saving is a much larger number than the extra rooms are worth? If the answer to any of those is no, the binding constraint was never the bank's.