Every mortgage advisor has had this conversation. A client is offered a rate of 3.9%, they can plainly afford the payment that rate produces, and yet the lender comes back with a maximum loan tens of thousands below what the arithmetic suggests. Nothing has gone wrong. The lender simply never underwrote the loan at 3.9% — it underwrote it at 6%, or 6.5%, or whatever its internal assessment rate happens to be this quarter. That gap between the rate a borrower pays and the rate the lender tests is the mortgage stress test, and it is the single most misunderstood number in the affordability process. This guide explains how the assessment rate is actually built, why it differs from lender to lender and product to product, what it costs a client in borrowing power, and what levers you can pull when someone fails it.
Start from what a mortgage really is: a long-dated loan on a short-dated price. A borrower fixes their rate for two, five, sometimes ten years, but the loan itself runs for twenty-five or thirty. At the end of the fixed period the loan reverts — to a standard variable rate, a tracker, or whatever the market offers on refinance. The payment the borrower signed up for is therefore a temporary payment. The stress test asks a simple question about the payment that comes after it: if this loan repriced to a materially higher rate, would this household still be able to pay it out of the income they have?
That reframing matters, because it explains why the test is not punitive or arbitrary. It is not the lender doubting the client's income. It is the lender refusing to underwrite a payment that only exists for the first sixty months of a three-hundred-month commitment. The 2008 crisis was, in no small part, a portfolio of loans underwritten on introductory payments that borrowers could never sustain once the teaser expired. Nearly every major regulator responded the same way: assess the borrower against a higher rate than the one they are sold.
How the assessment rate is built
The number a lender actually stresses at is rarely a single published figure. It is usually the higher of several components:
Reversion rate plus a buffer. The lender takes the rate the loan would revert to at the end of the fixed period — its standard variable rate, or a tracker margin over a reference rate — and adds a buffer, commonly one to three percentage points.
An absolute floor. Many lenders (and some regulators) impose a minimum assessment rate regardless of where the market sits, so that a period of very cheap money cannot inflate everyone's borrowing power at once.
A contract-rate buffer. Some regimes require the assessment rate to sit a fixed margin above the actual product rate, which keeps the test binding even for a borrower on an unusually cheap deal.
Whichever of those is highest becomes the rate that sizes the loan. The practical consequence is that the assessment rate moves on its own schedule, not the market's. When headline rates drop, a lender sitting on a floor may not pass any of that reduction into borrowing power at all. When headline rates rise, the reversion rate rises with them and the test tightens twice over. This is why, as covered in our guide to how interest rates affect borrowing power, client maximums often lag the news by a quarter or more in both directions.
Why the buffer is not the same for every product
A stress test guards against repricing risk, so it is logical to scale it to how soon repricing can happen. Most lenders do exactly that. A two-year fix reprices in two years, so the full buffer applies. A ten-year fix reprices after a decade, by which time incomes have moved and capital has been repaid, so many lenders apply a reduced buffer — or assess at the contract rate itself. Variable and tracker products are stressed hardest, because the payment can move next month.
This is not a technicality; it is a live advisory lever. Two products at near-identical headline rates can produce maximum loans that differ by a five-figure sum purely because of how their fix lengths are stressed. An advisor who knows which lenders relax the buffer on long fixes can sometimes rescue a purchase without the client changing a single thing about their finances. Buy-to-let sits in its own world again, assessed on a rental coverage ratio at a stressed rate rather than on personal income — the same principle, a different denominator.
A worked example
Take an illustrative household whose affordability assessment concludes they can sustain a mortgage payment of €1,750 per month on a 30-year repayment basis. Suppose the product rate on offer is 4.0%, the lender's reversion rate is 5.5%, and its stress buffer is 1.0 percentage point. The assessment rate is therefore the higher of the contract rate and 5.5% + 1.0% = 6.5%.
At the contract rate of 4.0%, €1,750 a month over 30 years supports a loan of roughly €366,000.
At the assessment rate of 6.5%, the same €1,750 supports only about €277,000.
The stress test has removed roughly €89,000 — about 24% — from the client's ceiling, and the client will still pay the 4.0% payment on whatever they borrow. Now change one variable: a lender that applies no buffer to a ten-year fix would assess at 5.5% and support about €308,000, over €30,000 more, on a client whose income, deposit and debts are identical. (All figures are illustrative and rounded, to show the mechanics; real outputs depend on the lender's exact model, term, reversion rate and buffer.)
What to do when a client fails
Failing the stress test almost never means "no mortgage." It means the loan that passes is smaller than the loan requested, and there are more levers than most clients realise:
Lengthen the term. Lower payment per euro borrowed, so more passes the test. The cost is total interest paid — a trade, not a free win, and worth spelling out.
Clear a debt — or a limit. Many lenders deduct an assumed payment from an unused credit-card limit, not just the balance. Closing a dormant card can move the number. The client's debt-to-income position is often the cheapest thing to fix.
Change the product. A longer fix at a lender with a softer buffer, as above.
Raise the deposit. A lower loan-to-value ratio usually unlocks a cheaper rate band, which lowers the assessed payment and can move the stressed maximum too.
Change lender. Assessment rates are not standardised. The same file genuinely does pass at one lender and fail at another.
Why agents should care as much as advisors
A buyers' agent who does not understand the stress test will keep putting clients into offers they cannot fund. The gap is invisible: the client quotes a budget from an online calculator that used the contract rate, the agent believes it, and the shortfall surfaces at underwriting with a deal already agreed. The habit that prevents this is simple — ask what rate the client's figure was calculated at, and if the answer is the rate they were quoted rather than the rate they were assessed at, treat the budget as provisional until the lender confirms it.
There is a second, quieter exposure. A loan is capped at the lower of the purchase price and the appraised value, so a stressed-thin buyer has no cushion if the valuation comes in under the agreed price. Checking the price against comparable sales before the offer is locked is the cheapest insurance available. This is where Biedradar earns its place in an advisor's or agent's workflow: enter an address and it returns comparable sales, a valuation range and market signals, then produces a branded property analysis report in minutes — so you can see whether a home is likely to value up before your client commits.
Framing it for the client
Clients hear "stress test" and think they are being distrusted. The most useful reframe an advisor can offer is that the test is running the same calculation the client should be running for themselves: what happens to this household when the fix ends? A borrower who takes the full stressed maximum is, by construction, someone whose budget survives the reversion with nothing to spare. A borrower who takes less is buying optionality. Showing a client both numbers — the loan at the contract rate and the loan at the stressed rate — turns a bureaucratic hurdle into the most honest affordability conversation they will have. If they can sit comfortably at the stressed payment, the mortgage is genuinely affordable. If they cannot, the lender has just told them something true, and for free. Pair that with a valuation you have actually checked, using a property analysis report the client can read, and you have advised on both halves of the risk: what they pay, and what they own.
Frequently asked questions
What is a mortgage stress test?
A mortgage stress test is a lender's check that a borrower could still afford their payments if the interest rate were meaningfully higher than the rate they are being offered. Instead of sizing the loan on the actual product rate, the lender recalculates affordability at a higher 'assessment rate' — often the reversion rate plus a buffer of one to three percentage points. The loan is capped at whatever passes that harder test.
Why do lenders test at a higher interest rate?
Because most mortgages reprice. A fixed period ends, the loan reverts to a variable or standard rate, and the payment can jump. Regulators after the 2008 crisis pushed lenders to underwrite for that future payment, not just the introductory one, so borrowers are not set up to fail in year three. It also protects the lender: stressed borrowers default more, and a loan book underwritten at today's cheap rate is fragile when rates move.
How much does the stress test reduce how much you can borrow?
It depends on the size of the buffer and the loan term, but a two-percentage-point stress buffer typically reduces the maximum loan by somewhere in the region of 15-20% on a standard repayment mortgage. The effect compounds with term: shorter terms are already payment-heavy, so the same buffer bites harder. The only reliable way to know is to run the client's figures at both rates.
Can you fail a mortgage stress test but still get a mortgage?
Usually yes, just for a smaller loan. Failing the stress test rarely means outright rejection — it means the amount that passes is lower than the amount requested. Borrowers can respond by lengthening the term, clearing a debt or credit-card limit, adding an applicant, raising the deposit, or choosing a longer fixed period at a lender that stresses long fixes more leniently. Different lenders use different assessment rates, so the same client can pass at one and fail at another.
Does a longer fixed rate mean a smaller stress test?
Often. Many lenders apply a reduced buffer — or none at all — to mortgages fixed for five years or longer, on the logic that the payment is locked and the repricing risk they are guarding against does not arrive for years. That can lift a client's maximum loan noticeably. It is one of the few levers that changes the answer without changing anything about the borrower, which makes it worth checking before you conclude a client cannot afford a home.
Is the stress test the same in every country?
No. The mechanics are broadly the same — assess at a higher rate than the one offered — but the specifics vary a lot. Some regimes prescribe a minimum buffer above the contract rate, some set an absolute floor rate, some let lenders build their own assessment rate within supervisory guidance, and some rely on income multiples with an affordability check underneath. Always work from the individual lender's current criteria rather than a rule of thumb.
