Skip to content
exactly.loansexactly.loans

Insights · Regulator

The 3% buffer designed for a different rate environment.

The vast majority of standard home loan applications are assessed against a rate three percentage points higher than the one you'll actually pay — a handful of lenders run a lower buffer on specific product lines, but for most borrowers 3% is the reality. That buffer is the single biggest factor compressing borrowing capacity, and the debate about whether it should change is now in earnest.

Reviewed · Adam King — 30 years in finance, Sunshine Coast

What the buffer actually is

APRA — the Australian Prudential Regulation Authority — sets the prudential standards that govern how authorised deposit-taking institutions (ADIs) assess residential mortgage applications. The headline guidance sits in Prudential Standard APS 220, with APG 223 providing residential mortgage lending guidance: APRA expects ADIs to apply a serviceability buffer of at least 3.0 percentage points over the loan rate when assessing capacity. It's supervisory guidance the regulator expects lenders to follow, not a statutory rule. Where it gets misunderstood is the assumption that every lender applies it identically. They don't. The 3.0% is a minimum the major banks apply to effectively all of their lending, and for the vast majority of applications across the market — the great bulk of scenarios — a 3% buffer is the reality you'll be assessed against. But the figure each lender uses, and the situations they'll flex it in, vary more than most borrowers realise. A handful of lenders run a consistently lower buffer on certain product lines, sitting at 1–2% rather than 3%, and a small number offer near-0% buffer assessments in specific, well-defined cases. That variation is where a broker earns their keep. In practice, the 3% means a borrower applying for a loan at 5.85% is assessed against a notional 8.85% rate to determine whether they can afford the loan. The same dollar repayment, modelled at the higher rate, has to fit inside the household's after-tax income net of declared expenses. If it doesn't, the loan is declined — regardless of whether the borrower would have been able to afford it at the actual contracted rate. The buffer was set at 2.5% before October 2021. It was lifted to 3.0% as the cash rate began its first move upward in years and the regulator wanted to ensure borrowers approved during the easing cycle would still be able to service if rates returned to long-run averages. The 3% has stayed put through the rate cycle since.

Buffer mechanics — what it does to capacity

  • Buffer over actual rate

    3.0%

    APG 223 minimum since October 2021

  • Capacity reduction

    ~20%

    Indicative impact vs. unbuffered assessment

  • Notional rate today

    ~8.85%

    If actual rate is 5.85% (indicative — varies by file)

  • Capacity example

    $540K

    Where unbuffered would model ~$650K for the same income

Why the buffer exists

The buffer is essentially insurance against a future rate environment that doesn't yet exist. It came out of the post-Royal-Commission tightening — the Financial Services Royal Commission of 2018 surfaced material weaknesses in how lenders had been assessing borrower capacity in the boom years, and the regulatory response was to push the floor of the buffer up and tighten the rules around how borrower expenses are verified. The argument for the buffer is straightforward. Variable rate borrowers are exposed to rate rises across the life of the loan. A loan written when rates are low and serviced at the low rate looks fine on day one. The same loan, three years later, after a 2.5% rise, can become unserviceable if it was written without a buffer. The buffer protects the borrower from being approved for a loan they wouldn't be able to afford if rates moved against them. The argument against — or for easing — is that the buffer is now binding capacity in a way that pushes borrowers either toward higher LVRs (more leverage, more risk), toward longer loan terms, or out of the market entirely. First-home buyers carry most of the binding. The buffer doesn't reduce the value of the home they're trying to buy; it reduces the amount they can borrow against the same income. The gap is filled by either a bigger deposit, a smaller property, or a different timeline. One mechanic that catches people out: the buffer isn't only applied to the new loan you're asking for. The lender adds it on top of your existing debts too — every other mortgage, and often other commitments — is re-assessed at the buffered rate, not at the rate you're actually paying. That's why capacity compounds against you as you accumulate property. A first or second purchase usually still clears; it's the third or fourth property where the buffer bites hardest, because the buffered cost of everything you already hold is stacked into the new assessment before the new loan is even counted.

Indicative capacity impact at different buffers

ScenarioBufferAssessed rateIndicative capacity
Current rule3.0%8.85%$540,000
Prior rule (pre-Oct 21)2.5%8.35%$575,000
Proposed easing scenario2.5%8.35%$575,000
Hypothetical 2.0%2.0%7.85%$610,000
No buffer0%5.85%$650,000

Indicative borrowing capacity for a single applicant on $130K gross income, no dependants, no other debt, owner-occupied P&I, 30-year term, at an actual rate of 5.85%. All figures indicative and subject to confirmation against the specific lender's serviceability model.

Practical implication

The buffer hits first-home buyers hardest.

Established buyers refinancing typically already own their home — the buffer doesn't lock them out of a market they're already in. First-home buyers see the full effect: the assessed repayment is what determines how much house they can buy, and the buffer trims roughly 20% off that number relative to an unbuffered assessment. Indicative — varies by file.

The live debate — what's actually being discussed

The debate about easing the buffer has been ongoing through 2024 and 2025, and is live again in 2026. The Council of Financial Regulators — APRA, the RBA, the Australian Securities and Investments Commission (ASIC) and Treasury — meets quarterly and has discussed the appropriate calibration multiple times. APRA's public position has been that the buffer should remain at 3.0% until the rate environment is more clearly settled. The argument for easing is that a buffer calibrated for a rising-rate shock can bind capacity hard when household incomes are already stressed. The argument against is that an easing translates fairly directly into more credit issuance, which can flow through to house prices. After the RBA's May 2026 increase, the rate environment is not clearly settled, which makes a near-term broad easing less certain than it looked earlier in the year. My own read is that any easing is more likely to be cautious and targeted than a blanket reset, possibly with carve-outs for specific borrower cohorts such as refinancers. That is not a forecast — it's a read of the public commentary from the regulators and the direction of the rate cycle. APRA's APS 220 standard, APG 223 guidance and quarterly statistical reports are all at apra.gov.au if you want to follow the technical commentary.

What the buffer means for your file, in practice

Several borrower types feel the buffer differently. Worth knowing where you sit.

  • First-home buyer with a 10–15% deposit: capacity is the binding constraint. Reducing other commitments (credit cards, BNPL, declared HEM expenses) is where the marginal capacity gain comes from.
  • Refinancer with existing equity: capacity usually matters less, because you're refinancing an existing balance, not borrowing more. And if the buffer is the very thing blocking a like-for-like refinance to a cheaper rate, some lenders will assess that improving-position move at a reduced or 0% buffer — exactly the case the standard 3% buffer shouldn't be trapping you in.
  • Investor adding to a portfolio: serviceability cascades, because the buffer is applied to your existing loans too, not just the new one. The third or fourth property is where it bites hardest. The structural lever is moving existing loans to interest-only with a lender that assesses the actual repayment rather than a buffered P&I figure.
  • Self-employed applicant: the buffer applies the same way, but the income side is harder to evidence. Two years of clean tax returns plus current BAS data is the standard documentation set.
  • Bridging or refinance with equity release: a partial release is sometimes possible at higher LVRs without the buffer biting if structured as a top-up rather than a full refinance. Depends on the lender and the file.

Workarounds — what is and isn't legitimate

The 3.0% is a minimum, not a universal setting. There's a common myth that some lenders quietly run a higher buffer — 3.5% on certain product lines — to be conservative. They don't; the majors sit at 3% and the variation is in the other direction. A handful of lenders run a consistently lower buffer (1–2%) on specific products, and a small number assess at or near 0% in defined cases. Separately, a number of non-bank lenders sit outside the APRA-regulated ADI universe — they fall under ASIC's responsible lending regime rather than APRA's prudential standards — so the 3.0% buffer doesn't bind them in the same way, though they still apply their own internal buffers because the loans need to be securitisable to be funded. The most useful lower-buffer case is a like-for-like refinance. If you're moving an existing loan to a lower rate and lower repayment — improving your position, not borrowing more — some lenders will assess at a reduced buffer (around 1%) or, in clean cases, 0%, provided your repayment history is clean and the new loan is still demonstrably affordable. This isn't automatic and APRA expects lenders to stay prudent, but it's real, and it's the lever that frees borrowers who are effectively locked into their current lender by the 3% buffer — unable to refinance to a cheaper rate purely because the buffered assessment says they can't 'afford' the loan they're already paying. If you've been told no on a refinance because of the buffer, this is the first thing worth checking. For investors, the lever is structural. Moving existing loans to interest-only, with a lender that assesses on the actual interest-only repayment rather than a buffered principal-and-interest figure, can lift capacity meaningfully. Beyond that, the income side does the rest — accurately declared overtime or bonus income, the right tax-return documentation, a partner's income added correctly, or rental income shaded at the right rate by the right lender. We map this lender by lender on every file.

From the practice

Capacity isn't a single number. It's a function of which lender reads which line.

The same applicant, same income, same debts, can model out at a $90K difference in borrowing capacity across our panel. The buffer is identical. What differs is shading on overtime, treatment of rental income, HEM benchmarks and dependant calculations. Knowing which lender treats which line favourably for your shape of income is the broker's edge.

Questions you might have

The honest answers.

Real numbers · honest answers

How much can you *actually* borrow today?

Twenty-minute call. We'll model your file against three lenders, factor in the 3% buffer correctly, and give you a real capacity figure — not a website estimate.

Keep reading

General information only — not personal credit advice. Figures are indicative and subject to confirmation against current lender pricing and policy.