Another OCR Rise Does Not Mean Another Housing Downturn

The Reserve Bank of New Zealand lifted the Official Cash Rate to 2.75% in September 2026, its second increase since tightening resumed in July.

The Reserve Bank has lifted the Official Cash Rate by 25 basis points to 2.75%. It is the second increase since tightening resumed in July, and every major bank economics team picked it.

For homeowners and property investors the question is straightforward. Does this send the housing market backwards again?

The answer is no. But not for the reason most commentary will give.

This is an inflation decision, not a housing decision

Annual CPI inflation reached 4.1% in the June quarter, up from 3.1% in March and well outside the Reserve Bank’s 1% to 3% band.

Most of that came from the fuel pumps. Petrol prices were 27.5% higher than a year earlier and diesel prices 71% higher. Stats NZ calculated that if petrol and diesel had not moved at all, annual inflation would have been 2.9%, comfortably inside the target band.

That does not make the inflation problem imaginary. Electricity rose 4.4% in the June quarter and construction costs rose 1.6%, the largest quarterly construction increase since late 2022. Council rates and insurance remain elevated. The Reserve Bank’s concern is that a temporary fuel shock becomes embedded in wider pricing and wage behaviour.

None of that has anything to do with house prices.

In the year to July, residential sales fell 10%, inventory rose 9.3% and new mortgage lending fell 13.1%. When the Reserve Bank left LVR settings unchanged this month, its Assistant Governor for Financial Stability described housing risks as currently contained.

The Reserve Bank is not lifting rates to cool the housing market. There is nothing there to cool.

What the July numbers actually show

There were 6,090 residential sales nationally in July, down 10% on July last year. Set against REINZ’s 35 years of records, that still sits close to the historical midpoint for the month.

The national median price was $760,000, down 0.7% on a year earlier. The REINZ House Price Index was down 0.4%.

Two numbers matter more than either of those.

The median Days to Sell was 50, two days longer than a year ago and the fifth slowest July since REINZ records began in 1992. Selling is not failing. It is taking longer.

And inventory was 9.3% higher than a year earlier while new listings were 0.5% lower. Stock is not accumulating because owners are rushing for the exit. It is accumulating because properties are clearing slowly.

That distinction is the whole argument. A market with rising listings and falling prices is a market under pressure. A market with flat listings, flat prices and a slow clearance rate is a market waiting.

The lending pullback is investor led

New residential mortgage commitments fell to $7.9 billion in July, the lowest monthly total since February and 13.1% below July last year. The number of borrowers taking out new commitments fell 7.2% over the year, and the average new loan value fell 6.3%.

Underneath the headline, the composition is more informative.

The investor share of new lending fell to 18.9%, down from 19.5% in June. The first home buyer share rose to 20.1%, up from 19.0% in June and slightly ahead of a year ago.

So the retreat is not general. Investors are stepping back while first home buyers are quietly stepping forward into a market with plenty of choice.

That is a demand and confidence story. It is not the profile of a market where rising rates are forcing owners to sell.

Affordability has already reset

The affordability equation has changed substantially since the peak of the last cycle.

The latest Staircase Affordability Index shows mortgage servicing on an 80% loan against the median home at around 39% of gross household income in Auckland, against an Auckland long term average near 44%. Nationally the figure is around 33%, against a long term average near 37%.

Housing has not become cheap. Plenty of households still cannot buy. But today’s market does not carry the servicing pressure that existed near the top of the previous cycle, and for most first home buyers the binding constraint is now the deposit rather than the repayment.

Borrowers should also note that the OCR is a lagging signal for them. Sharpest one year fixed rates at the major banks now sit near 4.95% to 4.99%, up from around 4.65% when the July increase landed. Two year rates are above 5.39%. Much of this tightening was priced before the decision.

The number that will actually be felt is the average rate on existing mortgages. It was 4.9% in March. The Reserve Bank expects it to reach about 5.3% within a year as borrowers refix. That is the real transmission channel, and it works slowly.

Why higher unemployment has not produced forced sellers

The obvious objection to all of this is the labour market. Unemployment reached 5.6% in the June quarter, the highest since 2015, and the Reserve Bank is tightening into it.

That objection deserves a proper answer, because job loss is the one thing that reliably turns a slow housing market into a distressed one. Households absorb higher repayments. They cannot absorb the loss of an income.

But look at how the June rate actually rose.

Employment increased by 13,000 over the quarter, to 2.91 million. The participation rate rose to 70.7%, the highest since early 2025. The unemployment rate went up because the labour force grew faster than the jobs available, not because employers were shedding staff.

That distinction decides everything for housing. Redundancy creates a forced seller. Population growth and rising participation do not. A person entering the labour force and not yet finding work is not a borrower refixing a mortgage they can no longer service.

This has also already been tested. Unemployment has been above 5% since late 2024 and has climbed through the correction and through the flat period that followed. If a rising jobless rate were going to generate distressed selling in this cycle, it would have shown up by now.

It has not. Mortgage arrears remain low. New listings in July were 0.5% lower than a year earlier, not higher. And when the Reserve Bank reviewed LVR settings this month it left them unchanged and described housing risks as currently contained.

The Reserve Bank’s own projection has unemployment sitting near current levels until the middle of 2027. That is a plateau at a high level, not a deterioration.

So the labour market explains a great deal about this market. It explains why buyers are patient, why they will not stretch, and why underutilisation at 13.8% keeps a lid on how fast demand can return. It does not explain seller distress, because there is very little seller distress to explain.

One part of this is genuinely untested. Through 2023 to 2025 unemployment rose while mortgage rates were falling, which cushioned households considerably. From here the jobless rate is high while the average rate on existing mortgages moves from 4.9% towards about 5.3%. That pairing has not been run before in this cycle. The number to watch is employment falling in absolute terms, not the unemployment rate drifting higher.

This is not one more rise. It is a cycle back to around 3%

The bigger point is being missed in most coverage. The question is not what one 25 point increase does. It is where the OCR settles.

Kiwibank describes this as the second of a likely three step move to 3%. ASB expects further increases in October and December to finish the year at 3.25%. Westpac sees roughly 3% by year end and warns against assuming the tightening runs on automatically after that. BNZ is the outlier, picking 25 points at every meeting to 4% by May 2027.

Notably, several of these economists are forecasting rises they do not think are warranted. Kiwibank has been explicit that it disagrees with the need to hike while agreeing that the Reserve Bank has signalled it.

For property, a peak near 3% to 3.25% is manageable. It leaves mortgage rates in the mid 5s, which is close to the servicing assumption most recent borrowers were tested against. A peak near 4% is a different conversation, and that is the scenario worth watching rather than tomorrow’s 25 points.

There is a timing wrinkle too. The election is on 7 November, which places the October review inside the campaign.

New Zealand no longer has one property cycle

The national figures also hide widening regional differences.

Auckland and Wellington remain weak. Both have now recorded 30 consecutive months of annual inventory growth.

Canterbury, Otago and Southland are considerably further advanced. Otago’s REINZ House Price Index reached a record high in July. Canterbury sits close to its previous peak, and all three regions recorded positive annual HPI growth.

Household expectations line up with that. Net house price expectations sit at 4% in Auckland and 6% across the rest of the North Island, against 23% in Canterbury and 18% across the rest of the South Island.

Nationally, a net 20% of New Zealanders still consider it a good time to buy, and buying sentiment is strongest in Auckland, one of the weakest markets in the country. The reason appears to be choice. Buyers have stock, time and no fear of missing out.

Meanwhile the net proportion expecting prices to rise has fallen from 30% at the start of the year to 9%. That is not a crash expectation. The largest group expects prices to hold roughly where they are.

Property cycles rarely turn at the same time in every region. Auckland being subdued does not mean the country is.

What would change the call

Vague watch lists are easy to write and impossible to be wrong about. Here are the specific readings that would move me off Recovery Consolidation, in order of importance.

• Employment falling in absolute terms, not simply the unemployment rate drifting up. Redundancies create forced sellers. Population growth does not.

• Mortgage arrears rising for three consecutive months rather than drifting sideways.

• New listings turning positive year on year while Days to Sell keeps lengthening. Today inventory is high because sales are slow. If it becomes high because owners are listing, the character of the market changes.

• A published OCR track pointing meaningfully above 3.25%.

And the confirmation running the other way is simple. Sales volumes rising while inventory falls. That is the signal that buyers are finally absorbing the stock already on the market, and it has not appeared yet.

ASB, for its part, expects the market to stay subdued through the rest of 2026 and improve through 2027, and will not rule out a small further fall in prices in the meantime.

That is a reasonable base case and not far from ours.

Where that leaves us

New Zealand has an unusual combination right now. Better affordability than several years ago. Subdued demand. Elevated inventory. Weak turnover. A labour market taking on more people than it can currently employ. And a Reserve Bank withdrawing stimulus before the housing recovery ever got going.

The correction has already done most of its work. Some regions have moved on from it entirely. Nationally, the next leg of the cycle is taking longer to emerge than it normally would.

Another 25 basis points will probably extend that wait. It does not, on its own, send the property cycle back to the beginning.