From November 1, people borrowing to buy a residential investment property in Auckland will need at least a 30 per cent deposit, the Reserve Bank has decreed.
This represents a tightening of the screws for this subset of borrowers within the regime of regulating loan-to-value ratios (LVRs), which the bank introduced two years ago.
Why is it doing this? The short answer is in order to douse some demand in Auckland's overheated property market.
"Demand" in this context is not physical demand - people needing roofs over their heads - but rather buyers in the market for houses and apartments. And especially in Auckland, those are two very different (albeit related) things.
Reducing demand by raising interest rates is not available as an option. In governor Graeme Wheeler's words, that is the last thing the economy needs right now.
The case for targeting Auckland is straightforward. The median house price in the city rose 25 per cent in the year to September.
In the rest of the country the rise was 8 per cent and that was boosted by the "halo effect" - the spillover of demand from Auckland to neighbouring regions.
Deputy governor Grant Spencer in a speech two months ago pointed to how stretched Auckland house prices have become.
The ratio of house prices to incomes has climbed from around six times in 2012 to nine, he said, a level seen in some of the world's most expensive cities like London, San Francisco and Sydney.
At the same time, the debt-to-income ratio for the representative Auckland buyer has climbed from around 4.5 times to seven.
But why pick on investors in particular? The Reserve Bank cites two things.
First, investors represent a growing share of purchases of Auckland properties and also of mortgage debt.
And they are, the bank argues, riskier borrowers - more likely to default in the event of a house price crash than owner-occupiers, who would lose their homes as well as their shirts if they did.
A third reason (which the bank does not advance) is that investors have a disproportionate effect on house prices.
The would-be landlord is often the marginal buyer, the one who would leave the market if the price were any higher.
He is the guy the owner-occupier has to outbid and if the price he is willing to pay is inflated by his expectations of enjoying all the benefits of leverage in a rising market until he is ready to collect his untaxed capital gain, then that is the price the would-be owner-occupier will have to better.
Research published by the Reserve Bank this month shows that purchases by investors rose from around a third of the total before the introduction of LVR restrictions two years ago, to just under 40 per cent by July this year.
So whose share has been correspondingly shrinking? Initially it was first home buyers, as you might expect.
But since mid-2014 their share has been trending higher and is now back where it was before the announcement of the LVR curbs, which hit them hardest.
Instead it is movers - owner-occupiers selling one home and buying another - whose share of property purchases has been declining.
And even though cash buyers are not affected by borrowing curbs, their share of the Auckland market has in fact declined (from around 23 to 20 per cent) since LVR restrictions were introduced.
Trawling through the data also reveals it is smaller investors (those who own two to four properties) who have been driving the increase in investor activity, as opposed to those with larger portfolios.
"This sort of profile - smaller investors who are reliant on credit - suggests that the new LVR restrictions on Auckland residential investors are likely to have an impact on overall demand," Spencer said.
"This is especially so when we consider that over half of investor lending is currently being written at high LVRs, that is, LVRs of over 70 per cent."
How much impact is a moot point and the bank has been careful not to paint this policy as any kind of panacea, emphasising instead the importance of a supply-side response.
But that does not mean it is not worth doing. This issue is bedevilled by people mounting arguments along the lines "That's not the problem.
This is the problem," or "That won't fix it. We need to do this other thing," as though we were only allowed one problem and one remedy.
The Reserve Bank's contention that lending to landlords is riskier than lending to owner-occupiers has not gone unchallenged, however, for example by Ian Harrison of Tailrisk Economics.
The Treasury sees some merit in some of the criticisms of the evidential basis for the policy, even though its bottom-line conclusion is to support it, "given the consequences of doing too little too late".
The level of risk to systemically important banks' solvency is not the only risk that is relevant here. A bank failure and the associated need for a costly bailout is not the only potential danger to be guarded against.
A credit crunch is another.
It is not just a matter of what nasty things the real economy might do to bank balance sheets, but what the banks in turn would do to the economy in response to such a shock - a feedback effect.
Last year the four largest New Zealand banks were subjected to a stress test which postulated a 40 per cent drop in house prices combined with a severe recession and rising unemployment.
"While this test suggested that banks would maintain capital ratios above minimum requirements, [they] reported that they would need to cut credit exposures by around 10 per cent, the equivalent of around $30 billion, in order to restore capital buffers," Spencer said.
Such a contraction in credit would only deepen the recession, leading to further falls in asset prices and even larger losses for the banks.
"A key goal of macro-prudential policy is to ensure that the banking system has sufficient resilience to avoid such contractionary behaviour in a downturn," Spencer said.