The Reserve Bank has a persuasive case for being given the power to impose debt-to-income (DTI) curbs on residential mortgage lending.
It is certainly more compelling than the stock arguments raised against such a move: that it is too hard on would-be first-home buyers and is regulatory over-reach - nanny state stuff.
Both the bank's May Financial Stability Report and its consultation document on DTIs, released last week, remind us how brittle the status quo is.
Measures like the ratio of household debt to incomes, house prices to incomes and house prices to rents are all very high by historic - and international - standards.
Even with the economy growing at an above-trend rate and employment growth strong, real income growth is sluggish. In the year to March, average weekly earnings for wage and salary earners did not increase at all when adjusted for consumer price inflation (never mind house price inflation).
While house price inflation nationwide has flattened off over the past six months, over the past year it has still outpaced wage inflation. Growth in the stock of mortgage debt has slowed too, but still outstrips growth in households' collective take-home pay.
Can we be confident the past six months are a better guide to the future than the past year, or four years? The Reserve Bank is not.
And the rise in house prices has been accompanied by a declining rate of home ownership, so the increase in debt is more concentrated among a smaller proportion of the population.
Another dimension to the risk is that New Zealanders are enthusiastic borrowers. Savers? Not so much.
About a quarter of the banks' funding is imported and lately the gap between deposit growth and lending growth has been widening.
So the risk to retail interest rates is not just - or even mainly - about what happens in this country. The danger is that some international shock drives up interest rates in the offshore markets which New Zealand banks rely on for funding.
The Reserve Bank's Financial Stability Report says that when it looked at recent borrowers (those taking out loans in the six months to March 2017), 19 per cent had a debt-to-income ratio above five - a possible threshhold for DTI curbs. Among all borrowers, 6 per cent had a DTI over five.
If mortgage rates rose to 7 per cent (just about the average two-year rate over the past 10 years) the bank estimates that 4 per cent of all borrowers and 5 per cent of recent borrowers would not have enough income left to cover their essential expenses.
"At mortgage rates around current levels, debt servicing consumes around 40 per cent of the before-tax income of a typical high-DTI owner-occupier, rising to around 50 per cent if mortgage rates rise to 7 per cent," the consultation document says.
"These debt servicing ratios leave very limited financial resources for expenditure on other items, increasing the risk of a large cutback in expenditure, forced sale, or loan default.
"In a scenario where mortgage rates rise to 7 per cent, a typical high-DTI owner-occupier is estimated to have residual income (after tax and mortgage payments) that is only just above our estimate of essential expenditure. If mortgage rates rise to 8 or 9 per cent, a typical high-DTI first-home buyer would have residual income below essential expenditure."
Its assumed essential expenditure for high-DTI households is about $23,000, based on Statistics NZ's household economic survey.
So we are talking about tens of thousands of households who are lying pretty low in the water, liable to be swamped by either a hit to employment and incomes, or by a rise in interest rates, to levels seen as recently as nine years ago.
Central to the Reserve Bank's argument is that this is not just a risk for the people involved, or their banks. The loan-to-value ratio (LVR) restrictions already in place are primarily designed to bolster the resilience of banks' balance sheets in the event of mass defaults on mortgage debt.
And it is not really the Reserve Bank's business to save individual borrowers from themselves.
Rather, its concern is the spillover effects on the wider economy if too many people are struggling to pay the mortgage at once.
One obvious effect would be a drop in household spending as debt servicing costs rose.
Another would be a fall in house prices if a lot of people needed to sell at once.
The wealth effect works both ways. High leverage amplifies the rise in housing equity in a rising market, but also amplifies the destruction of equity when prices fall.
In Ireland, when its housing bubble burst, private consumption fell sharply as housing wealth shrank, with retail sales falling nearly 20 per cent, year on year, by early 2009.
The snowballing effect on the wider economy would also be increased by banks tightening credit.
The Reserve Bank's ability to counter this is less than it was. During the global financial crisis it cut the official cash rate by 5.75 percentage points. With the OCR now at 1.75 per cent, it could not do that again.
Another ill effect would be on labour mobility. Even if they are managing to pay the mortgage and cover essential expenses, people may be unable to move to a better paying job in another city if their home loan is under water (more than the property is worth).
As for the potential blow that DTI curbs would deal to potential first-home buyers, the bank has suggested measures that would moderate that effect.
There would be a "speed limit", as with LVRs. That is, banks would still be free to make some proportion of their lending above whatever DTI limit was imposed. They might use that capacity for first-home buyers.
The Reserve Bank has also suggested an exemption for borrowers who want to buy and occupy a relatively low-priced home. And an exemption for new builds is likely.
None of this is imminent. Even if the bank manages to persuade whoever is Minister of Finance after the election that this tool should be added to its kit, a further round of public consultation on a more detailed policy proposal would be undertaken.