Where’s the evidence? asks expert critic of lending curbs.

The Reserve Bank's plans to curb residential property investors' access to credit are based on threadbare and unconvincing analysis and could do more harm than good.

At least that is the charge levelled by a former senior and long-serving economist at the bank, Michael Reddell, in a trenchant critique of its loan-to-value ratio (LVR) policy delivered last week to the Law and Economics Association.

Reddell's view is evidently shared at the Treasury. Advice released under the Official Information Act reveals a Treasury sceptical of the case for targeting high-LVR investors in Auckland and worried about unintended consequences. Those possible consequences include a heightened risk that investors will borrow instead from non-bank sources, or leverage up property outside Auckland, or demand will shift to cashed-up (code for foreign) buyers.

The Reserve Bank sees the rapid rise in Auckland house prices, from levels already high relative to household incomes, as a threat to the soundness of the financial system.


But Reddell points to the results of the stress tests of New Zealand banks last year, a few months after the original LVR curbs were imposed.

House prices were assumed to fall by 40 per cent nationwide and 50 per cent in Auckland, while the unemployment rate rose to 13 per cent. That is a higher unemployment rate than New Zealand has seen since the Great Depression and would require a greater increase in unemployment than any country with a floating exchange rate has seen since World War II.

Stress tests should be tough, Reddell says, and these were but the banks came through unscathed.

"In the most recent proposal, the [Reserve] Bank has concluded that New Zealand banks are so badly run, and New Zealand borrowers apparently so reckless, that not a ... cent, in huge balance sheets, can be lent safely to rental service providers in Auckland on LVRs in excess of 70 per cent. It differentiates by region and by type of borrower, not just by collateral or income. Where, we might ask, is the evidence that such lending is so unsafe that the coercive powers of the state should be exercised to ban it altogether?"

During the first five months of this year, of $21.7 billion of new residential mortgage lending, 39 per cent went to investors (and just 12 per cent to first home buyers). Just over half the investor borrowing was at LVRs above 70 per cent.

The Reserve Bank argues that loans to investors are more risky than to owner-occupiers, because in the event of a major fall in house prices they are more likely to default, as it is not the roof over their own heads that is at risk.

And it contends that the flow of new lending, not the change in the total stock of home loans on banks' books, is what matters for macro-prudential purposes.

"Grant [Spencer, deputy governor] says we have all this new debt and I say well, what was the ratio of new debt to old debt in 2003? We honestly don't know. More importantly, there is a lot of international literature on the incidence of crisis and all of it says that it is rapid growth in the stock [of debt] that is the vulnerability," Reddell said. In the year ended May, bank lending to all borrowers grew 5.8 per cent, and household mortgages 5.5 per cent - probably a bit faster than nominal GDP, but nothing like the double-digit rates prevailing during the mid-2000s boom.


And it is the quality of the loans that matters, Reddell argues. Banks very rarely fail from vanilla housing loans. "It is typically commercial property, and particularly property development loans, that bring down banks."

It is not enough, Reddell says, to simply point to other countries which have had a crash, without inquiring into the differences between their situation and ours.

How important in the US case was political pressure on Fannie Mae and Freddie Mac to encourage more low-income people into home ownership?

How much of Ireland's crash can be laid at the door of its outsourcing monetary policy to the European Central Bank, so it had German interest rates when it should have had rates more like ours? Ditto for Spain.

These are not rhetorical questions. They are questions the Reserve Bank should be able to answer before intervening on the scale it has.

Governor Graeme Wheeler was living in the United States when its property market crashed. He has seen, up close, what it does to an economy when one home loan in four is under water. He does not want to see that here on his watch.

A financial crisis is a low probability but high-cost event, as the Treasury says. If you focus on the low probability, like Reddell, the conclusion is that the bank should pull its head in. If you focus on the high potential cost, like Wheeler, you would want to do whatever you can to slow the growth in house prices and buy time to get more built and for the net inflow of migrants to return to more normal levels.