By Rodney Dickens*
The chart below is sourced from the Reserve Bank (RB) and shows household debt as a % of household disposable income, the average interest rate faced by households and the interest or debt servicing cost as a % of household disposable income.
Most of household debt is mortgage debt.
It shows that household debt as a % of income increased from 57.5% in the first quarter of 1991 to a peak of 153.4% in the second quarter of 2009, although it was down to 146.2% in the 2013 June quarter.
This massive increase in debt as a % of income may be a cause for concern about risks to the banking sector.
Debt servicing is where the rubber meets the road in terms of the risk of default by borrowers because banks can overlook people having negative equity as long as they are able to service the debt.
Debt servicing costs as a % of income was 8.8% in the 2013 June quarter compared to an average of 9.6% since 1991 and a peak in third quarter of 2008 of 14.4%. Low interest rates mean debt servicing isn't a problem at the moment.
But even if the average interest rate faced by households increased 2% over the next 2.5 years, which is in line with the upside the Reserve Bank predicted in the September Monetary Policy Statement for the 90-day bank bill yield, debt servicing as a % of income would only increase to around 11.3%.
This is somewhat above average but well below the peak level in 2008. This assumes that incomes grow by 4% per annum over the period and that debt as a % of income continues to fall a bit.
What happened after debt servicing costs peaked at over 14% of income in 2008 puts the issue in the right perspective.
The left chart shows the annual number of mortgagee sales increased from under 500 in 2007 to a peak of just over 3,000 2009/10.
But what matters in terms of risks to the banking sector is the number of mortgagee sales compared to the number of mortgages, which is shown in the right chart. Even at the peak, mortgagee sales were equal to only 0.2% of total bank mortgages.
This is infinitesimal.
It is of further use to put what happened in this period in the context of the state of the economy and especially the state of the labour market.
The biggest risk of mortgagee sales and debt default by mortgage borrowers that results in stress in the finance sector is where people lose jobs or businesses for whatever reason and aren't able to pay the interest bill.
Sickness will play a part in this, but I suspect the major risk is from job losses and business failures, with the two likely to be reasonably closely linked.
Fitting with this line of reasoning there is an extremely high correlation between the number of mortgagee sales and the unemployment rate (left chart). The 0.94 correlation compares to a maximum possible one of 1.0 and can be viewed as akin to getting a 94% mark in an exam.
The deterioration in the economy during 2008 and 2009, as measured by the increase in the unemployment rate from 3.7% to 6.9%, appears to be the most important factor that drove the increase in mortgagee sales.
The recession in this period was pretty severe and banks were under lots of stress because of the financial crisis and problems with access to overseas funding.
Despite this probably being as testing a time as we are likely to see, the level of mortgagee sales was low compared to the total number of mortgages and by implication the stress on the banking sector as a result of mortgage defaults was manageable.
I suspect that too much low quality lending to developers for residential and especially coastal subdivisions during the 2002 to 2007 property boom caused much more pain for financiers than did default by mortgage borrowers.
NZ's experience in 2008 and 2009 when debt servicing as a % of income was much higher than now, the economy experienced a major recession and the financial sector hit a major pothole suggests the RB's current concern about the risk of mortgage default to the banking sector is greatly exaggerated.
If my memory serves me well, just before the financial crisis arrived banks were approving plenty of low deposit mortgage loans and even with that thrown in there wasn't a major problem in terms of mortgage default.
The other concern seems to be the current high level of house prices compared to incomes and the risk that house prices will fall in response to eventual interest rate increases and that this will leave unwary or overly optimistic borrowers with negative equity, which could be a risk to the banking sector.
House prices may fall, but it is questionable whether falling house prices will spur a major increase in mortgage defaults on a scale that would threaten the soundness of the banking system.
The right chart above shows no correlation between the national average house price and the number of mortgagee sales.
Certainly, the fall in house prices in 2008 may have played a part in driving up mortgagee sales, but mortgagee sales kept climbing in 2009 even though house prices largely rebounded.
Mortgagee sales fell in 2010 when there was a moderate fall in house prices and have largely drifted sideways since 2011 despite a sizeable increase in house prices.
By contrast, what has happened with mortgagee sales over the whole period for which I can obtain data (i.e. back to 2007) fits pretty well with what has happened to the unemployment rate. I argue in our pay-to-view housing reports that housing prices are more than normally susceptible to interest rate hikes because debt to income ratios are high (see this link for info on these reports).
But even in what are likely to be the most severe circumstances NZ faces, in 2008 and 2009 house prices didn't collapse and cause massive mortgage defaults.
Not only are house prices not a major factor in driving debt defaults, the odds of a US-style fall in house prices being experienced in NZ are infinitesimal.
The NZ situation is nothing like the US situation was where lots of people who should have not been offered loans were for political reasons and, more importantly, dud loans were lumped together with lots of extra gearing and sold to unwary investors.
What happened in the US is in a different galaxy to what could reasonably be expected to happen in NZ.
In my assessment the historical experience in NZ and the current debt and debt servicing ratios don't justify the RB targeting low deposit borrowers in the way it has under the guise of protecting the banking sector from itself.
The RB is effectively saying that banks are not smart enough to protect their own survival in the context of mortgage lending.
NZ history teaches us that banks and more so finance companies have got drawn into property booms that brought about their demise (e.g. NZI Bank and BNZ in the 1980s, and many finance companies in the 2000s).
But the major failures have generally been associated with non-mortgage lending.
NZ history is short on examples of banks and finance companies failing on a scale that threatened the banking system as a result of mortgage lending in general or specifically low deposit mortgage lending.
So I see no sound basis for singling out low deposit borrowers in the way the RB has.
Are low deposit borrowers as a group more at risk of losing jobs or having businesses fail and default on mortgages when the next recession arrives than are moderate deposit borrowers?
If someone has a 10% deposit on their house rather than a 21% deposit are they much more at risk of defaulting if their business fails and they can't pay the bills? Obviously, to some extent the answer is yes.
But I believe the RB is missing the critical point.
Falling house prices don't pose a clear risk to mortgage default and the odds of NZ experiencing a US-style fall in house prices is infinitesimal.
The real risk of large scale mortgage default is a major recession that drives the unemployment rate up significantly and results in lots of business failures. But the next major recession isn't likely to come along for quite a number of years, by which time the government's housing initiatives will have hopefully solved the underlying housing affordability problem and mean new borrowers don't have to borrow to the hilt to be able to afford to buy a house.
And by which time the current low deposit borrowers will no longer have low equity because incomes will have increased relative to fixed mortgage amounts and some will have paid off parts of their mortgages. And, again, the 2008 to 2009 experience suggests that even when the next major recession arrives mortgage defaults won't be high enough to threaten the banking sector.
Average mortgages are surprisingly low
A client recently asked me what the average mortgage was and I had no idea, with his question spurring this Raving. Based on all the fuss the RB has made about the risk to the banking sector from mortgage lending I had assumed the average mortgage had skyrocketed in recent times, but it hasn’t.
The left chart shows the average mortgage compared to the national median dwelling price.
The RB data show that at September 2013 there were 1,430,689 residential mortgages with banks and $187,684 mln in residential mortgage lending. This equates to an average mortgage of $131,184.
Since January 2011 the median dwelling price has increased 21.1% and the average mortgage only 10.1%.
However, the increase in the average mortgage in the last couple of years has resulted in it increasing significantly as a percentage of the average income (right chart).
The right chart uses a different measure of average income than that used by the RB in the chart on page 2, but this doesn't change the picture in terms of the reported increase in the average mortgage as a % of income.
However, the average mortgage as a % of the average income is still well below the peak level in 2007 that didn't result in a high level of mortgage default of the scale that should concern the RB about the risk of mortgage lending to the banking sector.
Conspiracy theories are justified
Maybe the RB isn't just taking advantage of the recent meltdown in the US housing market that bears little relevance to NZ prospects to increase its sphere of influence when it isn't really justified from a banking sector risk perspective.
Maybe Governor Wheeler is taking over from Helen Clarke in telling us what we should and shouldn't do, like a benevolent dictator. If people and banks are too dumb or short-sighted to look after their own financial affairs, uncle Graeme will protect them from themselves.
The trouble with this is that, rightly or wrongly, people don't want this sort of protection.
Low deposit borrowers will, understandably, feel unjustly picked on and with the help of financial innovation will find ways to supplement deposits rather than accept the situation and wait longer before buying.
The result will in many instances be more expensive total borrowing costs for this group (e.g. higher interest rates for the supplementary deposit money; higher interest rates from non-bank lenders; and the higher interest rates they now face from banks).
Many low deposit borrowers are likely to end up in worse financial positions as a result of the 1 October lending restrictions than would have been the case otherwise, which means a higher risk of default.
In my assessment the lending restrictions will increase total finance sector risk even if they reduce banking sector risk.
They will result in some of the economy's scarce resources being wasted on finding ways around what are perceived to be unjust regulations.
Under this conspiracy theory the net result will be the group of people who the lending restrictions are designed to protect from themselves being hurt.
*Rodney Dickens is the managing director and chief research officer of Strategic Risk Analysis Limited.