The author has been described by News Ltd as an "iconoclast", "Svengali", a pollie's "economist muse", and "pungently accurate". Fairfax says he is a "Renaissance man" and "one of Australia’s most respected analysts." Stephen Koukoulas concludes that he is "85% right", and "would make a great Opposition leader." Terry McCrann claims the author thinks "‘nuance’ is a trendy village in the south of France", but can be "scintillating" when he thinks "clearly". The ACTU reckons he’s "an enigma wrapped in a Bloomberg terminal, wrapped in some apparently well-honed abs."

Wednesday, January 18, 2012

The housing 'destructionistas' dissembling...

If you trawl through CBA's covered bond promo materials there is some fascinating new data that confirms points about the housing market I've made here in the past. It also serves to highlight the quite misleading dissembling of the housing 'desctructionistas' (eg, the Steve Keens of the world).

The first extract of interest compares the default rates of CBA's first time buyers with their overall portfolio. You can see from the chart below that until 2010 first timers actually had lower default rates than your average borrower. Recall the hyperbole that folks like Keen were pushing during the GFC, when he claimed that the major banks lending to first timers combined with the government's first time buyers' boost would result in a mass of delinquencies? Sorry, that was just another mistruth.

In some respects, the parsimony of first timers makes intuitive sense: they are typically investing a larger share of their net worth and income in buying a home and servicing the repayments than older and wealthier buyers. While this brings with it some extra risk (eg, via lower equity), it also tells us that many first timers on lower incomes are committing higher shares of their labour earnings to keeping a roof over their head than older buyers. For some borrowers, this creates a better alignment of interests. For others, it may heighten the risk they default. What we can see from CBA's data is that first time buyers don't seem to have a fundamentally higher probability of default than other borrowers, although there has been convergence since 2010.

There are other explanations for this, of course: as I've argued before, our banks are very cautious in setting their lending criteria. It is possible that CBA is even more rigorous in its credit assessment when looking at first timers versus the rest of the borrowing population, which in turn helps to reduce credit risk.

The second CBA chart looks at the losses realised on CBA's home loan portfolio in comparison to all non-home loan assets between 1983 and 2011. It portrays a familiar story: historically, the real  risk for Australian banks lies not in lending to people buying homes (as many pundits claim), but rather in business borrowings and commercial property (especially the latter).

This is why, quite rationally, APRA makes it more expensive for banks to lend to businesses--via higher 'risk-weightings'--than to people acquiring established homes. In short, the historical risk of doing so has been demonstrably higher, so APRA makes you hold more capital against these loans.

Interestingly, we now have two clear case studies--the 1991 recession and the 2007-09 GFC--where residential lending has proven its vastly superior resilience relative to business credits.

This is also why some major bankers' calls for APRA to change the risk-weightings to favour business lending would appear to be either ignorant or subversive.

As a final aside, it is gratifying to see that my good friends over at the CBA are using our 'house price-to-income ratio' analysis in their covered bond promo materials, although I note that this is incorrectly labelled: this data comes from Rismark, not RP Data (see below).

A related observation would be that using the ABS's free-standing, established house price index, which ignores the circa 25% of all homes that are apartments, semi-detached, and terraces (especially in metro areas) to mark-to-market a covered bond portfolio, is hazardous, to say the least. It opens CBA up to the criticism that it is not properly assessing the collateral values of the circa 80% of all assets that are being used to secure the covered bond. For example, what happens if the higher priced detached house market is outperforming the apartment, semi and terrace sectors, and thereby giving artificially positive marks to the CBA portfolio? There is no need for these valuation errors since there are freely available indices that cover all property types--and which more accurately measure price changes (according to the RBA)--that are currently being used by CBA's competitors, and indeed promoted by CBA's economists. A very odd oversight.