Tougher LVR's obviously suggests some concern about valuation risks. Having been involved in bad debt scenarios in one of the banks, the big driver of credit losses (at least in bank models) has always been unemployment. Most people will cut back and cut back expenditure to avoid losing the house, particularly given Australian loans are full recourse. Thus, if you get a 20% fall in house prices, this is only a problem for the banks if they have to foreclose on a significant proportion of their portfolio and as most people manage to keep paying this remains a small proportion. However, if unemployment gets bad, people can not afford repayments, property gets dumped on the market, causing further falls in prices causing further losses in value and further credit losses. Thus, unemployment is the big issue. If it increases significantly, I foresee large credit losses and property price falls. However, this looks unlikely and unemployment appears to have peaked.
If unemployment continues to fall, given overvaluation by most metrics, I would expect that we would see an extended period of flat to low growth to allow incomes to catch up with property prices.
Mind you the models could be wrong. The mantra used to be that margin lending was effectively risk free. But the collapse of ABC, B&B etc where banks had single asset portfolios proved the models very wrong.
