Paul Ryan| Real Estate Insights| 31 March 2021
The decision on whether to intervene in a red hot housing market lies with our regulators. The problem is, they don’t have the right data to make the call.
Property prices are surging in 2021, as pent-up demand fuelled by record low rates ignites the market. Like buyers and sellers, regulators are keenly watching the market, but it’s not prices they’re worried about, it’s lending standards.
So far the regulators don’t have concerns. In their review of the December numbers, the Australian Prudential Regulation Authority concluded there had so far been no material relaxation in lending standards.
But the available indicators are flawed, and regulators must work with lenders to provide better ones.
Regulators, including APRA and the Reserve Bank of Australia, consider many indicators when assessing the riskiness of lending. Key metrics that have raised flags recently include high debt-to-income (DTI) ratio lending and high loan-to-valuation ratio (LVR) lending.
Debt-to-income ratios are a poor measure of risk
Regulators track loans extended at high DTI ratios – the amount borrowed relative to a borrower’s income.
Borrowers with high DTIs have higher repayments relative to their income, which may make them more likely to experience difficulty making repayments.
APRA data, released this month, showed the share of lending at DTI ratios of six or more (six times the borrower’s annual income) increased to 17% of new lending in the December 2020 quarter. This is the highest level recorded since data collection began in May 2019.
However, DTI ratios don’t provide a good measure of borrower risk over time.
In May 2019, a borrower with loan repayments equal to 25% of their income had a DTI ratio of 4.3. In December 2020, a borrower with the exact same monthly repayments had a DTI ratio of 5, but despite the increase it’s difficult to argue this borrower is now higher-risk.
That’s because interest rate falls allow borrowers to make repayments on higher levels of debt, so as rates have fallen over recent years it should come as no surprise that high DTI ratio lending has increased.
Additionally, there is only a short span of available data making it even more difficult to judge what a ‘safe’ level of high DTI ratio borrowing is.
Loan-to-valuation ratios are misleading
High LVR loans – the loan amount relative to the value of the property – effectively help the regulator track how many borrowers are securing finance with small deposits.
These loans pose a higher risk of losses to the lender; if the borrower defaults, there is a larger chance the sale of the property won’t cover the remaining loan.
Borrowers who choose to take out large loans with small deposits may pose a risk for multiple reasons. Are they less prudent with their money? Do they overestimate just how much property prices will keep increasing by? All contribute to borrower risk.
The share of loans made with a LVR of 80 or more has increased to almost 43% of all new loans in the most recent APRA data, but these figures are mostly meaningles
In reality, many people just increase their borrowing to a LVR of exactly 80 to get the best interest rates and avoid paying lenders mortgage insurance (LMI).
That doesn’t mean they don’t have extra funds, as many will keep other funds in offset accounts as a free form of liquidity. That’s why any measure of borrowing at a LVR of 80 doesn’t say much about risk, or even indebtedness.
More from Insights
However, the share of new loans with a LVR of 90 or more has ticked up to 11.5%, which could signal more risk in new lending. But the regulator noted this likely just reflects the high share of first-home buyers in the market.
First-home buyers tend to have smaller deposits, but often provide loan guarantors and lately many have insurance provided by the federal government’s First Home Loan Deposit Scheme. They are also lower risk on average.
First-home buyer outlook for 202102:27
The missing metric for regulators
To assess borrower risk, lenders calculate their net income surplus (NIS). Simply put, this is the amount of income left after making loan repayments (calculated with a ‘buffer’ added in case interest rates go up) and other expenses.
Borrowers with a lower NIS have less ‘spare’ income after repayments, so are riskier.
This metric better accounts for the effect of interest rate changes over time and also measures what both borrowers and lenders actually think about when loans are made. However, this data is not available to the regulator.
Banks actually have a fairly strong record of managing risk, hence why the share of borrowers experiencing difficulty repaying their mortgages remained low through both the global financial crisis as well as the COVID crisis.
However, regulators would be in a much stronger position to make calls on lending restrictions if they had access to the same data the banks lean on.
Until then, they’re largely flying blind.
Paul Ryan is an economist at REA Group. Prior to that he worked at the Reserve Bank of Australia for more than a decade, most recently specialising in mortgage lending risks.