Are Australia and Canada’s cooling housing markets a threat to financial stability?
After a long housing boom in Australia and Canada, some observers expect a sharp correction. Should this spark concern about the wider economic impact in these countries?
Property (and debt) boom
House prices in Australia and Canada have witnessed a spectacular rise in the past two decades.
Thanks to strong population growth, high net migration inflows, loose monetary policy and increasing domestic and foreign investor participation, house prices in these two countries have been among the fastest growing in the OECD, rising at 14% per annum since 1999 (Chart 1).
With the strong upward momentum in prices, some house price overvaluation has emerged and a rapid return to fair value in housing markets could potentially have a significant macroeconomic impact.
This is because such strong increases in asset prices have created significant housing imbalances and vulnerabilities. Boosting the value of collateral has made borrowing easier, thereby raising household debt. Debt-to-income ratios show a worrisome picture: household debt, at 190% of disposable income in Australia and 170% in Canada, is at an historical peak in both countries (Chart 2). Therefore, a sharp price fall to return to fair value will be translated into declines in income and wealth and significant reduction of consumer spending.
However, while household debt levels are high and rising, debt service ratios tell a different story and offer a more reassuring picture, at least for the medium term. Debt service ratios measure the amount of income used for interest payments and amortisations, giving an accurate picture of spending constraint tightness.
The Australian debt service ratio is substantially below its pre-crisis peak, while the Canadian ratio seems to have stabilised (Chart 3).
The reduction of debt service ratios from the pre-crisis highs was mainly driven by easier monetary policy in both countries, as falling interest rates have helped contain interest spending, offsetting the cost of servicing larger amounts of debt.
With economic growth recovering in both countries, monetary policy normalisation is on the cards, raising concerns over financial and macroeconomic stability. Higher rates could put pressure on borrowers while raising the debt service ratio, and hit consumer spending dramatically.
As shown in chart 4, the tightening cycle initiated by the Bank of Canada in July 2017 is raising the cost of borrowing for households through higher mortgage rates. As a result, the pace of expansion of residential mortgage credit has started to slow last summer, highlighting the impact of tighter monetary policy on household behaviour (Chart 5).
Residential investment and housing starts have remained robust in 2017 (chart 6), as they are the result of lagged household decisions and their response to changes in interest rates is not immediate. However, house prices, a more timely indicator, show clearly that the Canadian housing market is losing momentum (Chart 7), suggesting that residential investment and housing activity are likely to start deteriorating this year.
Higher rates will also increase the debt service ratio, posing severe risks to financial stability, with the potential to amplify the impact of a slowdown in the housing sector to the economy. Using data from 1999 to 2017, we estimate that the policy interest rate should rise from the current level of 1.25% to 3-3.5% for the debt-service ratio to reach its previous peak of 13.2%.
In Australia the beginning of the housing boom broadly coincided with the current monetary easing cycle, which started with the slowdown of commodity prices in November 2011. Low borrowing rates have stimulated credit growth, which began to accelerate in 2013, keeping the debt-service ratio above 15%.
Housing credit growth started losing momentum in 2017 (chart 5), thanks to the introduction of new macro-prudential measures aimed at restricting lending to the highly-indebted household sector. As a result, the housing market started cooling last year with declining residential investment and building approvals, and a slowdown in house price growth (Charts 8 and 9). In the meantime, household indebtedness remains elevated, primarily due to high levels of housing debt, although weak income growth is also contributing.
The potential rise of policy rates could further increase household indebtedness, making households more vulnerable to negative income shocks. We therefore run the same exercise we did for Canada in order to evaluate how vulnerable the housing sector is to higher interest rates. We find that the policy interest rate should increase from its current level of 1.5% to 6.5-7% for the debt-service ratio to return to its 17.9% peak.
Conclusion: no imminent crash
Housing markets in Australia and Canada are cooling, raising concerns over economic growth. We expect residential investment to be a drag on growth through 2018 in both countries. Moreover, declining house prices are likely to hit consumer spending, while monetary policy normalisation could represent a threat to household debt.
However, we do not expect an immediate housing market crash, and based on our analysis interest rates must rise substantially for the debt-service ratios to reach their previous peaks. As we think monetary policy normalisation will be gradual in both the economies, financial and macroeconomic stability will continue