Thank you for the opportunity to speak with you today. When the pandemic hit Australia in early 2020, policymakers entered a world of unknowns. The health situation required drastic curtailment of much economic activity. Fiscal and monetary policy both responded quickly and strongly. Governments provided substantial financial support to firms and individuals who overnight lost business and jobs. The Reserve Bank reduced interest rates to historically low levels and introduced a suite of other policy measures to lower funding costs and interest rates across the economy. The banks offered borrowers deferrals on their loan repayments. There were moratoriums on evictions and rent relief for businesses and households. The idea was that we were building a bridge to get us over the economic crevasse created by the pandemic.

At that time we were writing our Financial Stability Review. Things were changing so quickly that we seemed to be rewriting it daily. It ended up being quite a different style of Review than we would typically produce. Its focus was on the ability of the financial system to absorb the shock and support the economy. It also focused on the potential implications for household and business balance sheets. In particular, we noted the decline in demand in the residential property market and highlighted the risks this could pose to highly indebted households. To quote:

A key financial stability risk is the extent to which the weakness in economic activity spills over to the housing and commercial property markets. The prospect of large declines in property prices presents significant balance sheet risks for households, businesses and lenders.

Now, eighteen months later, as we are drafting our Financial Stability Review that will come out in October, we are not talking about financial stability risks from housing debt given falling prices. Rather, we are monitoring the surprising and dramatic rebound in the housing market and asking ourselves whether there are financial stability risks from these developments and the accompanying rise in housing credit.

Although it is not property prices per se that we care about from a financial stability perspective, developments in the housing market (including prices) provide information on the emergence of financial stability risks. Today I want to set out how we think about the housing market in the context of financial stability.

The housing market has been buoyant

After initially declining a bit at the onset of the pandemic, housing prices have recovered strongly (Graph 1). Despite some moderation during the recent lockdowns in NSW and Victoria, the strength has been broad based across states and across metropolitan and regional areas and particularly in detached housing.

In this environment it is unsurprising that housing credit growth has picked up and the rapid rebound in loan commitments suggests some further pickup in growth to come (Graph 2). Growth in housing credit is currently running at an annualised rate of around 7 per cent. Recent data on commitments, combined with several assumptions, suggest housing credit growth could peak at an annualised rate of around 11 per cent early next year.

There are a number of factors driving this strength. Low interest rates are clearly an important factor, making it easier for households to service their debt and changing the trade off on whether to rent or buy. Also important has been the government support for housing construction, including the HomeBuilder scheme. Growth in housing credit in late 2020 and early 2021 was driven by increases in loan commitments for the purchase of new dwellings or housing construction. More recently, however, there are signs that credit growth is increasingly being driven by purchases of existing dwellings, in which turnover is currently quite high.

The strength in the housing market is positive for the economy, and indeed an important channel for monetary policy to support the economy through housing construction, home improvements and purchases of household items. In fact, strongly rising housing markets and housing credit have been a feature of many other countries over the past 18 months as monetary policy has lowered lending rates and governments have provided fiscal support. However, while household debt to income in Australia hasn't increased much over recent years, it is at a high level, both historically and relative to other countries. So sustained strong growth in credit in excess of income growth may result in vulnerabilities building in bank and household balance sheets.

What does this imply for financial stability risks?

There are two interrelated channels through which the housing market could have implications for financial stability: bank balance sheets and household balance sheets.

Banks

The Australian banks are very exposed to the housing market. Around 60 per cent of their lending is for housing (Graph 3). And housing is also common collateral for loans to small and medium businesses. A decline in housing prices impacts the value of that collateral but of itself does not necessarily lead to losses. If there were an economic shock that resulted in borrowers being unable to service their debt, however, and at the same time housing prices fell sharply, the debt would be unable to be recovered in full by the sale of the property, resulting in a loss to the banks. If the losses were large enough the banks might exacerbate the decline by reducing their lending.

There are two broad mitigants against this – lending standards and capital.

Bank lending standards and practices determine how much risks bank take on. The Global Financial Crisis (GFC) demonstrated the impact that poor lending standards can have on bank balance sheets. Banks are in the business of taking risk – if they only lent money when they were 100% certain it would be repaid there would be too little credit. So there is a balance to be struck. Some loans will go bad but banks manage those risks by maintaining appropriate standards on who they lend to and how much. This involves, among other things, standards around serviceability and loan to valuation ratios.

Currently in Australia, the evidence suggests that lending standards overall have been maintained in the face of very strong demand for housing (Graph 4). This is in contrast to 2014-2017 when there were signs that lending standards were declining. During that period the Australian Prudential Regulation Authority (APRA) took a number of measures to bolster lending standards and curb growth in lending that was posing a higher risk to financial and economic stability.^[1] The share of new lending at high loan to valuation ratios rose last year, in part reflecting an increase in the number of first home buyers who tend to have lower deposits. Lending at high debt to income ratios has also risen since early 2020. APRA and the Bank are monitoring these trends closely.

The other mitigant is capital. The more well capitalised the banking system, the more ability it has to absorb losses. The GFC, which originated in a boom and bust in the housing market in the United States, demonstrated what can happen if banks are not well capitalised. The substantial losses sustained by many banks during that period severely depleted capital positions and resulted in banks pulling back from credit provision amplifying the economic downturn. This wasn't so much an issue in Australia though since the Australian banks were quite well capitalised at the time.

Following that episode, there was a worldwide effort to increase the capital in the banking system through what are known as the Basel III reforms, implemented in Australia by APRA. While the Australian banks did not have the undercapitalisation problems of some overseas banks, they now have even stronger capital positions than they did at the time of the GFC. Indeed, stress tests undertaken by the Bank and reported in our October 2020 Review indicate that even a very severe recession and substantial fall in property prices would still leave the banks with capital above their regulatory minima.^[2] The strength of the banks allowed the banks to respond positively to the pandemic shock, supporting the economy through lending and loan repayment deferrals rather than amplifying the shock.

Households

The second channel through which the housing market could impact financial stability is the household sector. The bulk of household debt in Australia is mortgage debt – people borrowing to buy homes. Households that have borrowed a lot to purchase a home relative to their income could, in the event of a shock to their employment status or income, find they are unable to continue to service their loans. This may not be a problem if the value of the loan is less than the value of the property. While it would be disruptive and stressful to have to sell their home, they would at least be able to repay the debt. However, if the value of the home has fallen substantially since they took out the loan, they will be unable to ‘self-cure’ in this way. In these circumstances, banks would incur losses and if there are enough, like in the GFC, there could be a substantial impact running from households to banks.

As I have already noted above, this currently seems an unlikely channel in Australia's case given the strength of the banks' balance sheets. But there is another channel through which a high level of household debt could have implications for the economy and by extension financial stability – through its impact on consumption. Over-exuberance in the housing market can amplify these risks in two ways. First, rapid price rises can increase the likelihood that some new borrowers will over-stretch their financial capacity in order to obtain a new loan, making them more likely to reduce their consumption in response to a shock to their incomes. Second, if rapid price rises ultimately prove to be unsustainable they could lead to sharp declines in price and turnover in the future. This in turn could result in reduced spending, both directly as a result of a decline in turnover and through the wealth effect.

Household debt in Australia has increased substantially over the past 30 years (Graph 5). Much of this occurred in the 1990s and 2000s as declines in inflation and interest rates, and rises in real income, meant that people were able to service higher levels of debt. Financial liberalisation in the 1980s also removed some unnecessary credit constraints. Both of these are permanent shifts and no cause for alarm. And although household debt to income in Australia is higher than many overseas countries, this can be partly explained by the fact that, unlike many other countries, households own the rental stock (and hence have the debt) in Australia, rather than corporate landlords owned by specialist firms or pension funds.^[3]

Nevertheless, there is quite a bit of international and domestic evidence that suggests that households that have high levels of debt are more likely to curb consumption in response to shocks in income or wealth.^[4] A large number of highly indebted households reacting in such a way to an economic downturn or decline in housing prices could amplify the economic shock. A boom in the housing market, accompanied by an increase in housing debt could therefore make the economy more susceptible to downturns.

For example, in a recession in which a large number of indebted households suffer reduced income, from say loss of employment or reduced hours, they might choose to reduce their consumption. It may be just precautionary. But if households are constrained (in the sense that they don't have a great deal of income left after meeting debt servicing requirements and the basics of their lifestyle) they are more likely to reduce consumption. This will amplify the initial impact of the economic shock.

Another way in which developments in the housing market and debt could impact the economy is through the wealth effect. Sharp declines in housing prices could result in households curbing their consumption, more so if they are highly leveraged, further exacerbating any downturn in the economy. In this sense, sharp rises in housing prices that are not associated with fundamentals could lead to instability by raising the risk of a subsequent decline. This is one reason why we watch investor activity in the housing market quite closely. Investors have the potential to amplify movements in housing prices, buying when prices are rising quickly but potentially selling when they fall in an effort to avoid a capital loss. Any amplified swings in housing prices will have an impact on the wealth of all households and hence potentially their consumption.

As I noted above, while picking up, investor activity in the housing market is currently nowhere near the levels it was in 2014 when APRA introduced its investor lending benchmark. And it is hard to judge in real time whether housing prices are out of line with fundamentals. Nevertheless, when prices are rising very rapidly and there are expectations that this will continue, borrowers are more likely to overstretch their financial capacity in order to purchase property. We are therefore watching developments in housing markets and credit very closely.

What can the authorities do?

Given the potential for risks to be building, there is a question about what the authorities can do aside from monitor developments. Unlike in 2014 and 2017, the concerns this time are not specific types of lending such as investor or interest only lending. So the tools used at that time are not really appropriate at this time. This suggests that if there were to be a need for so-called macro-prudential tools to address rising risks, they should be targeted at the risks arising from highly indebted borrowers. Tools that address serviceability of loans and the amount of credit that can be obtained by individual borrowers are more likely to be relevant. Indeed we have seen such tools used in a number of countries in recent times and they could be employed in Australia should the circumstances be judged to warrant it.

Conclusion

The unprecedented monetary and fiscal support to the Australian economy through the pandemic has helped ‘build a bridge’ to the other side. The strong recovery in the housing market is part of that bridge. But with the increase in housing prices and housing debt, risks to financial stability could be building. Even though the banks have strong balance sheets and lending standards are being maintained, there is a risk that in this environment, households will become increasingly indebted. A high level of debt could pose risks to the economy in the event of a shock to household incomes or a sharp decline in housing prices. It is these macro-financial risks that warrant close watching. Whether or not there is need to consider macro-prudential tools to address these risks is something we are continually assessing.

Thank you.

Endnotes

I would like to thank Jonathan Kearns, Natasha Cassidy, Michelle Wright and Amelia Gao for assistance in preparing this talk. [*]

See RBA (Reserve Bank of Australia) (2018), Financial Stability Review, October [1]

See RBA (2020), Financial Stability Review, October [2]

See J Kearns, M Major and D Norman (2020), ‘How Risky is Australian Household Debt?’, RBA Research Discussion Paper No 2020-05 [3]

See F Price, B Beckers and G La Cava (2019), ‘The Effect of Mortgage Debt on Consumer Spending: Evidence from Household-level Data’, RBA Research Discussion Paper No 2019-06 [4]