Booms and busts in housing markets not only contribute to a range of social problems but, as demonstrated by their role in the global financial crisis of 2007–2009, they also pose a very real threat to financial stability. Policy-makers can — and should — act to mitigate this threat, but how they can do this effectively is hotly debated.
The Committee on the Global Financial System (CGFS) — a central bank forum hosted by the Bank for International Settlements in Basel, Switzerland — has conducted an in-depth analysis of this problem in multiple countries and found that macroprudential policies are a key part of the solution. These policies work by maintaining sound lending standards and boosting lenders’ loss-absorbing buffers.
Housing markets as a source of risk
Housing market developments are closely monitored by central banks because of the pivotal role that boom–bust cycles have played in destabilising financial systems in the past. Mortgage payments often absorb a significant share of household income. Thus, higher interest rates, like the increases seen over the past two years, can lead to difficulties in servicing mortgages and ultimately trigger defaults, especially when combined with a drop in house prices that turns housing equity negative. At the same time, where the banking sector has large direct exposures to housing, defaults can result in large losses.
When lenders and borrowers have insufficient buffers, these effects can be amplified by lenders restricting credit and households cutting consumption. Weakness in housing markets thereby spreads to the broader economy.
Housing booms can exacerbate these effects by undermining the resilience of borrowers and lenders. Expectations that house prices will continue to rise can lead them to underappreciate the risks of taking on and providing ever more debt. The resulting accumulation makes borrowers and lenders more vulnerable to higher rates or a change in economic conditions. The failure of borrowers and lenders to appreciate the broader impact of their individual decisions can result in excessive leverage in the economy.
Over the past decade, authorities around the world have experimented with using macroprudential policies to tackle boom–bust cycles in housing markets. Drawing on the practical experience of policy-makers in 14 jurisdictions over a combined 168 years, the CGFS report offers actionable insights into what policies have worked in the past, how they have worked and ways to strengthen their effectiveness.
Matching tools with objectives
An overarching insight is that certain tools are better at addressing specific objectives than others. Macroprudential authorities have experimented with a variety of tools and now have a relatively good understanding of where and when certain tools work best.
Borrower-based policies, such as limits on loan-to-value (LTV) and debt service-to-income (DSTI) help to maintain sound lending standards by ensuring that the size of the loan is not too large relative to either the value of the property or the borrower’s ability to repay it.
Among these policies, ratios based on borrowers’ income, such as DSTI and debt-to-income limits, have become the go-to tools to maintain sound lending standards. Limits on LTV ratios are a less effective way to build borrower resilience than ratios based on borrowers’ income, although they do help strengthen lenders’ resilience if households default.
The effectiveness of such policies was recently revealed in Ireland, where households with lower loan-to-income ratios were much less likely to utilise Covid-19 mortgage payment breaks. As evidence has accumulated of the effectiveness of income-based ratios, more authorities have been granted powers to use them. In New Zealand, for example, the central bank was recently given permission to introduce income-based limits.
Capital-based policies, such as floors and add-ons to risk weights, boost the loss-absorbing buffers that lenders hold against their exposures. These policies are effective in building buffers against systemic risks not captured by microprudential capital frameworks. In Luxembourg and Belgium, which have not experienced a housing crisis, macroprudential recommendations on risk-weight floors and add-ons have been used to bring average risk weights on mortgages in line with other European economies.
With this in mind, the report distils four key lessons.
Consistency across housing-related policies
First, macroprudential policies should not be the only tools in town. Tax, planning and land supply policies all have a decisive influence on demand-supply imbalances in the housing market. While such policies are outside the remit of macroprudential authorities and decided by governments, successfully mitigating boom-bust cycles in housing markets requires consistency across all related policies.
In Singapore, for example, this takes the form of close co-operation between the central bank, the tax agency and land supply authority. The land supply authority moderates property prices by ensuring that there is sufficient housing supply to match demand. Tax measures on second and subsequent buyers dampen speculative activity. Macroprudential policies then complement these measures by strengthening the resilience of borrowers and lenders.
Governance influences policy effectiveness
Second, governance arrangements have an important influence on policy effectiveness. Policy is better targeted at risks when macroprudential authorities have a clear mandate, operational independence that shields policy actions from political considerations, and a legal basis to direct policy across the full range of macroprudential tools.
In France, for example, the macroprudential authority includes the central bank as well as the finance ministry, regulators and academics. This brings a diversity of perspectives, along with additional policy tools. At the same time, the central bank is the only member with the power to initiate macroprudential measures. This governance arrangement promotes the operational independence needed to fulfil the authority’s mandate.
A long-standing concern with macroprudential policies is circumvention by lenders outside the regulatory perimeter or by borrowers who set up separate legal entities. Having a clear legal basis to introduce tools that address all sources of housing risks, such as in Ireland, has been an important condition that has helped to ensure that such leakages are plugged.
Tools that do not require adjustment
Third, macroprudential policies can be made more effective by prioritising tools that achieve their objective without requiring adjustment. This reduces the lag between identifying risks and taking remedial policy action.
For example, income-based limits calibrated to lock in prudent lending standards — such as in Ireland and France, or for stressed conditions as in Australia and Hong Kong — help borrowers withstand higher interest rates without the need to adjust policies as market condition evolve. This contrasts with other tools which require periodic adjustment to meet their objectives: for example, LTV limits contribute less to lender resilience when house prices move well above fundamentals.
Openness about cost-benefit trade offs
Finally, macroprudential policies potentially impact output growth and can affect some groups more than others, so providing a clear cost–benefit rationale is essential. The benefits of successful policy actions are largely invisible because it is difficult to demonstrate the counterfactual of financial fragility. They are also dispersed across the economy and the population in general, so pointing to specific winners to justify these policies can be difficult, whereas the potential costs are more immediately visible and borne by a specific and sometimes vocal minority. For example, borrower-based measures might result in less lending to those with lower incomes or less savings.
However, authorities have taken important steps to mitigate potential costs. For example, borrower-based measures for first-time buyers (FTBs) or on low-value properties are calibrated so that they are often less strict than those for second-time and subsequent buyers. FTBs tend to have lower credit risk, for example, due to higher expected income growth early in their careers, which can justify less stringent limits.
Flexibility margins are another innovation that have helped mitigate costs. Also known as speed limits, flow limits or allowances, they exempt a share of new lending from borrower-based measures. They give lenders some leeway to use their private information to lend to creditworthy borrowers who do not conform with the blanket LTV or DSTI limits. This can be particularly useful when there are large differences in house prices within a country.
Looking back to look forward
Macroprudential policies cannot fix structural problems in housing markets. What they can and should do is mitigate the credit driven excesses that not only threaten the stability of the financial system, but also make housing less affordable. This is a silent — but highly significant — benefit that is easy to overlook.
The implementation of these policies in many countries over the past decade has helped to make lenders and borrowers more resilient to the sharp increases in interest rates that have been needed to contain inflation. Despite initial reservations, these approaches have generally been accepted by the public and also by financial institutions as a useful complement to other housing-related policies.
Looking forward, deliberate, coherent and consistent application of these four lessons can help to mitigate boom-bust cycles in housing markets. In addition, clear and open communication about the costs and benefits will help to maintain support for these policies, even as memories of housing crises inevitably fade.
Emmanuelle Assouan is chair of the Committee on the Global Financial System (CGFS) Working Group on policies to mitigate housing market risks and director-general of financial stability and operations at the Bank of France; and Ryan Banerjee is secretary of the CGFS Working Group and principal economist at the Bank for International Settlements.
