Closing remarks by Director for Financial Stability, Vasileios Madouros, at the Central Bank Conference to inform the 2021-2022 mortgage measures framework review

27 April 2022 Speech

Vasileios Madouros

Good afternoon

Let me start by thanking you all for taking part at the conference.

I would particularly like to thank our speakers, for sharing their insights on, and their own experiences with, macroprudential measures in the mortgage market.

And I would also like to thank everyone for their active participation, which made the event even richer.

As more and more countries use similar measures, and as more and more research on the effect of these measures is undertaken, coming together to share perspectives and learnings in events such as this one is a key element of our efforts to continue strengthening our policy frameworks.

For the closing remarks, I thought I might offer some reflections on how the issues raised during the conference relate to our own review of the framework of the mortgage measures. 

Let me start with yesterday’s session, on the insights from academic thinking around the lessons from previous credit-driven housing booms and busts.

First, and importantly, the discussion served as a reminder of the basic rationale for macroprudential measures in the mortgage market. The evidence – both over time and across countries – is very consistent: prolonged periods of unsustainable mortgage lending standards are associated with increased financial fragility, which often result in costly recessions, with long-lasting adverse implications.1 Of course, these are infrequent events – which also means that, as time passes, memories of the costs of financial crises begin to fade. But, while infrequent, credit-driven booms and busts have also been systematic and repeated patterns in the past. Learning from that experience, the introduction of macroprudential measures in the mortgage market since the financial crisis has sought to reduce the likelihood that similar patterns re-emerge in the future.

Yesterday’s discussion also highlighted how the costs of unsustainable mortgage lending – and therefore the benefits of macroprudential measures that guard against that – accrue to society as a whole. These policies are not about protecting banks. They are also not just about protecting borrowers that enter the mortgage market. The measures reduce the likelihood of economy-wide problems and, in doing so, provide society-wide benefits. Think, for example, of the younger generation in Ireland in the late-2000s and early 2010s. Because of the financial crisis, young people at the time faced very high rates of unemployment, even though the majority of them had not even engaged with the mortgage market by then.2 Put differently, excessive levels of indebtedness entail “negative externalities”, which pose risks to the whole economy.

A final lesson from yesterday is that macroprudential measures affect the economy, households and businesses through a multitude of channels.2 And, like all policy interventions, they entail both  economic benefits and economics costs, which we seek to balance. As macroprudential policymakers, we do not aim for resilience at any cost – that would not serve society well. So we need to continuously deepen our understanding of the transmission channels of these policies, learning from the growing domestic and international experience. This is an area that we are investing heavily in ourselves, through our research and analysis, and where further advances in academic thinking would be particularly beneficial for policymakers.

Now, turning to this morning’s session, this focused on learnings from policymakers’ experience across the world over the past decade.

Interestingly, there is a lot of commonality on what these measures are seeking to achieve – and, importantly, what they are not trying to achieve. At a high level, there is a consistent pattern looking across countries implementing these measures that they aim to limit amplification mechanisms between mortgage lending standards, the housing market and the broader economy. But another pattern that emerges looking across countries is that macroprudential measures in the mortgage market do not – and, indeed, cannot – target house price levels per se, which are affected by a multitude of factors beyond the control of central banks or regulators. This poses a communications challenge for central banks, especially given the very real, and very acute, housing affordability pressures facing people. In Ireland, the continued growth in both house prices and rents is pointing to an underlying imbalance between the demand for, and the supply of, housing services. In that context, clarity of communication around what macroprudential  measures can, and cannot, achieve is particularly important from an accountability perspective.

While objectives are broadly similar, the precise implementation of the measures across countries differ quite a bit. In some cases, the measures include both income-based limits and collateral value-based limits, in others it is only one type of limit. In some cases, the measures cover only bank lending, in others they cover both bank and non-bank mortgage lending. Now to some extent, this reflects local characteristics of credit markets. And it is important that measures are designed with those characteristics in mind, so that they can achieve their objectives. In other cases, though, they also reflect differences in policy judgements and the fact that we are collectively learning about the effectiveness of different types of implementation over time. That is not too surprising. To make an analogy with monetary policy, for example, the precise approach to monetary policy implementation has evolved significantly over time and still varies across countries.

Finally, and particularly important, is the question around the overall macroprudential policy strategy. Again, there are differences in approaches across countries here. Some jurisdictions, for example, have adjusted the calibration of the measures to cyclical shocks, others have not. There are also important questions around how borrower-based measures interact with other macroprudential policies, such as those around bank capital, and how – if at all – macroprudential measures in the mortgage market should be adjusted in response to other policies, such as monetary policy or even housing policies adopted by governments. This broad area of macroprudential policy strategy is still at a relatively early stage of development at an international level. Both academic thinking and policymaking experience is – naturally – at a less mature stage than, say, monetary policy. Our own framework review is a recognition of this. Indeed, one of the reasons we initiated our review is because it helps us take a step back and consider our overall strategy, so that the measures remain fit for purpose, not just now, but into the future.

Turning to the last session of the conference, this focused on what the evolving economy and financial system might mean for macroprudential mortgage measures in the future. In the near term, there has been increased focus on the drivers and implications of the global, synchronised upturn in house prices that we have observed recently.4 But I wanted to take a step back and focus on three structural trends that are relevant as part of our own framework review.

First, what has happened to equilibrium interest rates in recent years and what does that mean for the calibration of the measures? As you know, estimates of equilibrium interest rates have fallen in recent decades.5 Taken in isolation, lower equilibrium interest rates could – other things equal – make higher debt levels more affordable, supporting households to take on more debt without necessarily increasing aggregate risks to the economy. While conceptually this appears straightforward, especially when considering this over a long horizon, the implications for the here and now are a more difficult judgement. The equilibrium interest rate is, of course, unobservable and there is very large uncertainty around its estimate. Indeed, this uncertainty seems particularly pertinent in the current environment, given the swift reversal in market expectations around interest rates we have observed at a global level recently, with some also pointing to the potential that some of the structural disinflationary forces we have seen in recent decades may be waning.6

The second key structural factor is around trends in housing supply. Over the past decade, the recovery in housing construction has been slower than had been expected, despite continued house price growth. What happens to the housing supply curve matters, as it can affect the equilibrium level of house prices relative to household incomes in an economy. Put simply, if growth in the housing stock can only keep up with population growth at higher levels of prices, that could imply a sustained period of higher house prices relative to incomes. This also has implications for our strategy around the mortgage measures. From the perspective of society as a whole, it does not seem optimal that housing supply constraints lead to a permanently higher cost of housing relative to incomes and associated higher levels of indebtedness. The ideal policy mix would focus on the supply of housing – including affordable housing. But structural developments in housing supply also have implications for the costs and benefits trade-offs of the measures.

The third broad structural trend I would flag is around the increasing role of institutional investors in the housing market – a pattern gradually observed both in Ireland and globally recently. This can entail macro-financial benefits, by diversifying the source of financing into the housing market and in residential construction. But the growth in global institutional investment also increases exposure of the economy to global financial shocks. And, if that institutional investment were to become highly levered, it could also be a source of macro-financial vulnerabilities. The mortgage measures themselves do not apply to institutional investors, so do not guard against this. But we have recently been consulting on measures to limit leverage of property funds. Property funds partly invest in residential property, although not as much as they do in, say, offices or retail units, where they are a more important source of investment.7 So, while broader than the mortgage measures themselves, this is another dimension through which our broader macroprudential policy framework is seeking to evolve in line with structural changes in the economy and the financial system.

Finally, let me finish off, by underlining the importance of communication, accountability and transparency. The mortgage measures are important policy interventions. We are very conscious that they have very direct implications, with both benefits and costs for society as a whole, which we seek to balance. So it is critical that we are actively engaging with, and listening to, a broad range of stakeholders and that we are transparent about our judgements and the rationale behind them, so that we can be held accountable. The approach we have taken throughout this review has sought to achieve exactly that. And our commitment to openness and transparency will remain, as we approach the conclusion of our review in the second half of this year.

Thank you.

1Mian, A., Rao, K. and Sufi, A., 2013. Household balance sheets, consumption, and the economic slump. The Quarterly Journal of Economics128(4), pp.1687-1726. Jordà, Ò., Schularick, M. and Taylor, A.M., 2013. When credit bites back. Journal of money, credit and banking45(s2), pp.3-28.

2Bergin, A.K. and Elish Redmond, P., 2020. The labor market in Ireland, 2000–2018. IZA World of Labor.

3Aikman, D., Kelly, R., McCann, F., and Yao, F., 2021. The macroeconomic channels of macroprudential mortgage policies. Central Bank of Ireland, Financial Stability Note, Vol. 2021, No.11.

4See, for example, European Central Bank, 2021, Financial Stability Review.

5Lane, Philip. Determinants of the real interest rate. Address to the National Treasury Management Agency (28 November 2018). Rachel, L. and Smith, T., 2015. Secular drivers of the global real interest rate. Bank of England Staff Working Paper No. 571.

6Carstens, Agustín. The Return of Inflation. Address to the International Center for Monetary and Banking Studies (5 April 2022).

7Macroprudential measures for the property fund sector. Central Bank of Ireland, Consultation Paper 145.