Luigi Federico Signorini: Macroprudential policy - effectiveness, interactions and spillovers

Welcome address by Mr Luigi Federico Signorini, Deputy Governor of the Bank of Italy, at the fourth annual Macroprudential Policy Group research workshop, Rome, 10 October 2019.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
10 October 2019

It is my pleasure to welcome you to the fourth Macroprudential Policy Group research workshop. We are honoured to host an event that has become one of the most important opportunities to discuss macroprudential policy in Europe. I would like to thank the organisers, the keynote speaker, the contributors to the four sessions and the panellists for the roundtable that will conclude the conference this afternoon.

The papers to be presented today cover a wide array of topics and will help us to understand crucial features of the transmission mechanism of macroprudential policies. They take stock of the experience accumulated in recent years across the euro area. They also strive to contribute to the development of a measure of the prudential stance, an endeavour that I have always advocated as essential for the conduct of macroprudential policy. Let me spend a few words on this point.

The concept of stance should convey information on the overall impact of the policy instruments that have been implemented, on their adequacy in meeting policy objectives given the identified risks, and on the required policy orientation. A credible stance would help to enhance policy communication, reinforce transmission channels and counter potential inaction biases.

Certain steps towards setting a framework for a macroprudential stance in the EU have been taken. A notable contribution is a recent European Systemic Risk Board publication, which proposes a "risk-resilience" framework to define the macroprudential stance. In that framework, the amount of systemic risk faced by the financial sector is compared to the level of resilience against negative shocks stemming from microprudential provisions, regulatory aspects or public safety nets, and the additional resilience created by the active macroprudential tools. The macroprudential stance is defined as loose or tight depending on whether the difference between risk and resilience is above or below some  pre-specified level that measures the risk tolerance of the authorities, and thus, the "neutral" policy stance.

This is an interesting framework and I certainly welcome it as a good starting point for a policy discussion. It is a high-level framework, however, and certain concepts need further development. Without any claim to being exhaustive, I would like to mention three important dimensions along which I think the approach could be further improved.

First, the boundary between the effects of micro and macroprudential policies is more artificial than real. The two, while different in purpose, share instruments and are directed to the same participants in the financial markets. Indeed, it is the overall level of micro and macroprudential requirements that affects the ability of the financial system to finance the real economy, both in normal and stressed times. Consequently, a narrow notion of the macroprudential stance should, in my opinion, be replaced with the more comprehensive one of an overall prudential stance, one that would account for how both categories of prudential measures affect the economy. Let me emphasise that this is not just a question of semantics. Microprudential capital requirements, for instance, have been substantially increased after the crisis. In the short run, the contractionary effects on the credit cycle associated with such an increase were, directionally and qualitatively, of the same nature as those associated with the activation of a cyclical buffer. This of course does not mean that it was wrong to do it; an increase in the quantity and quality of bank capital was required to strengthen the banking system, in light of all the shortcomings of the old supervisory rules that the crisis had exposed. However, it was clear to all that there was a price to pay for this, in terms of tighter credit: at least temporarily, during the transition to the new standards. An overall prudential stance measure would capture this policy tightening; a narrower perspective misses it. Had the former existed, it would have supplied valuable input, for instance, to the discussion on transitional arrangements. The discussion took place anyway, because regulators were qualitatively aware of the issue, but in a sort of informational vacuum.

I would thus suggest to devote further thought to such issues of definition.

The conceptual framework should also more explicitly capture the idea that one central aim of macroprudential policies is to limit the endogenous pro-cyclicality of the financial sector. Risks tend to be underestimated by market participants during booms and overestimated in busts, resulting in an amplification of cycles. To lean against this procyclical pattern in risk perceptions, the framework for the macroprudential stance should have a clear cyclical perspective. One might also note that, in normal macro-economic discourse, "neutral" is a factual assessment of the stance and is not synonymous with "right"; in certain cases policymakers wish to have a loose, in other cases a tight stance. One way to set this idea within the ESRB conceptual framework might be to say that the macroprudential authorities would need to lower their risk tolerance in booms and thus activate  contractionary policies early; symmetrically, they would increase their risk tolerance as booms turn to busts, and thus provide accommodative policy and support growth when the cycle weakens. But whatever the specific phrasing, a more prominent cyclical element would, in my view, be welcome.

The use of macroprudential tools, especially the more wide-ranging ones such as the countercyclical buffer, should always be carefully considered in this light. Timing is essential. We may or may not be able one day to fine-tune the prudential stance so as to define an "optimal policy rule" over the cycle but, as a minimum, we should be alert to the risk that the countercyclical buffer may end up to be used like a procyclical one. "Better late than never" is a poor guide to countercyclical action; the Hippocratic injunction to physicians, "First, do no harm", is more apt. An objective evaluation of the stance, and a cyclical benchmark against which to measure it, would be most useful in this respect. This is an important reason why I look forward to any progress in this field, whether we can ever fine-tune macroprudential policy or not.

My third and final point on conceptual issues is that, ideally, the prudential stance should also account for the interaction of macroprudential policies across countries. Euro-area countries are deeply integrated. During stress periods, this may lead financial instability in one country to propagate to other countries, if an inaction bias prevails; the same holds, again symmetrically, for the effect of contractionary measures. How, and to what extent, this can be done is, I think, a matter for further potentially valuable reflection.

Before concluding, let me add that a different issue is sometimes raised about the relationship between monetary and macroprudential policy. This is an important and complex issue and it would not be possible for me to discuss it here at any meaningful length. Let me however make one observation that applies especially to tailored macroprudential policies, i.e., the ones that are aimed at preventing instability in specific markets. Such measures, I would argue, are best seen as working, not as a restraint on, but as a complement to, monetary policy. The latter is, by nature, pervasive, and as such it can have unwanted side-effects in certain areas or markets. Should such pockets of vulnerability emerge, authorities can activate tailored macroprudential policies to tackle them. In this context, macroprudential measures do not aim at "neutralizing" the effects of monetary policy. Rather, a targeted use of macroprudential tools creates more room for manoeuvre for monetary policy to follow its price-stability mandate. Recently, various euro-area countries have adopted macro measures to contain imbalances in the real estate sector, sometimes in response to the ESRB's warnings or recommendations. This is, in my view, a successful example of how monetary and macroprudential policies can complement each other.

*    *    *

It is time for me to close this address. I look forward to the valuable exchange of experiences, insights and evidence that will take place during this conference. I am sure it will help us improve our understanding of the way macroprudential tools can be best used to satisfy our financial stability mandate, and to complement other macroeconomic tools. I thank you all for your participation and your attention.