Macroprudential policies to contain systemic risks

Keynote speech by Ignazio Angeloni, Member of the Supervisory Board of the ECB, SUERF/Deutsche Bundesbank/IMFS Conference on “The SSM at 1”, Deutsche Bundesbank, Frankfurt am Main, 4 February 2016

Introduction [1]

I am grateful to the organisers of this conference on “The SSM at 1” for inviting me. The first birthday of the Single Supervisory Mechanism (SSM) was an important milestone, of special significance for those of us who have been involved in the build-up and activity of the new European supervisory body. On 4 November 2015, one year after the operational start of the SSM, the ECB launched a new conference on supervisory issues, whose proceedings are available on the SSM webpages. I am glad to see that the contents and the line-up of speakers today complement those of the conference organised by the ECB well.

Over the years, SUERF has built a tradition of fostering debates on macro, monetary and financial issues near the boundary between research and policymaking. Today’s programme is no exception. I would say that macroprudential policy is a topic that fits well with that tradition. Conceived originally by policymakers (Andrew Crockett is credited for introducing the concept in a 2000 speech [2]), macroprudential policy has recently stimulated a rich stream of academic research which shows no sign of abating. Yet, the ultimate test of success will require this area of policy to become part of the standard toolkit, regularly used alongside monetary policy, micro-supervision and financial regulation more broadly. When and whether this will happen remains to be seen.

With the launch of the SSM, the ECB has acquired macroprudential powers as well, sharing this competence with the national authorities. The modalities of interaction are described in the Capital Requirements Directive IV (CRD IV), the Capital Requirements Regulation (CRR) and the SSM Regulation. Moreover, since 2011 – so well before this – the ECB has hosted the European Systemic Risk Board (ESRB), the pan-EU body responsible for monitoring systemic risks and recommending macroprudential policy measures to the relevant authorities. In this dual role, the ECB has a strong interest, and actively participates, in the research and in the debates on macroprudential policies.

In my remarks today, I will offer some reflections on the present state of, and prospects of development for, this area of policy. I will discuss conceptual and implementation issues, the latter with particular reference to the euro area.

Where do we stand on the use of macroprudential policies?

Let me start with a personal impression, which I hope will not sound too disappointing or pessimistic. It seems to me that, up to now, the interest in macroprudential policies demonstrated by the research community has been much greater than the readiness of policymakers to make actual use of such policies in practice. Now, “interest” and “readiness” are hard to measure, so the word “impression” is in order, but the available evidence is suggestive. Consulting a popular catalogue of economic and financial research, I noted that the number of research papers published on the topic in question rose from less than five per year on average before 2009, to over 700 in 2014 [3]. By contrast, an IMF study shows that the frequency that macroprudential instruments are used around the world merely doubled between 2000 and 2013, also starting from a level close to zero. The more frequent users are emerging market economies, much more so than industrialised countries. [4]

To some extent, one should expect research to precede implementation, but there are likely to be other reasons as well. In this regard, the experience of the United States after the crisis is interesting. The Dodd-Frank Act, which entered into force in 2010, “requires [in the words of the then Chairman Ben Bernanke] that the Federal Reserve and other financial regulatory agencies adopt a so-called macroprudential approach – that is, an approach that supplements traditional supervision and regulation of individual firms or markets with explicit consideration of threats to the stability of the financial system as a whole.” [5] For this purpose, a new Financial Stability Oversight Council was created, chaired by the Secretary of the Treasury and including federal and state regulators. The Council’s mandate is to coordinate analyses and policies to identify and respond to systemic risks. Analytical backing is provided by a newly established Office of Financial Research (OFR), hosted by the US Treasury, which has produced regular reports and analyses since 2012. [6]

This important institutional and analytical build-up has not so far been matched by comparable policy action. In this regard, some remarks made last October by the President of the New York Fed, Bill Dudley, are revealing: “We need to do much more work in developing a coherent macroprudential framework before we start contemplating putting a number of countercyclical measures in place. Such a framework needs to take into consideration how it interacts with other policies [such as microprudential and monetary policy]. When are these policies substitutes? When are they complements? How will they interact? How will the governance work in coordinating across these three realms?” [7] He refers here specifically to countercyclical instruments, but some of his arguments have more general validity.

This experience may depend somewhat on US-specific internal arrangements, but is also paradigmatic of two orders of complexity that macroprudential policymakers have to face. One is intrinsic, inherent in the way the goals and instruments of this policy interact. The second is institutional, having to do with the way in which decision-making takes place. The first level of complexity is general, common to all countries. The second depends on the features of each jurisdiction.

Let me briefly comment on both, with specific reference, for the second, to the euro area.

Some conceptual issues

One may cast macroprudential policy within the familiar target-instrument scheme proposed by Tinbergen and Theil in the 1950s and 60s. In its simplest version, this scheme requires that there be as many independent instruments as there are targets. Independent means that instruments must be distinct and their effect on the goals (their “transmission”) sufficiently different. [8]

In this scheme, the separate identification of macro- and micro-prudential policy instruments is matched by a distinction, on the target side, between “safety and soundness” of individual financial institutions and “systemic stability”. Broadly speaking, the first implies that the risk taken by individual banks is measured and properly internalised; [9] the second that the system be stable, also taking into consideration the interactions and feedback in the financial sector and with the rest of the economy. [10] In this scheme, micro-prudential policy is supposed to take care of the first goal (safety and soundness), and macro-prudential policy the second (financial stability); monetary policy remains focused on price stability. [11]

This scheme has the merit of simplicity and of emphasising the distinction between traditional micro-supervision and macroprudential concerns. A broader instrument space may also be attractive if it helps alleviate policy trade-offs, for example between price stability and financial stability, which may arise especially at low levels of inflation and interest rates. Unfortunately, however, putting the scheme in practice is harder than it seems. First, distinguishing between “safe and sound banks” and “stable system” is not straightforward. Theoretical distinctions based on different types of externalities have been proposed, [12] but they are not easily operationalised. Second, macro- and microprudential instruments tend to coincide. Macroprudential instruments normally take the form of capital surcharges, liquidity requirements and other balance sheet restrictions, precisely the kind of tools employed by the micro-supervisor. Not only does this weaken the requirement that the instruments be distinct, but using the same instruments for different objectives may give rise to policy conflicts.

An alternative to the scheme just described would be to posit the recourse to two instruments, a prudential policy and monetary policy. While monetary policy remains geared to price stability, prudential policy uses the full range of supervisory instruments (pillar 1 and pillar 2 requirements, including macroprudential buffers) to pursue a broad notion of stability, including individual bank soundness as well as systemic stability. Conflicts between micro- and macroprudential goals over the cycle are addressed by setting appropriate countercyclical buffers. In essence, in this model the two authorities, if separate, must coordinate closely so as to attain jointly a broadly defined notion of financial stability. [13] This evidently requires appropriate institutional arrangements that facilitate timely information exchanges, unity of purpose and coordinated action.

This is where the institutional dimension comes in. Let me now move on to this, with specific focus on how Europe’s arrangements respond to the need for coordination.

Institutional features in Europe

In must be noted at the outset that Europe’s institutional framework adds a further element of complexity to those already considered: the presence of national and area-wide authorities. [14] Accordingly, the institutional set-up is built on two layers, at central (EU or euro area) and national level.

At the EU level, the European Systemic Risk Board (ESRB) has the mandate to analyse systemic risks in the entire EU financial system, covering both the bank and non-bank sector. It is a highly representative body (its General Board includes representatives of national central banks and regulators of all EU Member States, plus the ECB, the Commission and the ESAs, for a total of 95 members (of which 37 have voting right). Since its start, the ESRB has developed a rather extensive structure, with a large number of technical and research sub-groups to analyse various dimensions and sources of systemic risk. Notably, alongside an Advisory Technical Committee of central bank and supervisory officials, a large Advisory Scientific Committee, composed of outside experts and academics, prepares reports (including for publication) and is represented on the General Board with voting power. The ESRB, however, is not a direct decision-maker – it is only entitled to issue risk warnings and policy recommendations to the relevant authorities, national and European. In some of these dimensions, the ESRB is analogous to the FSOC in the United States.

In the euro area, following the entry into force of the Regulation establishing the SSM in November 2014, the ECB and national authorities share powers over the macroprudential toolkit provided for in the CRD IV and the CRR (including delegated acts issued by the Commission). This covers a number of lender-based tools, mainly on capital, liquidity and net stable funding. [15] However, it excludes borrower-based measures such as loan-to-value or loan-to-income limits, which act on the demand side; these instruments remain under the exclusive control of the national authorities. [16] The sharing of responsibility between the two levels is in the fact that the ECB can only tighten the policy stance, using the tools provided for in EU law, relative to what is determined by the national authorities. With this arrangement, the legislator has wanted to obviate a perceived tendency to inaction by national authorities, presumably deriving from cross-border externalities.

Within the ECB, pursuant to the EU Treaties all decisions are ultimately taken by the Governing Council, the same body that sets monetary policy. The SSM Supervisory Board (which includes heads of supervision and central bankers from all participating countries plus ECB representatives) prepares complete decisions in the microprudential area, submitted to the Governing Council for final adoption through a short non-objection procedure. In addition, the Supervisory Board can launch the adoption of macroprudential measures, within the limits of the ECB powers as described; complete proposals in this area are also adopted by the Governing Council. [17] In addition, a number of internal structures are involved to bring in the micro and macroprudential perspectives, both within the ECB and in the SSM countries, and to cater for the necessary coordination. In particular:

  • a Financial Stability Committee, involving banking supervisors and central bank representatives from SSM countries at a technical level, carries out analyses in the macroprudential field;
  • a Macroprudential Coordination Group supports intra-ECB cooperation between the micro- and macroprudential function of the ECB;
  • finally, a Macroprudential Forum, including all Governing Council and Supervisory Board members, operates as a platform for regular discussion at the most senior level.

This multiple interaction is new and still being tested in practice. It assembles all available information and expertise available at the ECB and among euro area authorities. But it is evident that all this comes at the cost of considerable complexity.

There are several consequences of this. First of all, up to six different institutions need to be involved for a decision by a Member State to be finalised (art. 458 CRR): the national authority, the ECB, the ESRB and the EBA, the European Commission and the European Council. Although short deadlines are provided for each step, the overall process can be cumbersome and lengthy.

In addition, the overlap of powers between the European and national levels may give rise to different goals. In the euro area, the instruments available for domestic macroeconomic management are often perceived to be too limited. There is a risk, in this situation, that macroprudential tools may be used as substitutes for “missing” instruments, notably monetary policy. For example, limits on bank credit (in the form of capital surcharges) could be imposed at the national level to compensate for a single monetary policy that is perceived to be overly expansionary in relation to national conditions. This could give rise to distortions in credit markets and fuel the development of the unregulated sector. [18]

Furthermore, the institutional complexity can lead to inaction bias or to the possible use of instruments that are sub-optimal for the purpose. According to a recent EBA report, [19] macroprudential decisions under CRDIV/the CRR have been heavily affected by governance and legal requirements. The EBA report contains examples in which different measures were used to address similar risks (for instance on real estate), probably influenced not by the fit of the instrument to the purpose, but by its governance.

Recent policy moves in the euro area

Let me now briefly mention some of the macroprudential measures that have been adopted in the euro area since the new rules of the CRDIV/CRR came into force in 2014. The purpose is not to give a full picture, but only a first indication of how the relevant authorities are beginning to move in this area. [20]

The new legislation required, first of all, that each Member State identifies the domestic authority in charge of macroprudential policy. All of them have done so, but with specific arrangements in each country, as already mentioned. Moreover, the law requires Member States to adopt decisions concerning specific instruments, within limits set by the law itself. I will focus on the two most important ones: the systemic risk buffers and the countercyclical capital buffer.

Buffers on systemically important banking institutions include a general systemic risk buffer (potentially applied to a country’s whole financial sector or a subset of it), as well as bank-specific buffers applied to globally important institutions (G-SIIs) and other systemically important institutions (O-SIIs). The determination of the former is done by applying a methodology established by the Basel Committee on Banking Supervision (BCBS) and implemented in Europe by the Commission, while for the latter, the choice of institutions and the extent of the buffer are up to the national authorities themselves. These capital surcharges are aimed at mitigating the “too big to fail” issue. Buffers should take into account each institution’s contribution to systemic risk which is assessed by their size, cross-border activity, interconnectedness, complexity and so on.

At the present time, only two SSM participating Member States have activated a general systemic risk buffer: Austria and the Netherlands. Slovakia will introduce it in in 2018. By comparison, among other EU countries, Sweden has adopted it and the UK has announced the launch of a public consultation on it.

For the G-SII, a gradual phase-in is foreseen until end-2018. [21] Once fully loaded, the buffer ranges between 1% and 2.5% (an upper bucket of 3.5% is not used at present). [22]

The range of buffers applied to O-SIIs varies considerably, also for banks that are comparable in size and business models. For instance, in Spain this buffer has been set at up to 1% (after the phasing-in period) on the major lenders, while in the Netherlands the corresponding value is at its top level of 2%. Italy has set a buffer of 0% on all its designated O-SII. Some countries, such as Lithuania, Portugal and Slovenia, have determined that buffers only apply to all banks from 2017 onwards, while in other countries they are already applied in 2016. There is also a wide variety in the number of banks designated as O-SII, even among countries with comparable size and banking structures.

The countercyclical capital buffer is meant to induce banks to build-up capital during periods of high aggregate credit growth associated with the build-up of systemic risk. No use has been made of this instrument among SSM members: all countries have set the relevant parameters at 0%. By comparison, Sweden has introduced it and in the UK the Financial Policy Committee has recently mentioned the possibility of introducing it in the future.

So far the ECB has decided not to exercise its macroprudential powers, not objecting to the measures adopted by the national authorities, based on an assessment of financial stability conditions. The initiative therefore has remained, so far, exclusively in national hands.

My brief summary of the measures adopted so far under the new legal framework ends here. I will not attempt to analyse their determinants or pass any judgement on their overall consistency. It is probably premature to do so. But it will be important, at some stage, to analyse them closely not only from the viewpoint of their overall coherence for systemic stability, but also in order to identify the scope of possible action by the ECB.

Conclusions

Let me conclude with three statements.

The novelty and importance of the macro-prudential approach has long been recognised, and rightly so. Systemic risks and the related policies merit the close and active attention of supervisors and regulators. Prudential policies must go beyond traditional supervision. What is in question is not the validity of the approach, but how to apply it in practice.

We should resist the temptation to regard macroprudential policy as a separate compartment in the policy construct, institutionally and operationally disconnected. The policy instruments available and the nature of the goals involved do not allow such rigid separation. Any attempt in this direction may fuel policy conflict and take macroprudential policy away from its mission, which is to promote financial stability.

In my view, we should work towards developing among regulators and supervisors a broader concept of banking and financial stability, encompassing those transmission links. The full range of available instruments should be used to attain that goal. As fittingly suggested by André Sapir in a recent debate, the macroprudential approach is a “state of mind”: an operational awareness from those in charge of regulating and supervising finance that to promote a stable financial sector, all relevant macroeconomic interconnections must be accounted for. The challenge in doing so is to assemble expertise and judgement from different sources and, where different decision-makers are involved, to closely coordinate them in the pursuit of that goal.

This is a complex task, but a feasible one. More analyses, debates and experience are needed to get there. This is why conferences like this are so helpful.

Thank you for your attention.



[1]I am grateful to Frank Dierick, Astrid Farrugia, Skander van den Heuvel, Giulio Nicoletti, Adam Pawlikowski, Evangelia Rentzou and Stephanie Stolz for helpful contributions and comments. This speech includes some material presented in a recent conference hosted by Bruegel (streaming available at http://bruegel.org/events/monetary-policy-after-the-great-recession/. The views expressed here are personal and should not be attributed to the ECB.

[2] See Crockett, A., “Marrying the micro- and macro-prudential dimensions of financial stability”; speech at the Eleventh International Conference of Banking Supervisors, Basel, September 2000. However, much earlier uses of this word have been reported by Clement, P. (2010). “The term "macroprudential": origins and evolution”. BIS Quarterly Review, March.

[3]IDEAS search engine; see https://ideas.repec.org/search.html.

[4]See Cerutti, E., Claessens, S. and L. Laeven, “The use and effectiveness of macroprudential policies: new evidence”, IMF Working Paper n. 61, March 2015.

[5]Bernanke, B., “Implementing a Macroprudential Approach to Supervision and Regulation”, speech at the 47th Annual Conference on Bank Structure and Competition, Chicago, Illinois, May 5, 2011.

[6]More details of the macroprudential institutional arrangements in the United States under Dodd-Frank are provided in the Annex.

[7]Dudley, W., “Is the active use of macroprudential tools institutionally realistic?”, panel remarks at the Macroprudential Monetary Policy Conference, Federal Reserve Bank of Boston, October 2015. Dudley’s remarks are partly based on a simulation exercise (“table-top experiment”), conducted within the Federal Reserve System in 2015; see the Annex for details.

[8]If these conditions are fulfilled, then the targets are, in principle, attainable. In practice, the ease with which they are attained depends on the strength and reliability of the transmission and on a correct assignment of instruments to targets, possibly including coordination schemes among different policies.

[9]The concept of safety and soundness of banks alongside that of systemic stability is enshrined in supervisory practices by the Basel Committee on Banking Supervision’s Core Principles for Effective Banking Supervision, issued in September 2012. Principle 1 states, inter alia, “The primary objective of banking supervision is to promote the safety and soundness of banks and the banking system. If the banking supervisor is assigned broader responsibilities, these are subordinate to the primary objective and do not conflict with it”.

[10]The ECB defines financial stability as a condition in which the financial system – intermediaries, markets and market infrastructures – can withstand shocks without major disruption in financial intermediation and in the general supply of financial services. See for example ECB Financial Stability Review, November 2015.

[11]For an illustration of this scheme, see, for example, the introductory chapter of Schoenmaker, D., Macroprudentialism, December 2014 (http://www.voxeu.org/article/macroprudentialism-new-vox-ebook ).

[12]For example, Hanson S. G., Kashyap A. K and Stein J. C., (2011), “A Macroprudential Approach to Financial Regulation”, Journal of Economic Perspectives, 25(1), pp. 3-28, emphasise the distinction between “safety net” externalities (the control of which falls in the remit of the micro-supervisor) and “network” or “systemic”externalities (which fall under the responsibility of the macroprudential authority).

[13]Coordination problems are greatly simplified when policy functions are concentrated in a single institution, albeit perhaps with some internal assignment criteria or firewalls. This is the case, for example, with the Bank of England, which houses both the micro- and macroprudential authorities (PRA and FPC respectively) and the monetary policy function (MPC). Recently the Bank of England has been comparatively active in this area, announcing a macroprudential tightening of lending standards to the real estate sector, announcing the future activation of the countercyclical capital buffer and last week introducing a systemic risk buffer for ringfenced banks beyond a given size threshold.

[14]To a lesser extent, this problem exists also in the United States between federal and state authorities; see the aforementioned speech by Dudley (see Annex).

[15]Pillar 2 requirements are have also been used in a macroprudential context in some cases; see some examples in EBA, “On the range of practices regarding macroprudential policy measures communicated to the EBA”; July 2015.

[16]There is evidence that borrower-based instruments are the most effective to curb excessive credit growth – see Hartmann P., “Real Estate Markets and macroprudential policy in Europe”, ECB Working Paper Series, No 1796 (May 2015).

[17]The Governing Council can also initiate the process to adopt macroprudential measures. In this case, it has to involve the Supervisory Board to prepare the decision, which is then re-submitted back to the Governing Council for adoption.

[18]A parallel that comes to my mind is with the imposition of credit guidelines or ceilings in the 1970s and 1980s. These were used, in some European countries, to maintain an expansionary monetary policy while mitigating undesired effects on inflation and the exchange rate. That experience was ultimately unsuccessful, and was followed by a generalised removal of credit controls in the course of the 1980s.

[19]See ESRB, “A review of macro-prudential policy in the EU one year after the introduction of the CRD/CRR”, June 2015.

[20]A full list of country notifications is available at http://www.esrb.europa.eu/mppa/html/index.en.html.

[21]See CRD IV article 162(5).

[22]See http://www.fsb.org/wp-content/uploads/2015-update-of-list-of-global-systemically-important-banks-G-SIBs.pdf.

Media contacts