Towards a macro-prudential framework for the single supervisory area
Remarks by Ignazio Angeloni, Member of the Supervisory Board of the European Central Bank,
Belgium Financial Forum,
Brussels, 20 April 2015
I am very happy to be here this morning to address the Belgium Financial Forum.
This is the first event sponsored by the National Bank of Belgium held since Jan Smets was appointed Governor. This has turned out to be a win-win situation for the ECB: the Governing Council has gained a valuable new member and the Supervisory Board has been able to draw on the support of the outgoing Governor, Luc Coene, as new member. After just a few weeks, the Board has already benefited significantly from Luc’s presence. I am especially glad and honoured to have him as a colleague.
My focus today is on the macro-prudential framework of the Single Supervisory Mechanism (SSM) area. I have chosen this topic for three reasons. First, while being a cornerstone of Europe’s new financial architecture, macro-prudential policy is still only partially explored and understood. Second, it is analytically complex, spanning macroeconomics, banking and finance; as such, it matches the strong research orientation of the National Bank of Belgium. The website of the Bank is rich in research contributions, many of which are precisely on this topic. Third, in Europe macro-prudential policy measures have recently been enacted predominantly by small-medium sized countries. Belgium itself has adopted measures of this kind recently.
If I had to choose two words to characterise the prospect of fruitfully using macro-prudential policy in the SSM area, I would pick the following: promising and difficult. It is promising because, as experience shows, a macro-prudential perspective is necessary to understand financial instability and how to deal with it. One cannot gain a proper understanding of the “safety and soundness” of a bank if the links between that bank, other banks and the broader economic and financial environment are overlooked. But it is also difficult, partly because some of the concepts are not yet sufficiently developed. In addition, the use of macro-prudential policy is made more complex in Europe by the simultaneous presence of national and area-wide policy actors.
I will first provide a short overview of the concepts, drawing from recent research. Then I will look at some recent concrete experiences. Finally, I will focus on the institutional landscape and describe how the ECB is organising itself to perform its role as macro-prudential policy-maker.
Macro-prudential policy: origins and rationales 
Two main lessons can be drawn from the recent crises.
First, the stability of individual financial institutions does not ensure the stability of the system as a whole. During booms, individual banks typically look healthy and well capitalised. Credit grows, asset prices rise and their volatility is low. In those phases, risk tends to be underestimated by market participants. Risk indicators only surge at a later stage, when risks are reassessed. Micro-prudential supervision, with its predominant focus on individual banks, alone cannot easily detect incipient systemic risks, because they derive from interconnections with other banks and with the economy. A more comprehensive approach is necessary.
The second lesson is that price stability alone does not ensure financial stability.  In the early 2000s, during the period often referred to as the “Great Moderation”, inflation was low and stable; yet financial risks were rising, in the form of leveraged balance sheets and asset price misalignments. Price stability was not the cause of the financial vulnerabilities, but did not prevent their development. How could that happen? Views differ on this. One reason is that the financial and business cycles are different;  I shall return to this point later. While aiming at price stability, monetary policy may, inadvertently and unwillingly, permit or even facilitate a build-up of risks in the financial sector.
These arguments can be framed within the well-known scheme connecting policy instruments and goals, introduced by Tinbergen and Theil in the 1950s.  That scheme predicts that two objectives, in this case price stability and financial stability, cannot be achieved by a single instrument, monetary policy. Moreover, one needs to distinguish between financial stability and its micro-prudential analogue, the “safety and soundness” of individual banks. The former neither implies nor requires the latter. One can have a few non-systemic, unsound banks in a stable system; if they fail, as they should, the stability of the system is not jeopardised. Conversely, a system composed of banks that are all individually “safe and sound” can become unstable when interactions among themselves, or with the financial markets, or with the real economy are taken into account. Ensuring the safety and soundness of each bank is the recognised task of micro-supervisors, while financial stability is what macro-prudential policy is supposed to achieve.
One may be tempted to conclude, following the Theil-Tinbergen logic, that the extra instrument, macro-prudential policy, allows achieving the extra goal, financial stability. The picture is more complicated than that, however. To understand, let’s consider more in detail the relations between macro-prudential policy on the one hand, and micro-prudential and monetary policy on the other.
Micro- vs. macro-prudential policy
To start with, the meaning of “safe and sound” needs to be clarified.  For sure, it does not mean riskless. Banks are risky by definition; they cannot conduct their business otherwise.  I think those terms mean, loosely speaking, that the risks borne by the taxpayer and by the creditors of the bank are appropriately contained and transparently disclosed. The exact extent and distribution of those risks, however, needs to be determined more precisely. 
Hanson, Kashyap and Stein  have proposed a definition of the goals of micro- vs. macro-prudential supervision, assuming, in accordance with neoclassical logic, that public policy intervenes only to correct market failures. In their view, micro-prudential policy should correct the distortion towards risk taking created by the safety net. Conversely, macro-prudential policy is meant to correct for other market failures, also giving rise to undue risk taking, generated by “systemic externalities”.
Systemic externalities can arise in two forms: from strategic complementarities and from interconnectedness.  The former are due to financial institutions’ tendency to adopt similar investment strategies. During upturns, exposure to similar credit and liquidity risks lead to the accumulation of financial imbalances. Similarly, during downturns financial institutions tend to reduce the size of their balance sheets by selling similar assets, leading to fire sales that generate funding stress and credit crunches. Externalities related to interconnectedness refer instead to reciprocal exposures among credit institutions.  The two types of externality tend to reinforce each other; for example, strong bank interconnections increase the risk of fire sales.
While the goals of macro- and micro-prudential policy are clearly distinct, their instruments tend to coincide. There is virtually no instrument commonly regarded as macro-prudential that cannot, in essence, be used also by the micro-prudential supervisor as part of its supervisory evaluation and intervention process. Macro-prudential instruments normally take the form of capital surcharges, liquidity requirements and other balance sheet restrictions. It is the logic of their use and the scope of application that are different. In particular, if an instrument is applied selectively to individual institutions, it is often considered micro, whereas if it is applied to groups of banks or to all banks in a country, it is macro.
We seem, therefore, to be facing a situation in which the number of objectives is larger than that of the instruments.  In this case, a tension between policy objectives arises. During upturns, macro-prudential authorities can decide to increase capital requirements; in the same context, micro-supervisors may wish to do the opposite, as risks to individual banks seem to be receding. A balance needs to be found in this trade-off.
A suitable empirical basis is essential here. To guide macro-prudential policy actions in a pre-emptive way it is essential to have reliable leading indicators of systemic risk. A vast literature on early warning indicators has developed. All in all, econometric estimates and case studies concur in suggesting that credit growth in excess of long-run norms, asset price booms and real estate market buoyancy, especially if manifesting themselves jointly, indicate impending systemic risks. 
Macro-prudential vs. monetary policy
Monetary policy, aiming at maintaining price stability by means of central bank open market and refinancing operations, is clearly distinct from macro-prudential policy in terms of both goals and instruments.  However, there are interactions between the two policies, as some recent literature has emphasised. In particular, there is evidence that the monetary policy stance influences the risk-taking behaviour of investors and financial institutions. When monetary policy remains accommodative for a protracted period, banks can seek more risk in their balance sheet, either by extending credit to riskier borrowers or taking up more leverage, or both.  According to some authors, market participants may also perceive that central banks are implicitly committed to support asset prices in case of a slump; anticipating that, market participants may engage in excessive risk-taking. 
The interaction between macro-prudential and monetary policy can give rise to complementarities and tensions.  In particular, macro-prudential policy can support monetary policy when the latter is constrained or overburdened. In the euro area, this may be perceived as a convenient avenue to recuperate degrees of national autonomy. But there are limits to that possibility, because credit and business cycles have different frequencies and magnitudes. As noted in several BIS papers recently,  the credit cycle tends to be longer, typically lasting for more than one standard business cycle. 
Specific European features
The argument so far was implicitly referred to a standard institutional and economic set-up, consisting of a single integrated economy with a central bank in charge of monetary policy and authorities responsible, separately or in combination, for micro-prudential supervision and macro-prudential policy.
Does Europe have this standard configuration? Yes, in many respects. For instance, the euro area has a single independent central bank conducting monetary policy. And since last November there has been a single independent supervisor, distinct from but closely connected to the central bank, possessing all the standard supervisory instruments and guided by a clear and strong charter – the SSM Regulation. In addition, in 2011 the EU set up a specific body, the European Systemic Risk Board – ESRB, analogous to the Financial Stability Oversight Council in the United States – providing macro-prudential policy analyses and recommendations. This body is responsible for conducting macro-prudential oversight of the EU’s financial system, which includes banks, other financial institutions beyond banks, financial markets and financial market infrastructure. Its broad scope is particularly relevant given the growing role of shadow banking and their interconnections to banks and the real economy.
Looking deeper, however, there are differences. Historically, banking regulation in Europe has been fragmented – though has become less so recently, after the creation of the European Supervisory Authorities and especially the European Banking Authority. Nonetheless, banking markets remain fragmented in many respects. There is a strong home-bias in bank portfolios, accentuated by the fact that large exposure rules exempt sovereign bonds. Retail banking remains predominantly domestic; only the interbank lending market has a real cross-border reach. During the crisis we have seen a re-nationalisation of many market segments, a process which has not yet been fully reversed.
But in a crisis, national financial clusters are not disconnected. Cross-border contagion in the euro area can be substantial, as we saw especially in 2011. Risks that are initially confined in individual countries can become systemic for other countries. The bank-sovereign nexus starts at national level, but can propagate area-wide. Therefore, in some circumstances, country-specific risks are better addressed at an early stage by national macro-prudential measures, hence helping prevent systemic instability for the area. This justifies the fact that, in the euro area financial architecture, macro-prudential policy is a shared competence between national and European authorities. The two components complement each other and need to be properly coordinated to avoid inconsistency.
Before turning to such coordination procedures, let me highlight some stylised facts regarding the recent experience in using macro-prudential instruments in Europe.  For further detail, see Annex 1. 
In 2014, i.e. during the year following the entry into force of the Capital Markets Directive IV (No 2013/36/EU, or CRDIV) and Regulation (No 575/2013, or CRR), the body of European law that provided a legal framework to macro-prudential policy, there were a total of 86 policy measures adopted by the 28 EU Member States, of which 37 were in the euro area (18 countries in 2014). The ESRB classifies a number of them as purely “administrative”, i.e. not expected to have a substantial impact.  Excluding those, there remain 47 measures in the EU (enacted by 15 countries) and 19 in the euro area (enacted by 7 countries).
Three observations emerge from this (still limited) experience. First, the frequency of use of those instruments is rather high. The data I have just mentioned imply an average of about three policy moves per year, for each country; a high number that however may reflect in part a backlog from the previous period in which the legal framework was less defined. Second, the countries outside the euro area have moved more frequently. This evidence would deserve more scrutiny; it could be due in part to the different position in the cycle of the euro area relative to the other EU member states. Third, most of these measures were adopted by small or medium-sized countries. The four largest countries in the euro area (Germany, France, Italy and Spain) made no use at all of macro-prudential instruments in 2014. In the EU as a whole, the average size of the countries that have activated the instruments in question is, roughly, close to that of Belgium.
The evidence shows that two main types of measure have been used.
Several Member States have introduced measures to address excessive credit growth, mainly related to mortgage lending. To deal with mortgage lending developments, Member States have often resorted to risk weights and to quantitative restrictions on lending standards, such caps on the loan-to-value (LTV) ratio. For example, the Belgian authorities have recently increased risk weights on mortgages granted by banks which use the internal ratings-based approach to levels similar to those of their peers. To take another example, the Central Bank of Ireland has decided to place ceilings on the proportion of mortgage lending at high LTV ratios and at high loan-to-income (LTI) ratios.
A number of other countries have taken steps to deal with the “too big to fail” issue, by means of systemic risk buffers (SRBs) and buffers for other systemically important institutions (O-SIIs). Such buffers, which address vulnerabilities related to size, concentration, common exposures etc., are primarily targeted at major banking groups and the rates are often differentiated according to each institution’s contribution to systemic risk. For instance, in 2014 Croatia, the Czech Republic, Denmark, the Netherlands and Sweden announced plans to activate either SRBs or O-SII buffers.
Coordination procedures in the SSM Regulation 
The CRD IV/CRR text mentions a fairly large number of macro-prudential instruments. The table of Annex 2 groups them in eight categories. Among them, the most frequently used are the counter-cyclical capital buffers, the systemic risk buffers, and the buffers for global systemically important institutions (G-SII) and other systemically important institutions (O-SII).
The SSM Regulation, while establishing a new banking supervisory authority within the ECB, also confers on it macro-prudential policy powers (see the relevant passages in annex 3). In particular, the Regulation empowers the ECB to activate such instruments, if included in EU law, but only to tighten them up, not loosen them. National authorities, though, can activate all measures, in both directions.  National authorities must consult the ECB before action is taken, and vice-versa. Specifically, if a national authority plans to introduce certain macro-prudential measures, it will have to notify the ECB about its intention ten days before a decision is taken. The ECB, in turn, can object within five days. Any objection should be addressed to the notifying authority.
Considering in particular that there is a degree of substitutability, not easily quantifiable, between the instruments which are included in the EU legal texts and those which remain under exclusive national control, such as caps on loan-to-value or loan-to-income ratios, there is a risk of one authority potentially offsetting measures taken by another. As a result, distortions can arise. In addition, macro-prudential policy instruments, especially if employed by countries that are small and relatively open, can easily have an impact across borders. This provides an argument for broad and systematic coordination of such moves between the national authorities and the ECB regarding the use of all available instruments.
At the ECB, the ultimate decision-maker regarding the activation of macro-prudential policies is the Governing Council, acting on a draft decision submitted by the Supervisory Board. In practice, the Governing Council interacts closely with the Supervisory Board, usually to strict deadlines. An effort is being made to combine micro-prudential and systemic considerations, against the background of the broader macro-financial situation. Every quarter, the Governing Council and the Supervisory Board convene in joint sessions to examine the macro-prudential situation. The Governing Council can also request the Supervisory Board to submit a proposal or to undertake studies concerning specific sources of vulnerabilities. At the ECB a Macro-Prudential Coordination Group has been established, comprising Board members and staff with the relevant expertise.
With reference to the EU as a whole, the ESRB, which brings together central banks, national supervisors and relevant EU institutions, promotes analyses of macro-prudential policies and can address policy recommendations to the relevant authorities.
I will now end my review with some tentative conclusions and suggestions for further reflection.
As I mentioned at the outset, macro-prudential policies open up attractive prospects but their use is complex. The overlap among policy areas, especially with micro-prudential supervision, the difficulty of drawing clear boundaries between micro- and macro-prudential instruments, and the likely occurrence of significant spill-overs suggest that more analysis and experience are needed. Certain specific features of the euro area are likely to compound those complexities further.
Specifically, the task of economic researchers is far from concluded. Theory and empirical analyses can hopefully tell us more, in the not too distant future, about how macro-prudential instruments interact with micro-prudential regulation and supervision as well as with monetary policy.
While all this is happening, there is a risk of an over-reliance on macro-prudential instruments, in two ways: first, they can be improperly used for different purposes (for example, to fine-tune the business cycle, especially in countries where other policy levers are lacking or overburdened); second, they may be activated without full awareness of their cross-border impact.
The above considerations suggest that active consultation and coordination is needed in the euro area when pondering the use of such instruments. If not binding, such consultation and coordination should at least be challenging. A checklist of necessary conditions for acting may include the following: (i) systemic nature of the risk involved; (ii) absence or relevant externalities, or ways to deal with them; (iii) impossibility or inadvisability of dealing with the underlying problem with other policy instruments; (iv) consistency with conditions and policies in non-euro area countries. The latter could be subject to the judgement of the ESRB.
The above represents a sizeable work plan. In the interim, the adoption of macro-prudential actions should be considered with caution. In this regard it may be apt to apply, or at least keep in mind, the quote attributed to Zoroaster: “When in doubt, abstain”.
Thank you for your attention.
I am grateful to Lorenzo Cappiello and Giuseppe Cappelletti for useful contribution to an early draft, and to Cecile Meys for further support. The views expressed here are personal and should not be attributed to the SSM Supervisory Board.
The notion of macro-prudential policy was introduced by A. Crockett, 2000, “Marrying the micro and macro-prudential dimensions of financial stability,” BIS Speeches, 21 September. It was then developed by economists at the Bank for International Settlements; see, for example, C. Borio and W. White, 2004, “Whither monetary and financial stability? The implications of evolving policy regimes,” in Monetary policy and uncertainty: adapting to a changing economy, Proceedings of symposium sponsored by the Federal Reserve Bank of Kansas City, Jackson Hole, 28-30 August 2003; also available as BIS Working Paper No 147, February 2004.
See, e.g., C. Bean, August 2010, “Monetary Policy after the Fall”, speech given at the Federal Reserve Bank of Kansas City Annual Conference, Jackson Hole, Wyoming.
See C. Borio, “The international monetary and financial system: its Achilles heel and what to do about it”, BIS Working paper n. 456, August 2014.
Tinbergen, J., On the theory of economic policy; Contributions to economic analysis; Amsterdam, North Holland, 1952; Theil, H. Economic forecasts and policy; Contributions to economic analysis; Amsterdam, North Holland, 1958.
The notion was introduced systematically in Basel Committee on Banking Supervision, Core Principles for Effective Banking Supervision, 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”.
D. Diamond and R. Rajan, December 1999, “Liquidity Risk, Liquidity Creation and Financial Fragility: A Theory of Banking”, NBER Working Paper Series No. 7430.
The “socially tolerable” level of banking risks and their allocation to taxpayers and bank creditors are choices that have profound distributional implications; as such they belong more properly to the political sphere.
See 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.
See, for example, G. De Nicolò, G. Favara and L. Ratnovsky, June 2012, “Externalities and Macroprudential Policy”, IMF Staff Discussion Note, SDN/12/05; S. Claessens, December 2014, “An Overview of Macroprudential Policy Tools”, IMF Working Paper, WP/14/214.
IMF, June 2013, “Key Aspects of Macroprudential Policy”; IMF, March 2011, “Macroprudential Policy: An Organizing Framework”.
This is in spite of the fact that the available macro-prudential instruments are many. It is doubtful, however, that all these instruments are independent of each other in the sense of their effects on policy goals being sufficiently different.
See, for example, M. Drehmann, C. Borio, and K. Tsatsaronis, 2011, “Anchoring Countercyclical Capital Buffers: The Role of Credit Aggregates”, Bank for International Settlements; L. Alessi and C. Detken, August 2014, “Identifying excessive credit growth and leverage,” ECB Working Paper Series No 1723. The financial stability area of the ECB regularly uses a set of indicators of this nature to prepare its semi-annual Financial Stability Review. See, for instance, the ECB’s Financial Stability Review, November 2014.
Incidentally, this has not always been true. Until the 1980s monetary policy was conducted in a number of European countries by imposing different forms of administrative constraints on or disincentives for the supply of bank credit. Some of the more modern macro-prudential instruments are in certain respects reminiscent of those controls.
Analyses of the risk-taking channel of monetary policy include G. Jiménez, S. Ongena, J. L. Peydró and J. Saurina, 2014, “Hazardous times for monetary policy: What do twenty-three million loans say about the impact of monetary policy on credit risk-taking?,” Econometrica, Vol. 82(2), pp 463-505. I. Angeloni., E. Faia and M. Lo Duca, “Monetary policy and risk taking”, Journal of Economic Dynamics and Control, Vol. 52, 2015, pp. 205-307.
See D. Diamond and R. Rajan, 2012, “Illiquid Banks, Financial Stability, and Interest Rate Policy”, Journal of Political Economy.
See, for instance, P. Angelini, S. Nicoletti-Altimari and I. Visco, November 2012, “Macro-prudential, Micro-prudential and Monetary Policies: Conflict, Complementarities and Trade-offs”, Questioni di Economia e Finanza, Banca d’Italia.
For example, C. Borio, “The international monetary and financial system: its Achilles heel and what to do about it”, BIS Working paper n. 456, August 2014.
When both policies are available, coordination between them can be beneficial. See O. Blanchard, G. Dell’Ariccia, and P. Mauro, April 2013, “Rethinking Macro Policy II: Getting Granular,” IMF Staff Discussion Note; I. Angeloni and E. Faia, April 2013, “Capital Regulation and Monetary Policy with Fragile Banks”, Journal of Monetary Economics; P. Angelini, S. Neri and F. Panetta, September 2014, “The Interaction between Capital Requirements and Monetary Policy”, Journal of Money, Credit and Banking 46, pp. 1073–1112.
A summary of global experiences in using macro-prudential instruments goes beyond the scope of my remarks today. For a review see I. Angeloni, 2014, “European macro-prudential policy from gestation to infancy”, in ‘Macro-prudential policies: implementation and interactions,’ Banque de France Financial Stability Review No. 18, and references therein.
The source is the ESRB’s Report A review of macro-prudential policy in the EU one year after the introduction of the CRDIV/CRR, (forthcoming).
These measures include, for instance, setting the countercyclical capital buffer rate at 0% or keeping the rate unchanged, and exempting small and medium-sized investment firms from the capital conservation buffer and/or the countercyclical capital buffer.
COUNCIL REGULATION (EU) No 1024/2013 of 15 October 2013 conferring specific tasks on the European Central Bank concerning policies relating to the prudential supervision of credit institutions.
See, for instance, I. Angeloni, 2014, “European macro-prudential policy from gestation to infancy”, in ‘Macro-prudential policies: implementation and interactions,’ Banque de France Financial Stability Review No. 18.