Mugur Isărescu: Recent developments in central bank governance

Panel presentation by Mr Mugur Isărescu, Governor of the National Bank of Romania, at the "Central Bank and Supervisory Governance" conference, organized by the National Bank of the Netherlands, Amsterdam, 12 February 2016.

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

Central bank speech  | 
24 March 2016
PDF version
 |  6 pages

Ladies and gentlemen,

It is a great honour to address the audience during this high-level conference on a subject of special interest to all of us, the recent developments in central bank governance leading to heated debates every time they are brought to the fore.

Mr. He has done a wonderful job in structuring the challenges of nowadays' central banking and in outlining the importance of effective governance arrangements in order to meet these challenges - his presentation is a testimony to IMF's unique ability to draw upon a vast set of worldwide institutional experiences. I, in turn, will draw mainly upon a narrower, but hopefully no less relevant, set of institutional experiences, gathered during my tenure as Governor of the National Bank of Romania.

In my address today I intend to deal with the issue of recent developments in central bank governance by looking at:

1. The central bank objectives as the starting point in designing its governance framework;

2. How the NBR dealt with the financial stability as an objective, from an institutional point of view; integrating or separating monetary policy and supervision;

3. A short history of institutional developments within the NBR for ensuring financial stability;

4. The internal control layers in the NBR;

5. Pros and cons of different institutional alternatives: banking supervision and macroprudential authority - inside or outside the central bank;

6. A historical perspective on the experience of becoming a modern central bank, a process undergone not only by the NBR.

1. Defining the objectives - the starting point in designing central bank governance

It goes without saying that ground zero in building the central bank governance framework is clarifying the objectives it has to pursue. For quite a long time, at least in Europe, "price stability" enjoyed the position of overriding objective, but since the outbreak of the crisis it became obvious that the "inflation first" approach may turn counterproductive in an environment of financial instability. Therefore, financial stability concerns made their way up on central bankers' agenda.

Is that really a novelty? As the representative of an emerging economy central bank that never had the luxury of being concerned exclusively with price stability, I dare say it is not. Actually, central banking has always been about the preservation of, to quote Charles Goodhart, "the three aspects of stability", namely price stability, (external) currency stability and financial stability. Moreover, the fact that these aspects of stability do condition each other was not discovered in the aftermath of the crisis - central bankers, maybe even more so emerging-market central bankers, have long been keenly aware that failure to ensure one kind of stability is bound to jeopardise the achievement of the others.

What the crisis has indeed challenged is the widespread conviction that meeting one particular stability criterion, while necessary in order to achieve the others, is also a sufficient condition to do so - after all, maintaining price stability, while valuable in itself, has proved insufficient to ward off major financial or macroeconomic instability. As a matter of fact, Andrew Crocket claimed ever since 2003 that the "peace dividend" yielded by the successful war against inflation had not lived up to expectations and therefore the battlefront against financial instability should not be overlooked.

Regardless of how this battlefront is approached, it is of the essence to make sure that the tasks entrusted to the central bank are feasible. Otherwise, if the monetary authorities are overburdened with possibly conflicting goals, the major risk is to render them inefficient and unable to deliver a single one of their objectives. This is the very reason why the issue of central bank governance is so intricate and delicate: one needs to reconcile objectives that, under certain circumstances, might even prove incompatible.

2. Integrating or separating price and financial guardianship

Given that the conference focuses on central bank and supervisory governance, allow me to touch on the issue of integrating price guardianship and financial guardianship within the central bank. As you probably remember, the idea that the two tasks should be completely separated has gained ground in the 1990s. The rationale behind this view was to free the monetary policy conduct from conflicting objectives and thus render policymakers' mission and life easier. It was advocated that banking sector supervision should no longer be assigned to the central bank, but rather to an independent agency (which might also ensure the consolidated supervision of all financial markets).

I, myself, found the idea rather interesting at the end of the 1990s, when the collapse of several Romanian banks influenced the decision-making process at the National Bank of Romania. I realised at the moment how costly it was in terms of reputation for a central bank to be in charge of supervision and prudential regulation - not to mention that the separation appeared quite fashionable after the newly-established Financial Services Authority had become operational in the United Kingdom.

At the same time, I vividly remember the conflict between price stability and financial stability, when I had to mediate the divergent positions taken during Board meetings by the deputy governors coordinating the two functions, not by virtue of personal affiliations, but due to the very nature of the specific tasks they had been entrusted with. I was wondering, in light of these Board disputes, whether separating the monetary policy and the supervision functions was preferable.

At hindsight, I believe that Romania has made the right choice preserving banking supervision as a prerogative of the central bank, sorting out such disputes under the same roof.

Coming to the current NBR governance framework, I would like to start by pointing out that it ensures a complete separation as regards the coordination of the relevant departments: while the Governor is in charge of the monetary policy area, the First Deputy Governor coordinates the Supervision Department; one Deputy Governor takes care of financial stability and the other, of bank resolution.

The nine-strong Board of Directors runs the central bank, but the process of preparing its decisions in the areas of monetary policy and prudential supervision takes place in two distinct structures: the Monetary Policy Committee and the Supervisory Committee, respectively, which comprise members of the executive management and experts in the relevant areas. As separation is not the same thing as complete isolation, I have to add that both committees are chaired by the Governor...

Pursuant to the NBR Statute, the Board is tasked to act as a first appellate forum for sanctions on credit institutions. In order to avoid conflicts of interest that may appear, the following institutional arrangements are in place: separate Board meetings, a certain degree of autonomy of the Supervision Department and recusal from the proceedings of the First Deputy Governor, in his capacity as coordinator of the Supervision Department.

I would like to add a special remark regarding the new role of the NBR as the resolution authority, arising from the provisions of the Bank Recovery and Resolution Directive. We have configured the activities and reporting lines in order to ensure full operational independence and to avoid conflicts of interest between the supervision and resolution functions. Of course, this takes place without prejudice to the exchange of information and cooperation obligations.

To be more specific, supervisory decisions do not pose constrains to resolution-related ones. Moreover, the staff involved in performing the resolution function is structurally separated from the other NBR functions.

3. Romania: a short history of institutional changes for ensuring financial stability

The global crisis has made obvious for all of us that price stability and financial stability are much more intertwined than we used to think. I believe this development has farther-reaching consequences when it comes to the future of central banking. As I pointed out in November last year at the World Policy Conference organised in Montreux, I cannot imagine a future where the explicit concern for financial stability and the macroprudential policy are not central banking fixtures.

In Romania, the concern for ensuring and strengthening financial stability was visible in several significant changes of the institutional framework, operated already before the outbreak of the global economic and financial crisis. Specifically, the NBR's Financial Stability Department was created as early as 2004, with the first edition of the annual Financial Stability Report - a report that will be published on a half-yearly basis starting this year - being released in 2006.

Of course, no matter how much or how responsibly the central bank would get involved in safeguarding financial stability, this public good can only be delivered through cooperation and coordination with the other institutions tasked with financial market supervision and regulation. Relevant in this respect is that in Romania a National Committee for Financial Stability was established as early as 2007. The Governor of the National Bank of Romania, the Minister of Public Finance, the President of the Financial Supervisory Authority and the Chairman of the Bank Deposit Guarantee Fund are members of this committee, whose key objective is to ensure the exchange of relevant information, as well as to prevent, appraise and manage any issues with a potentially systemic impact.

Efforts are currently underway to turn this entity set up based on an agreement among member institutions into a National Committee for Macroprudential Oversight, whose composition, tasks and areas of competence shall be set forth by law. This inter-institutional structure will act as a macroprudential authority, within the meaning of ESRB Recommendation 2011/3.

4. NBR: the internal control layers

I imagine we all agree that improving the governance structure is essential for a central bank and I strongly believe the sharing of views and experience made possible by reunions such as the one we are attending is extremely beneficial for the process.

As far as the NBR is concerned, enhancing the governance has been a steady concern over the course of time, the ultimate goal being the alignment with the best practices, often with IMF support.

Let me just point out that the NBR shifted from an internal auditors' Commission to the auditing of financial statements by external auditors as early as 2001, while the following year saw the establishment of the Audit Committee as a substructure of the NBR Board (at a time when not even the ECB had such a committee in place!).

In addition to the Audit Committee, the NBR's internal control system includes several control layers. In fact, there are three lines of defence against any factors that might jeopardise the fulfilment of the NBR objectives.

First of all, the National Bank of Romania takes a functional approach to risk management, meaning that each organisational unit has primary responsibility for identifying, assessing and managing the risks associated with its own activities and operations. The risk management process is an integral part of the NBR's internal control system and represents the bank's first line of defence. Each organisational unit implements operational control procedures within its area of responsibility, in accordance with the levels of risk tolerance set in advance by the executive management. Furthermore, the NBR applies the approach known as the "Chinese wall", which separates the departments in charge of monetary policy formulation and implementation, on one hand, from the departments entrusted with other statutory tasks, on the other.

The second line of defence is functional on certain business segments with high-risk financial exposure, e.g. international reserve management. This line of defence monitors compliance with the risk limits approved by the bank's executive management and reports any breach.

A major role in this line of defence is played by the decision-making/advisory committees within the central bank. Apart from the two committees dedicated to monetary policy and supervision that I have already mentioned, particularly significant is the Foreign Reserve Management Committee.

In addition to the risk monitoring by the operational structures, the NBR's Internal Audit Department - which functions as the third line of defence - examines the overall controls in place and determines whether they are properly designed and functional, so as to ensure the reliability and integrity of financial and operational information, the effectiveness of the activities carried out, asset protection, as well as compliance with applicable legal and contractual provisions.

5. Different institutional arrangements for banking supervision and financial stability

As we all know, the responsibility for banking supervision rests in some countries with the central bank, whereas in others it lies with a distinct authority. Both options are valid and a one-size-fits-all solution will probably not be reached, given the country-specific financial conditions and institutional arrangements. However, this does not necessarily exclude looking into the arguments for or against either alternative.

As I already said, I believe that it is better to have supervision integrated in the central bank. First of all, the ongoing complex challenges to financial and economic stability call for identifying, using and corroborating tons of information in order to articulate the adequate policies. The central bank relies on its database and expertise, which are prerequisites for managing such a challenging environment. Beyond this major advantage, there are also other benefits of such an institutional arrangement. Independence from political influence is essential for effective supervision. And central banks are institutions enjoying a high level of independence that a newly-established government agency would not be able to gain, at least in the short run.

Besides, clearly acknowledging the links between price stability and financial stability means that they need to be taken into account when formulating the monetary policy. Moreover, as Benoît Cœuré pointed out, data collected and analyses conducted as part of banking supervision provide valuable information about the financial sector and may feed into the assessment of the state of the economy, being useful for the monetary policy decision-making process.

Of course, literature identifies some disadvantages of integrating supervision in the central bank. The risk of this arrangement resulting in a too loose monetary policy, due to the conflicting objectives, which I have already referred to, is an important drawback. However, as Beck and Gros stressed in an article from 2012, with the rise of macro-prudential regulation as an additional tool, it is not clear whether price and financial stability still conflict with each other.

Another disadvantage is that the reputation and/or the independence of the central bank might be undermined: if its decisions and actions in the area of supervision and resolution are perceived as wrong or suboptimal, this may negatively affect the credibility of the monetary policy, as it is conducted by the same institution. Moreover, the temptation of politicians to capture the central bank may rise, as it has much more power than in the case when its only mission is to deliver price stability.

The various arrangements on assigning banking supervision also have a bearing on the design of the national macroprudential authority. Depending mainly on who conducts the supervision of banks, different institutional solutions are adequate for this authority: it may be represented by a committee or board comprising representatives of the relevant institutions, by the central bank or by another authority tasked with financial market supervision. Boards are in place or envisaged in 13 EU countries, the central bank is - or is projected to be - the national macroprudential authority in 13 other EU countries, whereas two opted for - or are considering - entrusting this role to the national supervisory authority.

Opting for the central bank as a national macroprudential authority can be warranted by the experience accumulated in the field, especially in the case of a dominant position of the banking sector, and by the possibility of a quick implementation of adopted measures (since the institution is responsible for identifying systemic risks and, at the same time, has available the macroprudential tools to put into practice the measures considered adequate).

Some disadvantages of having supervision integrated in the central bank are manifest in this case as well, such as the backfiring of potential errors in the area (be they real or merely perceived as such by the public) on monetary policy credibility or a keener interest in politically capturing the institution as a result of the substantial rise in central bank powers.

As Klaas Knot has showed more than once, the key argument in favour of creating a board is that it enables the pooling of the expertise of all financial market supervisors and of the government; the drawbacks would be more diffused accountability and the risk of delayed implementation of recommendations, given the involvement of several institutions. A prominent role of the central bank in the board (either by chairing it or through the number of members assigned) might mitigate the risk of inaction.

In the case of Romania, where the supervision of banks and non-bank financial institutions is the task of the NBR, while the Financial Supervisory Authority (FSA) oversees the other financial markets, the solution that emerges as adequate is the one identified in the draft law promoted by the Government: a National Committee for Macroprudential Oversight. The decision-making body of this committee is to be made up of nine members: the Governor, the First-Deputy Governor, the two Deputy Governors and the Chief Economist of the National Bank of Romania; the President and the First-Vice President of the Financial Supervisory Authority, as well as two Government representatives appointed by the Prime Minister.

Addressees of National Committee recommendations - the NBR and the FSA, as sectoral financial supervisory authorities, and the Government - should act on them unless inaction can be adequately justified ("act or explain" mechanism).

6. Reforming central bank governance after the fall of communism

I have always been of the opinion that one cannot truly understand the present or have a glimpse into the future without knowing the past and learning its lessons. These lessons begin with the transition from a monetary authority in a command economy to a modern central bank operating in a market economy, an experience that is, for sure, familiar to many of us present here - I believe half of the central banks in this constituency, including Romania's, are "new kids on the block".

Of course, there are different stages of development in terms of institutional framework. Some central banks are presently undergoing the process of shaping and consolidating their governance, an endeavor intended to transform them into modern monetary authorities. And, believe me, I know from experience how complicated such a process can be!

In the early nineties, when I took over as Governor of the National Bank of Romania, a tremendous effort was needed to put into practice the transition from a monobank-type system to a two-tier system. It was by no means a simple task to reform an institution that had operated in the context of centralised planning of the entire economic activity and to separate commercial banking operations from central bank operations, which remained the responsibility of the NBR.

The new NBR only counted less than 5 percent of the former central bank employees, most of them in the cash management and accounting units! Therefore, we had to build from scratch several departments that are key to the functioning of a modern central bank, such as monetary policy, economic research, banking supervision and regulation...

Fortunately, the NBR was given a helping hand in improving its governance structure when it needed it the most. Just to give two examples, banking supervision was set up at the NBR with the extremely useful support of De Nederlandsche Bank and the forex market was put in place with the assistance from the central bank of Belgium. We are extremely grateful to both institutions for their generous support.

Like they say, what goes around, comes around - it is now our turn to share our experience. I am glad to say that the NBR - in a consortium with De Nederlandsche Bank - has been involved since June 2015 in a Twinning project dedicated to the National Bank of Moldova, the NBR providing expertise in the areas of regulation and authorisation, supervision, financial stability and human resources.

While speaking about my own experience in reforming the NBR in the early 1990s, after the fall of communism, the words of Walter Bagehot have come to my mind. Referring to the needed changes in the governance framework of the Bank of England, he wrote in his book "Lombard Street: A Description of the Money Market", published as early as 1873, that "putting new wine into old bottles is safe only when you watch the condition of the bottle, and adapt its structure most carefully".

I find this metaphor particularly adequate for guiding us in the demanding task of tailoring the structure of central banks to the challenges of today and tomorrow. Let me add only that - as I once put it - good wine is as difficult to make as good monetary policy. And proper dosage is of the essence in both cases.