After the big bang and before the next? Reforming the financial supervision architecture and the role of the central bank

Donato Masciandaro, Marc Quintyn 03 February 2009



In June 1998, the Bank of England turned over banking supervision responsibility to the newly established Financial Services Authority (FSA), charged with the supervision of all segments of the financial system. That transfer marked a significant shift in thinking about financial sector supervision. Indeed, it was the first time that a large industrialised country – and major international financial centre – decided to assign the task of supervising the entire financial system to a single authority other than the central bank.1 This column introduces CEPR Policy Insight No. 30, released today, "Reform of financial supervision and the role of central banks: a review of global trends, causes and effects (1998 to 2008)".

A wave of reform

Since that landmark date, the number of unified supervisory agencies has grown rapidly worldwide, particularly in Europe. In addition to the UK, three “old” EU members – Austria (2002), Belgium (2004), and Germany (2002) – have assigned the task of supervising the entire financial system to a single authority other than the central bank. Switzerland moved to the same model in 2009. They have been accompanied by five “accession countries” – Estonia (1999), Latvia (1998), Malta (2002), Hungary (2000) and Poland (2006).2

The reform wave caught on outside Europe as well. Unified agencies have been established in Kazakhstan, Korea, Japan, Nicaragua, Rwanda, and a number of other countries. The landscape became even more diversified with the adoption, in the Netherlands and Australia, of the twin-peak model – the assignment of one supervisor per public objective. In addition, a large number of countries, while refraining from drastic overhauls, made changes to the supervisory landscape by moving away from the traditional “silo” model (one supervisor per segment of the financial sector) and merging two or more sector supervisors based on local circumstances and preferences. As a result, the world of supervisory architectures is now much more diversified than 20 years ago when the sector-specific supervisor dominated the scene. In addition, the role of central banks in supervision has been changing all along.

Why this reform wave?

To understand the root causes of these reforms, we need to go back to the financial liberalisation waves that started in the 1970s and sprawled in all directions in the subsequent decades. Liberalisation unleashed competitive forces that created pressure to take on more risks and search for innovations. Great advances in information and communication technologies spurred many new developments and opportunities, first within the banking system, then in other segments of the financial system, and finally across them, blurring previously distinct segments of the financial sector. Policymakers responded by reforming their supervisory apparatuses.

Is this reform process coming to a close, now that the financial crisis has turned attention to redefining regulatory frameworks, incentive structures, safety nets, and several other features central bank and supervisory intervention? Quite to the contrary! While supervisory architectures per se are not among the root causes of the crisis, weak oversight, regulatory gaps, and lack of coordination and information exchange among supervisors are often related to the features of the supervisory architecture. The architecture plays an important role in making supervision effective and efficient.

Some countries that were considering institutional reforms prior to the crisis are now convinced that their supervisory architectures need to be tackled. Past reformers are reconsidering their architectures in light of the lessons of the crisis. And the two financial giants most affected by the crisis – the US and the EU – are drawing up plans to strengthen their supervisory structure. In the former, weaknesses in the architecture have become very apparent in the run-up to the crisis. In the latter, the concern is coordination of national supervision in the face of dominating cross-border groups, which has consequences for the architecture.

Emerging supervisory architectures

The remarkable diversification of the supervisory landscape that we are witnessing has generated a lot of intellectual curiosity with respect to the motives (underlying and revealed) of the reformers, the role of politicians, central banks, and markets in the direction of the reforms, and the impact of these reforms on financial sector performance. More specifically, the literature has addressed three broad questions:

  • What are the main features of the emerging supervisory architectures?
  • What explains the cross-country diversity of the emerging institutional settings – economics, politics, or something else?
  • What effects have regime shifts had on the quality of regulation and supervision?

Our CEPR Policy Insight No. 30, released today, summarises the answers provided so far. One of the interesting features of the current supervisory landscape is the emerging dichotomy between the role of the central bank in supervision and the trend toward supervisory unification. In a majority of countries that have opted for a unified supervisor, the central bank has been given no role in supervision. On the other side of the spectrum, in those countries that stayed close to the “silo” approach to supervision, the central bank is the main (or sole) bank supervisor in a majority of cases. This phenomenon – that the degree of supervision unification seems to be inversely correlated with central bank involvement – has been labelled the “central bank fragmentation effect”

The central bank fragmentation effect

The literature has dug deeper to find some explanation for this effect. Several potential determinants, including the quality of public sector governance, country size, influence and composition of financial markets, a possible demonstration effect (others reform so we follow), and the impact of a financial crisis, have been tested. The empirical analysis supports the view that the central bank fragmentation effect dominates. If the central bank is historically the bank supervisor, the likelihood that a unified supervisor outside the central bank will be established diminishes. The trendsetter – the UK – turns out to be an exception to the rule.

Good public sector governance increases the probability that a unified supervisor is established outside the central bank. And the size of the country also matters –smaller countries, with smaller financial systems and capacity constraints, seem to be more willing to adopt a unified model to reap scale economies.

Finally, the (thinly populated) literature on the impact of supervisory reform on the quality of supervision seems to indicate that unified regimes, on average, are associated with higher quality and consistency of supervision. Whether the unified supervisor is located inside or outside the central bank does not seem to matter.

Supervising the ever-changing markets

It is clear from the review that attention to the supervisory architecture remains important. It is often stated that the architecture is a second-order issue – that the quality of regulation and supervision is the important issue from a financial stability point of view. This is correct, but the impact of the architecture on the efficiency and efficacy of regulation and supervision should not be underestimated. The crucial question is: how can we best supervise the ever-changing markets, which are becoming increasingly complex and intertwined with each other?

While there is no such thing as a “best model” or “best practice” when it comes to the architecture, the general formula for effective supervision is always the same; it is necessary to have exhaustive and up-to-date information. But applying that today is tricky. In markets that were fundamentally static and segmented – banks, the stock exchange, insurance – a simple “photograph” of the situation every now and then was sufficient. The vertical or silo model was the natural and effective answer. Nowadays, to have exhaustive and up-to-date information, the best architecture will vary across countries and circumstances. We need to explore innovative models of supervision – the unified model and the horizontal model – and adopt what is best given the context.

However, the economic rationale for modifying the supervisory settings is not always sufficient. Whether financial supervision reforms see the light of days depends on politics and politicians. More research on why and when politicians do (and do not) act is needed. If the government’s helping or grabbing hand leaves fingerprints all over the topic, the mission of the researcher is to find them.

Views expressed herein are those of the author and should not be attributed to the IMF, its Executive Board, or its Management.

1 Admittedly, the UK was not the first one to unify its supervisory structure. The Scandinavian countries (Norway (1986), Iceland and Denmark (1988) and Sweden (1991)) had preceded the UK in the aftermath of a domestic financial crisis. But it is no exaggeration to state that the establishment of the FSA was the first reorganisation to become front page news in the international press.
2 In Ireland (2003) and the Czech and Slovak Republics (both 2006), the supervisory responsibilities were concentrated at the hands of the central bank.




Topics:  Financial markets Institutions and economics

Tags:  reform, financial regulation, supervisory architectures

Full Professor of Economics, and Chair in Economics of Financial Regulation, Bocconi University

Division Chief, Africa, at the IMF Institute


CEPR Policy Research